RELATIONSHIP BETWEEN PERSONALITY AND VIDEO GAME PREFERENCES

A Thesis

Presented to the faculty of the Department of Psychology California State University, Sacramento

Submitted in partial satisfaction of the requirements for the degree of

MASTER OF ARTS in Psychology (Counseling Psychology) by Joseph B. Borders SUMMER 2012

RELATIONSHIP BETWEEN PERSONALITY AND VIDEO GAME PREFERENCES

A Thesis by Joseph B. Borders

Approved by: , Committee Chair Dr. Lee Berrigan , Second Reader Dr. Lawrence Meyers , Third Reader Dr. Marya Endriga

Date

ii

Student: Joseph B. Borders

I certify that this student has met the requirements for format contained in the University format manual, and that this thesis is suitable for shelving in the Library and credit is to be awarded for the thesis.

, Graduate Coordinator

Dr. JianJian Qin

Date

Department of Psychology

iii

Abstract of RELATIONSHIP BETEEN PERSONALITY AND PREFERENCES IN VIDEO GAME PLAY by Joseph B. Borders Video games are a popular form of media that are enjoyed by individuals with a wide range of ages. However, to date there has been very little research conducted examining the relationship between personality and preferences for different types of video games. The majority of studies have focused on preferences for violent or prosocial games and have failed to demonstrate empirically distinct video game types. The current study examined the relationships between several personality traits as measured by the NEO Five Factor Inventory and the California Psychological Inventory and preferences for different types of video games as measured by a modified version of Zammitto’s (2010) Gaming Preferences Questionnaire. Principal components analysis yielded three types of video games that were found to be related to sex and a narrow set of personality traits. , Committee Chair Dr. Lee Berrigan

Date

iv

DEDICATION

To Pat Floyd, my mother, for saving me. There was a time when I knew I wouldn’t amount to anything in life. Thank you for giving me the foundation that enabled me to reach for the stars and become the person I am today. Orin Borders, my father, for creating in me an appreciation of education and psychotherapy. Chloe Borders, my sister, who gives me hope for the future and provided positive reinforcement to help me work on completing this thesis. My wife, Angela Mae Borders, who encourages me daily with her strength and bravery. This thesis is for her and the dreams we will pursue together now that I am done with school.

v

ACKNOWLEDGEMENTS I am deeply grateful to my thesis chair, Dr. Lee Berrigan, without whom this thesis would have been significantly less substantive and detailed. His enthusiasm and weekly guidance enabled me to complete this thesis in a timely manner and for that I am thankful. I am also grateful to Dr. Lawrence Meyers who guided me in the data analysis of this study. Throughout construction of the results and discussion section of this thesis, Dr. Meyers was constantly available to elucidate statistical concepts that were confusing to me. I would also like to thank Dr. Marya Endriga for being my third reader, and for providing words of support and encouragement. Also, many thanks to my friend E Ting Lee who gave me a lot of encouragement and assistance with the statistics involved in this study.

vi

TABLE OF CONTENTS Page Dedication ............................................................................................................................v Acknowledgments.............................................................................................................. vi List of Tables ........................................................................................................................x Chapter 1. INTRODUCTION ..........................................................................................................1 Children and Adolescents ....................................................................................... 2 Differences between the Sexes ................................................................................2 Violent Video Games ...............................................................................................4 Positive Effects of Video Game Play ......................................................................5 Classification of Video Game Types .......................................................................7 Personality and Video Game Preferences .............................................................16 The Affect-Dependent Theory of Stimulus Arrangement .....................................19 Measures of Personality .........................................................................................20 The Five-Factor Model of Personality .......................................................21 The California Psychological Inventory ....................................................26 The Present Study .................................................................................................31 Hypotheses ............................................................................................................34 2. METHOD ....................................................................................................................36 Participants ............................................................................................................36 Materials ................................................................................................................36 vii

Procedure ...............................................................................................................42 3. RESULTS......................................................................................................................44 Invalid Packet Exclusion........................................................................................44 Preliminary Data Analysis ....................................................................................44 Principal Components Analysis ............................................................................49 Canonical Correlations...........................................................................................63 Analysis of Variance .............................................................................................67 Differences between the Sexes .............................................................................72 Relationships Not Addressed by the Study Hypotheses .......................................80 4. DISCUSSION ...............................................................................................................86 Hypothesis One .....................................................................................................86 Hypothesis Two......................................................................................................90 Hypothesis Three ..................................................................................................93 Relationships Not Addressed by the Study Hypotheses .......................................96 Limitations and Implications of this Research.......................................................98 Appendix A. Descriptions of the Entertainment Software Rating Board (ESRB) Ratings .......................................................................................................102 Appendix B. Demographic Sheet ...................................................................................103 Appendix C. Gaming Preferences Questionnaire ...........................................................104 Appendix D. Gaming Patterns Questionnaire .................................................................107 Appendix E. Consent Form............................................................................................. 110

viii

Appendix F. Debriefing.................................................................................................... 111 References ........................................................................................................................ 113

ix

LIST OF TABLES Tables

Page

1.

Ethnicity of Participants .........................................................................................45

2.

Reported Frequency of Video Game Play .............................................................46

3.

Age When First Played Video Games ....................................................................46

4.

Genres of Reported Top 3 Favorite Games ............................................................48

5.

Variance Accounted for by the Eight Factor Solution Yielded by Principal Components Analysis of the 52 items From the Gaming Preferences Questionnaire with a Promax Rotation ..................................................................50

6.

Correlations of the Eight Components Yielded by Principal Components Analysis of the 52 Items from the Gaming Preferences Questionnaire with a Promax Rotation.....................................................................................................51

7.

Structure Coefficients Based on Principle Components Analysis with a Promax Rotation for the 52 Items from the Video Game Preferences Questionnaire .........52

8.

Reliability and Descriptive Statistics for the Eight Video Game Preference Scales Resulting from a Principal Components Analysis of the 52 Items from the Gaming Preferences Questionnaire ..................................................................57

9.

Variance Accounted for by Each of the Three Components Yielded by a Second-Order Principal Components Analysis (PCA) of the Eight Components Yielded by a PCA Performed on the 52 Items of the Gaming Preferences Questionnaire .........................................................................................................58 x

10.

Correlations of the Three Components Yielded by a Second-Order Principal Components Analysis (PCA) of the Eight Components Yielded by a PCA Performed on the 52 Items of the Gaming Preferences Questionnaire ......59

11.

Structure Coefficients Based on a Second-Order Principle Components Analysis with a Promax Rotation for the 52 Items from the Gaming Gaming Preferences Questionnaire ........................................................................60

12.

Reliability and Descriptive Statistics for the Three Video Game Preference Scales Resulting from a Second-Order Principal Components Analysis (PCA) of the Eight Factors Yielded by a PCA of the 52 Items from the Gaming Preferences Questionnaire ......................................................................................63

13.

Cumulative Percentage of Explained Variance, Eigenvalues, and Squared Canonical Correlations for the Two Canonical Functions .....................................64

14.

Structure Coefficients for Predictor Canonical Variates for the Two Functions ................................................................................................................66

15.

Structure Coefficients for the Dependent Canonical Variates for the Two Functions ........................................................................................................67

16.

Differences in Openness Scores between Those Who Indicated a First Favorite Video Game that was an RPG and Those Whose First Favorite was a Racing, Shooter, Platform, or Sports Game ...........................................................69

17.

Differences in Openness Scores among Those Who Indicated a Second Favorite Video Game that was an RPG and Those Whose Second Favorite was a Racing, Shooter, or Sports Game ...................................................71 xi

18.

Coefficients and Alpha Levels for Three Pearson rs Performed to Examine the Relationships Between Sex and Preferences for Action, Cognitive, and Strategy Games ......................................................................................................72

19.

Observed Frequencies for the Chi-Square Performed to Examine the Relationships between Participants’ Sex and Genres Corresponding to Participants’ First Favorite Video Games...............................................................74

20.

Observed Frequencies for the Chi-Square Performed to Examine the Relationships between Participants’ Sex and Genres Corresponding to Participants’ Second Favorite Video Games ..........................................................75

21.

Observed Frequencies for the Chi-Square Performed to Examine the Relationships between Participants’ Sex and Genres Corresponding to Participants’ Third Favorite Video Games .............................................................76

22.

Chi-Square Statistics and Alpha Levels for Each of Three Chi-Square Tests Performed to Examine the Relationships between Participants’ Sex and Genres Corresponding to Participants’ First, Second, and Third Favorite Video Games ...................................................................................................................77

23.

Observed Frequencies of Male and Female Participants in Each of the Genre Groups Corresponding to Participants’ First, Second, and Third Favorite Video Games Examined via One-Way Chi-Square Analyses ...........................................79

24.

Correlation Coefficients and Alpha Levels for Several Pearson rs that Yielded Statistically Significant Relationships That Were Not Addressed by the Study’s Hypotheses ................................................................................................82 xii

25.

Descriptive Statistics for All Variables Involved in Significant Relationships Explored in Addition to Relationships Specifically Addressed in the Hypotheses of This Study ..........................................................................................................83

26.

Observed Frequencies for the Chi-Square Performed to Examine the Relationship between Sex and Participants’ Indications That They Would or Would Not Choose to Spend More Time Playing Video Games if More Free Time Was Available to Them .................................................................................84

27.

Observed Frequencies for the Chi-Square Performed to Examine the Relationship between Sex and Participants’ Indications That They Do or Do Not Only Play Video Games on Facebook or another Similar Social Networking Site .....................................................................................................85

28.

A Comparison of the Factors Found by Lucas and Sherry (2004) and the Second-Order Principal Components Found in the Current Study .......................88

xiii

1 Chapter 1 INTRODUCTION

The first popular home video game, Pong (Atari Inc., 1972), was released on the Atari console system in 1975 (Funk, 2005). Since then, video games have evolved into a medium of art [Shadow of the Colossus (SCE Studios Japan, 2005)], storytelling [Final Fantasy X (Square, 2001)], education [My Spanish Coach (Sensory Sweep, 2007)], and social networking [World of Warcraft (Blizzard Entertainment, 2004)]. Video games are now available on a wide variety of systems: computers (PC and Macintosh), hand held devices (Nintendo DS and Sony PSP), console systems (Xbox 360 and Nintendo Wii), and arcade video game machines. In the literature, games played on personal computers are referred to as computer games, whereas games played on systems designed for the sole purpose of playing video games, and in some cases movies, are known as video games. Throughout this study, the term “video games” will be used and should be taken to mean any game that is played using either a personal computer or a dedicated gaming system. Over the years video games have become a mainstream part of modern culture. According to the Entertainment Software Association (ESA), the primary organization responsible for managing the business and public affairs of many video and computer game companies in the United States, 67% of American households play video games. The ESA also reports that 25.1 billion dollars were spent by consumers on video games in 2010 (Entertainment Software Association, 2011). Given the fact that video game use is

2 so prevalent, it is important that research be conducted to examine the correlates of video game play. To date, there has been much research conducted concerning video game play; however, the breadth of potential correlates examined in the literature has been quite limited. Children and Adolescents The vast majority of research on video game play has been concerned with the study of video game playing among children and adolescents (Colwell, 2007; Von Salisch, Oppl, & Kristen, 2006). This is partially a reflection of the perception that children and adolescents are impressionable individuals who primarily learn appropriate social behavior through modeling. Another reason for this focus on children and adolescents is the stereotype that children are the primary consumers of video games. When video games were a new form of entertainment in the early 1980’s, the majority of players were children and adolescents. However, partially due to the aging of these original gamers and the increasing availability of titles for more mature audiences, today the average video gamer is 34, and 26% of gamers are over the age of 50 (Entertainment Software Association, 2011). Differences between the Sexes It is also stereotypically thought that male video game players vastly outnumber female players. Research has consistently shown that males enjoy and play video games more than do females (Royse, Lee, Undrahbuyan, Hopson, & Consalvo, 2007; Lucas & Sherry, 2004). However, this trend appears to be diminishing. Current estimates of the concentration of females among video game players range broadly. In a report published

3 in 2011, the ESA reported that 40% of gamers are female (Entertainment Software Association, 2011). In contrast to this, other research has found the percentage of female gamers to be much lower. In a study that sampled 2,000 undergraduate students Terlecki et. al. (2010) found that only 27% of female participants reported that they were currently playing video games at the time of the study as compared to 74% of male participants. Terlecki et al. (2010) also found that males reported having played video games for a significantly longer time in their lifetimes, with females reporting an average of having played video games for two to five years and males averaging ten years. This suggests that females, on average, devote less time to playing video games than do males but engage in game play nonetheless. Therefore, the distinction between the sexes may be that males spend more time on average playing video games than do females but that the sexes are more comparable when simply considering who plays video games and who does not. In a study examining the differences in genre preferences between the sexes, Consalvo and Treat (2002) found that, when given a list of eight common video game genres and asked to select which ones were their favorites, males tended to list sports, action/adventure, and simulation as their top three, whereas females tended to list puzzlesolving, platform, and sports as their favorite genres. Boys have also been shown to prefer fighting/combat games more than girls (Terlecki et. al., 2010). One study (Lucas & Sherry, 2004) that more exhaustively examined the differences between the sexes in genre preferences found statistically significant differences between males and females, with males more strongly preferring fighter, shooter, sports, racing, fantasy/role playing,

4 action/adventure, and strategy games, and females preferring card/dice games, classic board games, quiz/trivia, puzzle, and arcade games. Research has also shown that males are more likely to purchase video games with higher violence content and Entertainment Software Rating Board (ESRB) ratings (the current established system for rating video game content) than are women (Pryzbylsky, Ryam, & Rigby, 2009). For a more complete discussion of ESRB ratings please refer to Appendix A. These differences suggest that males may be more likely to prefer violent, action oriented games, whereas females are more likely to prefer less violent, more prosocial games. Violent Video Games The majority of research regarding video game play has been concerned with the effects of playing violent video games (VVGs) (Anderson et al., 2004; Ferguson, 2007; Ferguson & Kilburn, 2010). The rationale given for this focus has often been the fact that some adolescent perpetrators of violent crimes had histories of playing VVGs (Anderson et. al., 2004). The most commonly cited instance of this perceived relationship is the Columbine High School shooting in 1999. The two adolescents who went on a shooting spree at Columbine High School were known to have been players of violent video games such as Wolfenstein 3D (Id Software, 1994) and Doom (Id Software, 1993). After the Columbine shooting, there was much discussion in the popular media that VVGs were potentially to blame for the actions of these adolescents. This focus on VVGs in the literature may also be a reflection of the popular perception that most gamers are male. The thinking appears to be that VVGs provide testosterone ridden males a platform through which to express violent urges. This in turn

5 is seen as priming them for further aggressive behavior. Research has also shown that there is a clear publication bias for studies examining the effects of VVGs as opposed to prosocial, educational, or simply less violent games (Ferguson, 2007). This publication bias has resulted in a substantial lack of published research pertaining to video games other than those examining violence and aggression. Positive Effects of Video Game Play Video game play has been associated with many positive effects, including the development of positive attitudes toward the use of technology, computer literacy, improved cognitive and attention skills (Lucas & Sherry, 2004), and moral development (Baranowski, Buday, Thompson, & Baranowski, 2008). Relatively recently, games such as Dance Dance Revolution (Konami TYO, 2001) that require players to control games through aerobic activity have been providing players with health benefits (Baranowski et al., 2008). Most studies that have found positive effects resulting from video game play recognize various different popular genres of video games and tend to focus on games that are considered prosocial or educational (Gentile et al., 2009). Research has shown that playing prosocial video games is related to increases in empathy and decreases in schadenfreude (happiness at the misfortune of others) (Greitemeyer, Osswald, & Brauer, 2010). One particularly interesting study composed of four experiments exposed participants to either a prosocial, neutral, or aggressive video game and measured intergroup differences in prosocial behavior (Greitemeyer & Osswald, 2010). In the first experiment a confederate dropped a jar of pencils and recorded whether participants

6 voluntarily helped collect them or not. Sixty seven percent of those who played a prosocial game helped the confederate as compared to only 33% of those who played a neutral game and 28% of those who played an aggressive game. In the second experiment, after playing either a prosocial or neutral video game, participants were asked if they would be willing to participate in future research and how much time they could give. One hundred percent of participants in the prosocial video game group said that they were willing to participate in future research as compared to 68% of those who played the neutral game. Participants in the prosocial video game group also indicated that they would be willing to devote more time to a future study than did those in the neutral game group. In the third experiment, after participants had played a prosocial or neutral video game for eight minutes, a confederate posing as an angry boyfriend entered the experimental room and verbally and physically harassed the female experimenter. In this experiment prosocial behavior was defined as participants making verbal or physical efforts to intervene and help the researcher. Fifty six percent of participants in the prosocial video game condition helped the researcher as compared to 22% of those in the neutral condition. In the fourth experiment participants were given the opportunity to write down what they were thinking while playing either a prosocial or neutral video game. Those who played a prosocial game displayed significantly more prosocial thoughts (M = 1.26, SD = 1.15) than those who played a neutral game (M = 0.06, SD = 0.24) (Greitemeyer & Osswald, 2010).

7 Classification of Video Game Types Surprisingly few studies have been performed to examine the validity and accuracy of current video game genre classifications (Apperley, 2006). For research to succeed in examining the correlates of video game play it is essential for there to be an accurate system of video game classification. Without such a system, researchers are left without a valid and reliable means of distinguishing any one video game from another. In his insightful article, Apperley (2006) argued that the current genre classification system used for video games is inefficient and flawed in that it is based on genre classifications of films and other narrative media forms. In his paper, Apperley discusses the two opposing camps in this debate. “Narratologists” hold that video games should be defined based on their narrative whereas “ludologists” hold that video games should be classified by other features unique to video games such as the rules and action of play. The current video game classification system primarily uses a ludological approach and categorizes video games mostly based on the ways they are played. This can be confusing, as two games can share the same genre but be extremely different in many regards. For example The Legend of Zelda: The Wind Waker (Nintendo, 2003) and Resident Evil 5 (Capcom, 2005) are both classified as action/adventure games. The Legend of Zelda: The Wind Waker (Nintendo, 2003) follows the story of a young hero who primarily fights enemies with a sword on his way to save a princess. The game world is vast and open, allowing players to explore, find treasures, and complete quests on the way to the ultimate goal of saving the princess. This game has relatively low violence content and places as much emphasis on puzzle solving as it does combat.

8 Resident Evil 5 (Capcom, 2005) on the other hand, is the story of a military weapons company that designs a virus that turns people into mutated monstrosities. Game play is relatively linear, requiring players to follow a designated sequence of unfolding events. Players use guns to shoot gruesome monsters, and there is a significant element of fright and horror. These two games share the same genre but are strikingly different in many regards. This reflects one of the problems with the current video game classification system. Reflecting the ludologist approach, current video game genres are intended to tell players what kind of game play they can expect from any particular game, but tell them nothing about the graphic and narrative content of a game. From a narratologist perspective, The Legend of Zelda: The Wind Waker (Nintendo, 2003) would likely be categorized as a drama whereas Resident Evil 5 (Capcom, 2005) would likely be considered a horror. The current video game classification system arranges video games into different “genres” based on the ways games are played. There is much disagreement as to how many distinct genres exist and what should be considered inclusion criteria for each genre. Some researchers have argued for the existence of as few as three distinct genres (Lucas & Sherry, 2004) whereas others have argued for the existence of as many as 42 (Wolf, 2001). In an attempt to solve this problem, many individuals conceptualize video game categories in the form of super-genres and sub-genres. Many games within a genre have conceptual differences that qualify them as belonging to various sub-genres. For example, the relatively broad super-genre of puzzle games encompasses all games that primarily require players to solve puzzles. However, there are many puzzle games that

9 specifically involve the arrangement of letters and words, such as Words with Friends (Newtoy, 2010) and Bookworm Deluxe (PopCap Games, 2005). These games could be considered to belong to a sub-genre known as “word puzzles.” In a manner similar to this, Apperley (2006) named four main genres (simulation, strategy, action, and role playing games) and gave each of them various sub genres. Zammitto (2010) did this in her study as well by creating sub genres of action-no shooting, action-shooting, and action-fighting for the super genre action. A comprehensive description of all possible genres is beyond the scope of this study. In place of an exhaustive description, only the most popular genres will be described here. The ESA (2010) reported that the top selling super genres in 2009 were (in order of popularity): sports, action, family entertainment, shooter, racing, adventure, strategy, role-playing, fighting, children's entertainment, flight, and arcade. As some of the most popular, these genres would be recognized and understood by most experienced gamers. Of these 12 most popular genres, family entertainment, arcade, and children's entertainment are the easiest to define. Family entertainment games are those that are intended to be played by families. These games involve light competition and usually take the form of board game type games such as Mario Party 8 (Hudson Soft, 2007). Children's entertainment games include all games that are intended solely for the consumption of children. These games are simplistic, contain very little violence, if any, and typically take the form of interactive stories tied to popular movies or television shows. Arcade games are characterized as being simple, skill based games where players

10 play through levels of progressing difficulty. Unlike many other games, arcade games require very little time commitment. The name “Arcade” may imply to some readers that these games are only found on arcade machines. This is not true. The genre arcade was named so in reflection of the fact that most games found in arcades in the early days of video games shared many qualities with those included in the genre, namely, that they are short, simple, and require little time commitment. Some Arcade games include Angry Birds (Rovio, 2009) and Ms. Pac-Man for the Super Nintendo Entertainment System (Williams Entertainment Inc., 1996). Sports games such as Madden NFL 12 (Tiburon, 2011) allow players to play simulations of various athletic sports. These games usually require players to manage several characters and are often played in multiplayer mode with multiple players controlling different teams (Smith, 2006). Racing games such as Dirt 3 (Codemasters, 2011) simulate the experience of racing and allow players to control a vehicle through which they race other players to a finish line (Smith, 2006). Flight games, more commonly referred to as “flight simulators”, simulate the experience of flying by allowing players to control an airplane, spaceship, or other flying machine. Some, including Zammitto (2010), argue that flight and racing games belong to a much broader super genre referred to as “simulation.” Simulation games are those that attempt to simulate, as closely as possible, real life situations. Some simulation games include Nintendogs: Lab and Friends (Nintendo, 2005), where players are given a pet dog which they are required to care for, and The Sims (Maxis, 2000), where players control every detail in the lives of several characters. Sports games could arguably be included

11 in the super genre of simulation games because they simulate the experience of being a sports player and/or managing a sports team. Fighting games require players to fight their way through several adversaries on their way to an end goal. Fighting games embody an odd dichotomy in that they are some of the simplest games to play yet also some of the hardest to play well. Fighting games such as Street Fighter IV (Capcom, 2009) primarily require players to push sequences of buttons to carry out various attacks on opponents. This means that just about anyone can pick up a fighting game, push random buttons, and experience a relative degree of success playing the game. For this reason, fighting games are often referred to as “button mashers.” However, doing well, and often times succeeding in completing the end goal in fighting games is contingent upon mastering sequences of timed button presses. Shooters, as a genre, represent games which primarily require players to shoot things. This is conceptually distinct from fighting games where players primarily use hand to hand combat and weapons such as swords and sticks (Smith, 2006). Shooters are generally sorted into the sub genres first-person shooter and third-person shooter. In first person shooters players play from the visual perspective of the character, as if the player were looking out into the game world through the character's eyes. Third-person shooters, on the other hand, give players a bird's eye view of the battlefield and require players to aim on two dimensions (up/down and left/right) as contrasted to first-person shooters which require players to aim and navigate in three dimensional space.

12 Adventure games are typically described as those where players control a character who sets out on an adventure. In these games, players are faced with smaller goals or quests on their way to completing an end goal. Unlike many other genres, adventure games often place an emphasis on exploration of an open game world (Wolf, 2008, p. 81). This definition is broad and can conceptually be applied to several games that end up in other genre categories, specifically action games and RPG games. Action games are defined as those that require players to take on the role of a protagonist who they must guide through a series of physical challenges. Action games are typically arranged into levels of progressing difficulty that players must complete in order to reach an end goal. Typically Action games have a “boss” or enemy who is more difficult than the average at the end of every level. Much like adventure games, action games are defined broadly and include the sub-genres fighting, shooter, platform, and real time strategy. Fighting, shooter, and real time strategy games are discussed elsewhere in this paper. Platform games can be briefly described as those which require players to navigate an avatar through obstacles. This primarily takes the form of players jumping from one platform to another. One of the most famous video games ever, Super Mario Brothers (Nintendo R&D4, 1985) is a platform game. Quite often adventure games have aspects of action games, and vice versa. In an attempt to resolve this, the genre action/adventure was created. Action/adventure games embody qualities of both action and adventure genres. Players control a character who sets out on an adventure, overcoming obstacles on the way to an end goal, and game play is characterized as being action oriented, typically with an emphasis on combat in some

13 form (Smith, 2006). Games in the action/adventure genre include God of War 3 (SCE Studios Santa Monica, 2010) and Tomb Raider for the PlayStation 3 (Crystal Dynamics, 2012). One of the most recognized yet poorly understood video game genres is role playing games (RPGs). The most defining characteristics of RPGs are that they place a heavy emphasis on players completing quests, and “leveling up” their character(s) (Smith, 2006). Characters in these games have attributes such as strength, defense, stamina, and health points that are set at certain values at the beginning of the game. As players progress through the game they gain “experience points” through completing quests and defeating enemies. Once players reach a designated amount of experience points they “level up” and their character(s) gain a permanent increase in their attributes. It is this leveling up of a character that is usually thought of as the hallmark of RPG games. For many players, leveling up is a primary motivator for playing RPGs. Some RPGs even require players to engage in what is known as “level grinding”; spending a significant amount of time doing nothing but battling enemies for the purpose of leveling up a character. Many players would say that another thing that sets RPGs apart from other games is that they typically have intricate story lines and involve a significant amount of reading in the form of character dialogue. These characteristics are not unique to RPGs but are undoubtedly more common among them. Many RPGs also have characteristics that are typical of other genres. One such game “Mass Effect” (BioWare, 2007) is considered by most to be an RPG because it places heavy emphasis on the

14 completion of quests, leveling up characters, and has an intricate storyline. However, in this game, players largely interact with the game world through shooting things with guns. In this way, Mass Effect (Bioware, 2007) could be considered a shooter. However, in general, if a game has a system of leveling or evolving a character's attributes, and involves the completion of quests, it is typically considered to be an RPG. One of the more recent additions to video game genres is massively multiplayer online (MMO) games such as World of Warcraft (WOW) (Blizzard Entertainment, 2004) and Guild Wars (ArenaNet, 2005). MMOs are video games which people play over the internet with unprecedented numbers of other players physically located across the globe. MMOs provide an interesting mix of competitive and cooperative play that did not exist before the genre came into being (Cole & Griffiths, 2007). In their MMO play, players can join with other members in personal alliances or larger organized groups referred to as “guilds.” Players can also engage in what is called “player versus player” (PVP) and attack, sometimes even hunt other players for bounty. Thus, MMOs provide a social outlet that can create lasting friendships between people over the internet and in some cases in person. Most commonly seen in the form of massively multiplayer RPGs or MMORPGS, MMOs have been the focus of much video game addiction research in recent years (Charlton & Danforth, 2007; Peters & Melesky, 2008). Peters and Melesky (2008) found addictive MMO behavior to be correlated with agreeableness, neuroticism, and extraversion. MMOs are of interest to the current study due to the potential for players to engage in relatively more prosocial play and the juxtaposition of that prosocial play with competitive play and the formation of in versus out groups.

15 There has been very little research conducted on the classification validity of many popular genres. One study used an intuitive approach to genre classification that was replicated in part in the current study (Lucas & Sherry, 2004). In their study on sex differences in video game play, Lucas and Sherry (2004) composed a list of 13 common genres. Along with their other testing measures, they gave subjects this genre list accompanied by short descriptions of each genre and asked them to rate their liking of each on a 7 point Likert scale. After data collection was completed, Lucas and Sherry performed a factor analysis of subjects' scores for different genres and derived three factors they then used as the three levels of their genre variable. The three factors were labeled “traditional” (puzzle, card/dice, quiz/trivia, and classic board games), “physical enactment” (fighter, shooter, sports, and racing games), and “imagination” (role playing, action/adventure, simulation, and strategy games) (Lucas & Sherry, 2004). Another author, Zammitto (2010), attempted to validate a classification system more comparable to the current genre system. In her study, Zammitto constructed an inventory designed to measure preferences for several video game types (action-shooting, action-no shooting, action-fighting, sports, simulation-vehicle, simulation–artificial intelligence, adventure, puzzle, and online). Zammitto's inventory breaks these types down into their various conceptual components which participants then indicate their preferences for, using a four point Likert scale. Scoring is achieved through obtaining composite scores of the components pertaining to each video game type. Most other studies that examine preferences for different video game types simply define several game types and then ask participants to rate their preference for each one. Zammitto's

16 study is the only one that was found during the literature review phase of this study that attempted to divide video game types into their various descriptors. Personality and Video Game Preferences When an individual chooses to play a video game, that person is often making a conscious decision to dedicate a significant portion of time to one activity. The decision to engage in any behavior is affected by situational constructs such as mood, and consistent constructs such as personality (Hartmann & Klimmt, 2006). A person who is feeling depressed may be more likely to choose a video game with dark undertones as compared to someone who is feeling happy. However, much as individuals show consistent favor for certain movie or book genres, individuals tend to make consistent choices in video games as well (Hartmann & Klimmt, 2006). It is plausible that this enduring choice pattern is the result of personality. It follows then that choice of video games may be a reflection of enduring personality traits. A person who is high in trait competitiveness may, for example, show a consistent preference for games that give him/her the opportunity to express that competitiveness (Hartmann & Klimmt, 2006). This supposition is dependent upon the idea of functional equivalency (Hartmann & Klimmt, 2006). Functional equivalency refers to the phenomenon of an object or behavior having a similar cognitive meaning to an individual as another object or behavior. A competitive person may not be interested in competitive video games if engaging in such play does not have the same cognitive meaning to him or her as other outlets of competitive expression. Just as an individual may be competitive in sports but not in chess, so too can personality traits result in different behaviors when it comes to

17 video game play than might be exhibited in other situations (Hartmann & Klimmt, 2006). A good example of this was demonstrated in a study by Bushman and Whitaker (2010) in which belief in a cathartic effect of playing violent video games was found to be positively related to participants’ attraction to such games. This example demonstrates the idea of functional equivalency by essentially showing that angry people are likely to engage in violent video game play if they believe that doing so will offer them the same kind of relief they believe other aggressive expressions might. In this way, the choice to play video games has much to do with what a person expects to get out of doing so. Some video game players cite the expectation that playing games will help them to recover from work related fatigue and daily hassles as their primary reason for playing video or computer games (Reinecke, 2009). Two other factors that may affect the cognitive meaning of video game play are the phenomena of wishful identification (Konijn, Bijvank, & Bushman, 2007) and telepresence (Lachlan & Maloney, 2008). In one interesting study by Konijn, Bijvank, and Bushman (2007), 99 Dutch adolescent males (mean age = 14 years) were partnered with confederate opponents after playing a video game for 20 minutes. The adolescents were given a control they were told would punish their opponents with a loud blast of noise. Results of the study indicated that the amount of punishment administered to opponents was significantly affected by wishful identification with a violent character. Adolescents who played a violent video game and experienced wishful identification toward a violent character were the most aggressive toward their opponents and even administered punishments they were told would cause permanent hearing damage. Social

18 learning theory holds that observing and modeling play a significant role in childhood development (Bandura, 1977). As such, wishful identification with violent video game characters may affect the shaping of children’s personalities. Such identification may be predictive of future behavior, including patterns of violent video game use (Cohen, 2001). According to the ESA (2011) adult video game players in the United States have on average played video games for 12 years of their lives. As such a longstanding presence in people’s lives, video games may play a role in the formation, modification, and maintenance of personality. In addition to the importance of wishful identification, Konijn, Bijvank, and Bushman (2007) also noted that adolescent participants were more likely to identify with violent characters in video games they felt immersed in. Through narrative, realistic environments, and direct control of a protagonist, video games allow players to temporarily immerse themselves in the worlds of the games they play. This immersion, where one feels more like one is existing in a media environment than in one’s own environment, has been termed “presence” or “telepresence” (Lachlan & Maloney, 2008). Depending on their subjective telepresence, players come to identify themselves with characters while they are playing and temporarily experience an alteration of selfperception through association with the character. Players associate with a character’s desirable traits and experience themselves as having such traits themselves (Klimmt, Hefner, & Vorderer, 2009). Personality may affect the extent to which players consider character traits to be desirable and concurrently affect their patterns of video game choice.

19 The Affect-Dependent Theory of Stimulus Arrangement Freud’s psychosexual theory of personality places emphasis on drive reduction and holds that behaviors that result in drive reduction are likely to be repeated and eventually become an integral part of personality (Tyson & Tyson, 1990). Indeed, Freud’s psychosexual stages are primarily concerned with the ability of individuals to effect drive reduction throughout each stage (Engler, 2003). One theory applicable to selective video game exposure that is based on the idea of emotional regulation and drive reduction is known as the affect-dependent theory of stimulus arrangement (Bryant & Davies, 2006). The basic premise of the theory is that individuals selectively expose themselves to media based on their drives and emotions at the time of choosing to engage in game play. The theory is dependent upon the drive reduction idea that individuals strive toward reducing and avoiding negative stimuli/outcomes and increasing the occurrence of positive outcomes. The affect-dependent theory of stimulus arrangement involves four primary elements: excitatory homeostasis, hedonic valence, intervention potential, and messagebehavioral affinity (Bryant & Davies, 2006). Excitatory homeostasis refers to the idea that individuals base their media choices on optimal levels of arousal. According to this thinking, individuals who are highly excited are more likely to choose video games that are relaxing so that they may attain a more homeostatic excitatory state, whereas individuals who are bored may choose games that are higher in excitatory content to counteract their boredom and establish excitatory homeostasis. Intervention potential refers to the ability of a message to capture and hold a person’s attention. It has been

20 postulated that highly engaging messages can disrupt cognitive rehearsals related to emotions and thereby reduce the perceived intensity of those emotions (Bryant & Davies, 2006). Message-behavioral affinity refers to the similarity between the content of the media being consumed and the affect of the individual. It has been shown that messages that have a high degree of similarity to individual affect have a lower chance of altering that affect than do messages that are dissimilar to the individual’s affect. As such, people who are in a bad mood may be more likely to choose video games that are more light hearted and prosocial in an attempt to diminish the negative feelings experienced at that time. The last component of affect-dependent theory is hedonic valence, which refers to the extent to which a message is positive or negative. The affect-dependent theory of stimulus arrangement holds that messages of hedonic value opposite to that of a person’s current affect will reduce that affect. In this way, a person who is depressed may choose to play a video game that is uplifting and happy to change his/her depressive mood. Video games, being high in hedonic valence, are easy sources for individuals to turn to for pleasure and mood regulation (Bryant & Davies, 2006). Measures of Personality The main goal of the current study is to test whether personality traits correlate with predictable patterns of selective exposure to several types of video games. The primary means of accomplishing this goal was through correlating reported video game type preferences with subjects' five factor model (FFM) personality traits as measured by the NEO-Five Factor Inventory (NEO-FFI) (Costa & McCrae, 1992) and with subjects' scores on six scales from the California Psychological Inventory (CPI) (Gough, 1987).

21 The six CPI (Gough, 1987) scales used were: dominance, empathy, intellectual efficiency, self-acceptance, self-control, and socialization. The Five Factor Model of Personality The (FFM) represents a common belief among personality researchers that personality as measured on most inventories can be reduced to five global personality factors (Costa & McCrae, 1985; O'Connor, 2002). The five factors of the FFM have been labeled agreeableness, conscientiousness, extraversion, neuroticism, and openness to experience. In a study by O'Connor (2002) examining the universality of the FFM across personality inventories, factor analyses were performed on 28 popular personality inventories including the Italian version of the 16 Personality Factor (16PF) (Cattell, Eber, & Tatsuoka, 1970), the Myers Briggs Type Indicator (MBTI) (Myers & McCaulley, 1985), and the basic scales of the Minnesota Multiphasic Personality Inventory (MMPI) (Hathaways & McKinley, 1983). The results of O'Connor's study showed that 26 of the 28 personality inventories examined could be reduced to five primary factors that closely resembled those of the FFM, and that the FFM was able to account for an average of 38.8% of the scale variance across inventories. When compared to a mean of 50.1% of the variability being accounted for by the inventories examined, one can see that some data are being lost when reducing other inventories to match the FFM, but the commonality is substantial nonetheless (O'Connor, 2002). The FFM has been found to be robust across a wide range of ages, sexes, and cultures (Markey & Markey, 2010). There has been much debate concerning how best to describe the five dimensions of the FFM (McCrae & John, 1992). It is important to note that researchers have

22 experienced difficulty coming to a consensus regarding how the five factors should be described due to linguistic challenges. Statistical data obtained through factor analysis have offered clear support for the concept that personality as measured by most inventories can be reduced to five conceptually orthogonal factors (O'Connor, 2002). The difficulty in defining these factors arises through limitations of the language they are being described with. Just as one might have difficulty explaining the color of the sky to someone in a language that does not have a word for blue, so too can describing the five factors be difficult because the English language lacks adjectives that would adequately describe them in their entirety (McCrae & John, 1992). Of the five factors, neuroticism is the most agreed upon in definition and is generally conceptualized as “individual differences in the tendency to experience distress, and in the cognitive and behavioral styles that follow from this tendency” (McCrae & John, 1992, p. 1954). Individuals who score high on neuroticism tend be prone to psychopathology stemming from self-consciousness, guilt, depression, and other negative psychopathological symptoms. Those who score low on neuroticism are seen as being calm and even tempered but not necessarily high in psychological health (McCrae & John, 1992). Extraversion has a far broader definition and can be thought of as “venturesomeness, affiliation, positive affectivity, energy, ascendance, and ambition” (McCrae & John, 1992, p. 196). Individuals who score high on extraversion display higher levels of these traits than do those who score low. The latter may be seen as withdrawn, shy, and quiet. Extraversion is often simplified as being a measure of

23 sociability. However, individuals with high scores on extraversion tend to exhibit personality traits that are somewhere between warmth and dominance. It is this tendency toward dominance that undermines the simplistic view of extraversion being a measure of sociability. To counteract the simplification of extraversion as being synonymous with sociability, many researchers have come to describe extraversion as primarily being a measure of positive emotion (McCrae & John, 1992). Agreeableness and conscientiousness are two of the most easily described traits. A high score on agreeableness indicates a person who is friendly, altruistic, and caring, whereas a low score is indicative of a person who is the exact opposite: cold, hostile, and self-centered. Agreeableness is often simplified as being a measure of “good” versus “evil”, although, to avoid using such loaded terms, it may be more appropriate to refer to the polar extremes as being prosocial versus antisocial (McCrae & John, 1992). Similar to agreeableness, conscientiousness has clearly definable polar opposites. A high score on conscientiousness indicates a person who is driven and motivated to achieve whereas a low score indicates a person who is impulsive and experiences difficulty tolerating delayed gratification. The polar extremes of conscientiousness can be described simply as “strong willed” versus “weak willed” (McCrae & John, 1992). The extent to which an individual is able to tolerate delayed gratification and/or is weak or strong willed may relate to an individual's ability to tolerate boredom and other such unpleasant situations. Those who are “weak willed” or low in ability to tolerate delayed gratification may be more likely to play video games as a method of self-regulation (Hartmann & Klimmt, 2006). As mentioned earlier in the discussion of the affect-

24 dependent theory of stimulus arrangement and hedonic valence, video games may serve as a way for some individuals to emotionally regulate themselves. Through the hedonic value of video games, individuals who have difficulty tolerating delayed gratification or prolonged moments of boredom may be more attracted to video games than others. The final factor, openness to experience, is the hardest to describe. Many people think of this factor as representing intelligence because individuals who score high in openness are characterized as being creative, having unconventional thoughts and values, and having wide interests (McCrae & John, 1992). Although this may at face value appear to indicate that openness is related to intelligence, low scores do not necessarily indicate that someone is less intelligent than one who scores high on openness. Individuals who score low on openness are seen as being more conventional in thought and behavior and less appreciative of aesthetics and differing perspectives. As mentioned earlier, there has been relatively little research conducted investigating the effects of personality on video game play, with the exception of the study of violent video games and aggression (Anderson et al., 2004; Ferguson, 2007; Ferguson & Kilburn, 2010). Research has shown that the FFM traits measured by the NEO-FFI (Costa & McCrae, 1992) act as moderators of VVG selection tendencies and the effects that exposure to VVGs have on aggressive behavior. Chory and Goodboy (2011) found openness and extraversion to be positively related to violent video game selection tendencies and agreeableness and neuroticism to be negatively correlated with the tendency to select VVGs. Interestingly, neuroticism appears to be positively correlated with aggressive behavior post exposure to VVGs (Markey & Markey, 2010).

25 These findings suggest that those who score low on neuroticism are more likely to select VVGs and less likely to display aggressive behavior after playing them than are those who score high on neuroticism; whereas those who score high on neuroticism are less likely to select VVGs but more likely to display aggressive behavior after playing them. Conscientiousness and agreeableness have been found to be negatively correlated with aggressive behavior after playing a VVG (Markey & Markey, 2010). Bruggeman and Barry (2002) found that their test subjects who were high in psychoticism showed a tendency toward preferences for violent movies. Psychoticism has been found to be negatively correlated with the FFM traits of agreeableness and conscientiousness (McCrae & Costa, 2002). This finding may suggest that individuals low in agreeableness and conscientiousness may choose violent video games more often than others. In general, research has shown that trait aggression is positively related to neuroticism and negatively related to agreeableness (Sharpe & Desai, 2001). In the literature relevant to video game selection patterns, the FFM is one of the more common forms of personality assessment used. One study conducted by Zammitto (2010) examined the relationship between personality as measured by the NEO-FFI (Costa & McCrae, 1992) and 12 conceptual video game genres: action-fighting, action-no shooting, action-shooting, adventure, artificial intelligence simulation, construction simulation, puzzle, real time strategy, role playing, sports, turn based strategy, and vehicle simulation. Zammitto found neuroticism to be a significant predictor of preference for shooters, action-no shooting, fighting, and sports games. Extraversion was found to be a significant predictor of preference for these same genres, as well as for online play.

26 Openness to experience was found to be significantly negatively related to preference for shooters, sports, and online play and positively related to preference for simulation, adventure, and puzzle games. Agreeableness was found to have a significant negative relationship to preferences for shooters, action-non shooters, fighting, sports, and online games while being positively related to preference for adventure games. Lastly, Zammitto's study found a significant positive relationship between conscientiousness and preference for action-non shooters and puzzle games and a negative relationship between conscientiousness and driving games. The California Psychological Inventory The FFM is a data reduction approach to personality assessment. Although it is true that factor analysis has shown the five factors of the model to be consistent across many personality inventories, some more specifiable data is lost in reducing personality to five broad facets (O'Connor, 2002). To assess comparatively more specific personality traits, the California Psychological Inventory (CPI) (Gough, 1987) was administered to measure subjects' loadings on six scales of interest to this study. The six scales that were used were dominance, self-acceptance, empathy, socialization, self-control, and intellectual efficiency. For the most part, the personality structures represented by each of these scales can be inferred by their titles. Individuals who score high on dominance are assertive, confident, and dominant, whereas those who score low on dominance are unassuming and passive. Individuals who score high on empathy are generally well accepted by others, comfortable about themselves, and able to understand/empathize with the feelings

27 of other individuals. Low scores on empathy are characteristic of individuals who are relatively unempathic and uncomfortable in many social situations. High scores on intellectual efficiency are associated with individuals who are efficient in intellectual tasks and are able to maintain focus on them where other individuals might get bored or discouraged. Low scores on intellectual efficiency are indicative of individuals who have a hard time starting projects or tasks and following them out to completion. Individuals who score high on self-acceptance have good opinions of themselves and see themselves as personally attractive and talented whereas those who score low on self-acceptance doubt themselves, readily accept blame, and often think that others are better than they are. Individuals who score high on self-control are very controlling of their emotions and see themselves as self-disciplined, whereas those who score low on self-control voice their frustrations when annoyed or angry and have strong emotions that they make little attempt to hide. Lastly, high scores on socialization are characteristic of individuals who readily conform and accept rules and regulations whereas those who score low on socialization are unconventional and resist rules and conformity (Gough, 1987). Many video games are highly competitive in nature. Because of this, the CPI (Gough, 1987) scale of dominance, in which high scorers are described as being confident and dominant, is of interest to this study. Research has shown that individuals high in competitiveness show only slightly more competitive behavior in video games unless they are motivated to compete within the game (Vorderer, Hartmann & Klimmt, 2003). However, highly competitive video game players have been shown to have a tendency toward playing competitive video game genres such as real time strategy and

28 shooter games (Hartmann & Klimmt, 2006). This suggests that individuals high in dominance may prefer video games characterized as competitive. Five popular genres that are typically competitive are racing, first-person shooter, fighting, real-time strategy, and online games. The abundance of research that has found that video game play increases aggressive cognition and behavior suggests that video game play may correlate with the CPI (Gough, 1987) self-control scale in which low scores describe individuals who have strong emotions and feelings and have difficulty controlling them (Ferguson, 2007; Ferguson & Kilburn, 2010: Gough & Bradley, 1996). This suggests that individuals low in self-control may be more likely to turn to video games as a means of emotional regulation. Such individuals may prefer action oriented games such as shooters, fighting, and racing games as an outlet for uncontrollable emotions. Online video games allow players to communicate with other people without being in the same physical space as them. Individuals low in emotional regulation may be more likely to play online games because they provide them with a format to express uncontrollable emotions without fear of reprisal. Use of the CPI (Gough, 1987) empathy scale is justified by findings that show that exposure to violent video games is negatively related to empathy (Barlett, Anderson, & Swing, 2009; Funk et al., 2004). This suggests that individuals low in empathy may be more likely to prefer video game genres that typically contain more violence than are individuals high in empathy. First-person shooters and fighters are two such genres.

29 As there has been limited research conducted concerning the relationship between personality and video game preferences, there appears to be no research available to justify using the CPI (Gough, 1987) scales of intellectual efficiency, self-acceptance, and socialization. However, the inclusion of these scales can be justified through consideration of the affect-dependent theory of stimulus arrangement. In many ways, video games provide players with immediate gratification. Through playing video games high in hedonic valence and intervention potential, players can readily regulate their moods and emotions through video game play. Players can also regulate excitatory homeostasis through video game play (Bryant & Davies, 2006). Intellectual efficiency is related to an individual's ability to stay on task, follow through to completion of a task, and not get bored or discouraged. Given this definition, it seems reasonable to suggest that intellectual efficiency may be related to ability to delay gratification. Through immediate gratification and affect regulation, video game play may reduce an individual's ability to tolerate delayed gratification. This would in turn make an individual more susceptible to boredom and becoming discouraged when faced with tasks that require perseverance. This suggests that individuals high in intellectual efficiency may prefer video games that require mental focus and perseverance through completion of tasks such as puzzle and music games, whereas those low in intellectual efficiency may prefer games that provide more immediate feedback and are more action oriented such as those in the popular genres action, action/adventure, fighting, racing, and shooter. Socialization can be similarly justified for inclusion in this study. Individuals who score low on socialization are unconventional, resist rules, and experience difficulty

30 conforming. Many popular video game genres such as role playing games (RPG) and simulation offer players the ability to freely explore a virtual world, without ties to any particular rules or regulations. Perhaps individuals low on socialization would prefer these types of video games because they allow players to operate outside of conventional rules. A significant portion of current video game play occurs over the internet. Playing in this way allows individuals to reach out to other players and establish meaningful relationships that can turn into friendships. It seems reasonable to think that individuals low in socialization might be attracted to the safety of anonymity and indirect communication inherent in internet relations and thereby prefer online games more than others. Individuals who score low on self-acceptance readily doubt themselves and tend to think that others are better than they are. This suggests that such individuals may be less likely to prefer competitive genres such as fighting, racing, real-time strategy, online games, and shooter than are those who are high in self-acceptance. It seems reasonable to suggest that individuals low in self-acceptance might isolate themselves from others. Similar to those low in socialization, individuals low in self-acceptance may prefer online games because they allow them to interact with other individuals without being in the same physical space as them.

31 The Present Study The present study replicated many aspects of the studies by Zammitto (2010) and Lucas and Sherry (2004). The present study used a modified version of Zammitto's (2010) Gaming Preferences Questionnaire to measure preferences for different types of video games. This was done in an attempt to create a more sophisticated and accurate means of measuring video game preferences than has been used in most previous studies. Much like Lucas and Sherry (2004), after obtaining measures of participants' video game preferences, the present study used principal components analysis to examine empirically the existence of differing video game types. To date, there is a clear lack of empirically derived video game categories in the literature. Principle components analysis was performed in an attempt to aid in furthering the understanding of the existence of distinct video game types. Because Zammitto’s questionnaire was constructed as a collection of various video game descriptors, it is important to note that the results yielded by the current study’s principal components analysis demonstrate sets of descriptors that tend to be preferred together. As such, it is more accurate to think of the current study’s video game types as groupings of descriptors that appear to be distinct from each other. Throughout this study the term “video game types” should be read as meaning distinct groupings of video game descriptors. In addition to examining video game preferences, the present study examined the relationships between several personality variables and preferences for different types of video games. Similar to Zammitto (2010), the present study employed the use of the NEO-FFI (Costa & McCrae, 1992) to measure openness, conscientiousness, extraversion,

32 agreeableness, and neuroticism. In addition to this, the CPI (Gough, 1987) was used to measure dominance, empathy, intellectual efficiency, self-acceptance, self-control, and socialization. The relationships between these personality variables and the video game types resulting from the principal components analysis were then examined. Similar to Zammitto (2010), the current study also asked participants to indicate their top three video games of all time. Zammitto used this information to validate her Gaming Preferences Questionnaire by comparing the genres of these top three favorites to the preferred genres indicated by her measurement in a small sub-sample of participants. In the current study, the relationships between genres corresponding to participants' top three favorite games and the measured NEO-FFI (Costa & McCrae, 1992) and CPI (Gough, 1987) personality variables were examined. This was done as a contingency for the possibility that the principle components analysis yielded no significant results. In addition to examining the relationships between personality and preferences for different types of video games, the present study also examined the relationships between sex and video game preferences. Many previous studies have examined the relationship between sex and video game preferences (Consalvo and Treat, 2002; Lucas & Sherry, 2004; Terlecki et. al., 2010). Previous studies have focused almost exclusively on the differences between the sexes in preferences for violent and/or prosocial video games. The present study differs from these previous studies in that it examines sex differences in preferences for all video game types regardless of their violence content or prosocial play.

33 Statement of Problem Previous research on video games is lacking in empirically supported means of classifying video games into differing categories. As a result, it is difficult to measure individual video game preferences with any degree of certainty. Possibly due to this absence of a reliable and valid means of measuring video game preferences, there is a notable lack of research to date that has been performed examining the relationships between personality and video game preferences. Previous research examining the relationships between personality and video game preferences has almost exclusively focused on preferences for violent video games. Similarly, very little research has been performed examining the relationships between sex and video game preferences beyond the exclusive consideration of violent or prosocial games. Because video games have become such a pervasive part of our modern world, played by individuals of all ages, it is important that research be conducted examining the relationships between personality and video game play. Statement of Purpose The purpose of this study was to examine empirically the existence of distinct video game categories and to investigate the relationships between preferences for those categories and several personality traits as measured by the NEO-FFI (Costa & McCrae, 1992) and the CPI (Gough, 1987). Additionally, the present study aimed to examine the relationships between sex and these video game categories. With the exception of violent video games, very little research has been conducted examining the relationships between personality and video game preferences and sex and video game preferences. In addition,

34 previous research has lacked an accurate means of measuring video game preferences. The current study furthers knowledge in these areas. Hypotheses Hypothesis 1 Principle components analysis will yield results indicating the existence of fewer independent video game types than are currently in popular use. Based on the lowest estimate found in the literature (Lucas & Sherry, 2004), it is expected that no fewer than three distinct types of video games will be identified Hypothesis 2 Based on personality traits as measured by the NEO-FFI (Costa & McCrae, 1992) and the CPI (Gough, 1987), individuals will differ in their preferences for the video game types yielded by principle components analysis and in the genres corresponding to their top three favorite video games. Because the video game types yielded by the current study’s principal components analysis may differ from the video game categories examined in other studies, it is difficult to predict how individuals will differ in their preferences for these types of video games. In this way hypothesis two is exploratory in nature. Despite this, it is specifically hypothesized that dominance, neuroticism, and conscientiousness will be positively related to preferences for action oriented games and that empathy, intellectual efficiency, and self-control will be positively related to preferences for games that place emphasis on puzzles and other cognitive challenges.

35 Hypothesis 3 It is hypothesized that a statistically significant difference in preferences for different types of video games will be found between the sexes. Similar to the findings of Lucas and Sherry (2004), it is specifically hypothesized that males will show a preference for more violent, action, and combat oriented games whereas females will show a preference for more prosocial games and less violent, combat oriented games.

36 Chapter 2 METHOD

Participants A total of 312 (126 males and 186 females, mean age = 21, SD = 3.6) undergraduate students enrolled in introductory psychology courses at California State University, Sacramento (CSUS) comprised the sample for this study. Students enrolled in this study in partial fulfillment of the psychology department’s compulsory research participation required of all students enrolled in introductory psychology courses. Materials Demographic Sheet Participants were given a six item demographic sheet (Appendix B). The sheet assessed for age, sex, ethnicity, major focus of study, year in college, and marital status of participants. NEO Five-Factor Inventory (NEO-FFI) The NEO-FFI is designed to assess personality as delineated by the Five Factor Model (FFM) of adult personality (Costa & McCrae, 1992). The FFM is the result of significant research that has suggested that the majority of adjectives used to describe personality can be divided into five broad dimensions (Digman, 1990; Goldberg, 1993; John, 1990). Studies have shown that many popular personality inventories, including the Myers Briggs Type indicator (MBTI) (Myers & McCaulley, 1985), and the basic scales of the Minnesota Multiphasic Personality Inventory (MMPI) (Hathaways &

37 McKinley, 1983) can be reduced to five factors similar to those of the FFM (O’Connor, 2002). These findings support the construct validity of the FFM. The five dimensions of the FFM have been labeled neuroticism, extraversion, openness to experience, agreeableness, and conscientiousness. Neuroticism broadly refers to an individual’s general tendency toward experiencing emotional distress. Extraversion refers to the extent to which an individual is sociable, assertive, active, energetic, and enjoys the company of others. Openness to experience represents the degree to which an individual is open minded, imaginative, curious, and willing to try new things. Agreeablenees refers to the extent to which an individual is friendly and altruistic. Conscientiousness refers to the extent to which an individual is strong willed, determined, purposeful, and actively plans, organizes, and carries out tasks (Costa & McCrae, 1992; McCrae & John, 1992). It is important to remember that these categories are domains and exist on a continuum. The dimensions should not be considered as categorizations with polar opposites but rather as spectrums along which individuals can be placed to indicate the extent to which a certain domain characterizes their personalities. The NEO-FFI (Costa & McCrae, 1992) is an abbreviated version of the NEO Personality Inventory Revised (NEO-PI-R) (Costa & McCrae, 1992). It is intended for individuals aged 17 and older and requires a sixth grade reading level. The test items take the form of first person statements which participants are asked to rate on a five point Likert scale ranging from “Strongly Disagree” to “Neutral” To “Strongly Agree.” The inventory typically takes 10-15 minutes to complete (Costa & McCrae, 1992).

38 In the construction of the NEO-FFI, Costa and McCrae performed a validimax factor analysis on all 240 items of the NEO-PI-R (Costa & McCrae, 1992) and isolated the 12 items with the highest factor loadings for each dimension. Some of these items were replaced with ones that had lower factor loadings to reduce redundancy in wording. The result was an inventory consisting of 60 items; five domain scales consisting of 12 items each. The NEO PI-R measures each of the FFM dimensions and six narrower facets for each dimension; however, as an abbreviated form, the NEO-FFI does not measure these facets (Costa & McCrae, 1992). The Neo-FFI (Costa & McCrae, 1992) is one of the most widely used measures of personality as defined by the FFM (Pytlik Zillig, Hemoenover, & Dienstbier, 2002). Internal consistency coefficients range from .68 (agreeableness), to .86 (neuroticism) (Costa & McCrae, 1992). The inventory also has good test-retest reliability, with 30 month reliabilities ranging from .73 (agreeableness) to .86 (openness to experience) (Murray, Rawlings, Allen, & Trinder, 2003). The NEO-FFI (Costa & McCrae, 1992) scales strongly correlate with the full 48 item domain scales of the NEO-PI-R (Costa & McCrae, 1992). The correlation coefficients between each NEO-FFI (Costa & McCrae, 1992) scale and its respective NEO-PI-R (Costa & McCrae, 1992) scale are: .88 (agreeableness), .89 (conscientiousness), .90 (Extraversion), .93 (neuroticism), and .94 (openness to experience). Costa and McCrae (1985) designed a measure of the FFM using adjective self-reports. This measure yielded five adjective factors that resemble those of the FFM. The correlation coefficients between each of the NEO-FFI (Costa & McCrae, 1992) scales and their respective FFM adjective factors are: .57 (agreeableness),

39 .61 (conscientiousness), .60 (extraversion, .62 (neuroticism), and .56 (openness) (Costa & McCrae, 1992). Each of these were significant at an alpha level of .001. For more information regarding the reliability and validity of the NEO-FFI, please refer to Costa and McCrae (1992). California Psychological Inventory (CPI) The CPI is a personality inventory that assesses individuals on twenty folk concept scales. Gough described these folk scales as being “constructs about personality that all people, everywhere, make use of to comprehend their own behavior and the behavior of others.” (Gough, 1996, p. 2). The twenty folk scales are: dominance, capacity for status, sociability, social presence, self-acceptance, empathy, responsibility, socialization, self-control, good impression, communality, well-being, tolerance, achievement via conformance, achievement via independence, intellectual efficiency, psychological-mindedness, flexibility, and femininity/masculinity. Of the 20 folk scales, the current study made use of six: dominance, empathy, intellectual efficiency, selfacceptance, self-control, and socialization. There are currently four versions of the CPI: The original 480 item CPI (Gough, 1957), the second edition consisting of 462 items (Gough, 1987), the third edition consisting of 434 items (Gough & Bradley, 1996), and the relatively brief 260 item CPI 260 (Gough, 2002). The third edition of the CPI can only be interpreted after submitting answer sheets to Consulting Psychologists Press for scoring. The second edition of the CPI (Gough, 1987) was chosen for use in this study because the ability to hand score data made administration less costly. The second edition of the CPI (Gough, 1987) is

40 comprised of 462 items which participants are asked to rate as True or False or as Agree or Disagree in regard to how they feel each item pertains to them. The CPI is a widely used inventory that has had much research conducted to support its validity (Gough, 1987). The CPI scales used in this study have been shown to be correlated with other inventory scales measuring similar personality constructs and covariates (Gough, 1987). For further information on these correlations and other validity data, please refer to Gough (1987). Internal consistency scores for the scales used in this study are: .79 (dominance), .58 (empathy), .72 (intellectual efficiency), .52 (self-acceptance), .80 (self-control), and .71 (socialization). Test-retest reliability over a one year period for these scales has been found to be: .62 for males and .68 for females (dominance), .56 for males and .58 for females (empathy), .72 for males and .79 for females (intellectual efficiency), .60 for males and .74 for females (self-acceptance), .76 for males and .72 for females (self-control), and .69 for males and .74 for females (socialization) (Gough, 1987). Gaming Preferences Questionnaire The Gaming Preferences Questionnaire (Appendix C) was created by Veronica Zammitto (2010) to measure empirically video game players’ preferences for different popular video game genres. Zammitto consulted six professional game designers who were asked to review items and provide suggestions for bettering the scale. All six of the experts agreed that they would approve of the questionnaire as an appropriate tool for measuring video game preferences. In Zammitto’s study, participants were each asked to indicate their top three favorite games and their favorite video game genre. Following

41 the data collection phase of her study, Zammitto randomly sampled ten percent (55) of participants’ data and compared the genres of their top three favorite games and their indicated favorite genres with the top three genres indicated by their Gaming Preferences Questionnaire scores. She found that the Gaming Preferences Questionnaire was able to predict participants’ favorite genres in 91% (50) of the 55 cases sampled (Zammitto, 2010). For the current study the format of this instrument was changed so that participants were asked to indicate how much each statement applied to them using a five point Likert scale ranging from zero “very little” to five “very much”. The pronoun “I” was removed from each statement and the wording of statements was changed so that participants could rate them based on the extent to which they apply to them rather than their agreeing or disagreeing with each statement. This change was made in an attempt to simplify the inventory and reduce possible confusion among participants. Veronica Zammitto was also consulted for suggested points of revision. She suggested that the conceptual category of “music games” was underrepresented by the questionnaire in its original form. Since the original conceptualization of the questionnaire, music games as a distinct category of video games have become more popular and widely recognized. The questionnaire presented to participants therefore included two additional items intended to capture the characteristics of this category. The final modification made to the original inventory was that the wording of statement number 21 was changed from “I enjoy more taking decision on the fly”(sic) to “Enjoy games that require players to make quick decisions.” This change was made in

42 consideration of the fact that the original questionnaire was created for a Canadian audience. The wording was changed to something considered more germane to American readers. Gaming Patterns Questionnaire A ten item Gaming Patterns Questionnaire (Appendix D) was constructed to assess video game play tendencies among research participants. The questionnaire consisted of two items that assessed amount of time typically spent playing video games, one item that assessed participants’ ages when they first engaged in video game play, three items that assessed preferences for single or multiplayer video game play, one item that assessed current video game systems used, one item that assessed participants’ top three favorite video games, an item that asked participants if they would spend more time playing video games if they had more free time, and one item that assessed the frequency of experiencing motion sickness while playing video games among each participant. Procedure Data for this study were collected in multiple 30 minute sessions conducted by a male researcher. Sessions were held in one of two research rooms that seated up to eight participants. Upon reporting to the research session participants were given a consent form (Appendix E) discussing the potential benefits the study might yield, the amount of time the study would require, and the researcher‘s contact information. Consent forms were stored separately from data packets to protect confidentiality. After signing an informed consent form, each participant was given a packet of materials which he/she was then asked to read carefully and respond to appropriately. Each packet consisted of a

43 demographic sheet, The NEO Five-Factor Inventory (Costa & McCrae, 1992), 195 items from the California Psychological Inventory (Gough, 1987) that comprise the scales of dominance, empathy, intellectual efficiency, self-acceptance, self-control, and socialization, and a 52 item Video Game Preferences Assessment paired with a ten item Gaming Patterns Questionnaire. These materials were presented to each participant in one of six possible permutations. Each permutation consisted of the demographic sheet followed by the NEO-FFI (Costa & McCrae, 1992), 195 CPI (Gough, 1987) items, the Video Game Preferences Assessment, the Gaming Patterns Questionnaire, all presented in one of six possible randomized arrangements. Within each randomized presentation of the materials, the Video Game Preferences Assessment was immediately followed by the Gaming Patterns Questionnaire. After completing their material packets, participants were provided with a written debriefing (Appendix F) and thanked for their participation.

44 Chapter 3 RESULTS

Invalid Packet Exclusion Thirty five of the 312 participants’ data packets were omitted from inclusion in the study. Of these 35 omitted packets, 24 were excluded from the study because participants gave the same response to 40 or more items on the 52 item Gaming Preferences Questionnaire. These packets were determined to have insufficient variability among item responses and were therefore not included in the study. Nine of the 35 packets removed were excluded from the study because of participants’ failing to complete enough of an inventory to make it scoreable. Two final packets were excluded because they had a significant number of invalid item responses. Preliminary Data Analysis After the removal of 35 invalid data packets, the study sample consisted of 277 participants (108 male and 169 female, mean age = 20.96, SD = 3.27). Of the participants, 14.1% were freshmen, 28.3% were sophomores, 42% were juniors, and 15.6% were seniors. Also, the study participants were ethnically diverse. Table 1 illustrates this ethnic diversity.

45 Table 1 Ethnicity of Participants Ethnicity

N

Percent

Caucasian

114

41.5

Asian

61

22.2

Hispanic

56

20.4

African American

21

7.6

Native American

2

0.8

Middle Eastern

1

0.4

Portuguese

1

0.4

Two ethnicities

16

5.6

Three or more ethnicities

3

1.1

Note. N = 275.

Study participants reported a wide range of time they spend on average playing video games every week. Of the 274 participants who provided an average amount of time they believe they spend playing video games every week, 30 reported playing zero hours every week, while the remaining 244 ranged from 0.5 to 50 hours on average spent playing video games every week (M = 5.99, SD = 8.30). As illustrated in Table 2, study participants varied in their reported frequency of video game play. Participants also reported a range of ages when they played video games for the first time. The distribution of ages at which participants first played video games is illustrated in Table 3.

46 Table 2 Reported Frequency of Video Game Play Frequency of Play

N

Percent

Have never played

1

0.4

Rarely

47

17.0

Several times/year

38

13.7

Several times/month

69

24.9

Several times/week

68

24.5

Almost every day

29

10.5

Every Day

25

9.0

Note. N = 277. Table 3 Age When First Played Video Games Frequency of Play

N

Percent

Have never played

1

0.4

5 years or younger

97

35.1

6 to 11 years old

156

56.5

12 to 18 years old

19

6.9

19 to 35 years old

3

1.1

Note. N = 276. No participants circled the response option “Older than 35.”

47 One hundred and forty three (51.6%) participants indicated that they would not choose to spend more time playing video games if they had more free time whereas 134 (48.4%) indicated that they would. One hundred and eighteen (42.6%) participants indicated that they only play multiplayer video games. Only 24 (8.7%) participants indicated that they only play video games on Facebook or another social networking website. In the Gaming Patterns Questionnaire, participants were asked to indicate their top three favorite video games of all time. These data were collected as an additional measure of participants’ video game type preferences. Zammitto (2010) used this approach in her study on video game genre preferences. In her study, Zammitto examined the genre classifications of participants’ top three favorite video games and compared them to participants’ genre preferences as indicated by their scores on her Gaming Preferences Questionnaire. Zammitto found that, in a sub-sample of 55 participants, 91% of participants indicated top three favorite video games that belonged to the same genres as those they showed a preference for in the Gaming Preferences Questionnaire. This suggests a relationship between video game type preference and the types of video games individuals are likely to report as their favorites. In the current study, video game genres for each participant’s indicated top three favorite video games were obtained from a popular video game review website (http://www.ign.com). The distribution of genres corresponding to participants’ reported top three favorite games are illustrated in Table 4.

48 Table 4 Genres of Reported Top 3 Favorite Games

Genre

___1st favorite N (Percent)

__2nd favorite N (Percent)

3rd favorite N (Percent)

1st person shooter

71

(26.6)

59

(22.8)

48

(21.5)

RPG

19

(7.1)

25

(9.7)

16

(7.2)

sports

22

(8.2)

30

(11.6)

14

(6.3)

puzzle

4

(1.5)

9

(3.5)

5

(2.2)

strategy

5

(1.9)

3

(1.2)

8

(3.6)

simulation

16

(6.0)

8

(3.1)

7

(3.2)

action

14

(5.2)

15

(5.8)

16

(7.2)

platform

29

(10.9)

30

(11.6)

20

(8.9)

racing

20

(7.5)

22

(8.5)

22

(9.8)

action/adventure

16

(6.0)

21

(8.1)

19

(8.4)

fighting

16

(6.0)

8

(3.1)

19

(8.4)

music

10

(3.7)

10

(3.9)

11

(4.9)

action/RPG

7

(2.6)

7

(2.7)

8

(3.6)

MMO-action

3

(1.1)

2

(0.8)

3

(1.3)

MMO-RPG

9

(3.4)

4

(1.6)

2

(0.9)

shooter

2

(0.8)

1

(0.4)

3

(1.3)

party

4

(1.5)

4

(1.6)

3

(1.3

Note. N for 1st, 2nd, and 3rd favorite is 267, 258, and 224 respectively. Some games were listed as belonging to sub-genres that were not strongly represented in this sample. These games were listed as belonging to the most relevant super-genre. Example: Action/racing became racing.

49 Principal Components Analysis To assess the existence of distinct video game categories, a principal components analysis (PCA) with a promax rotation of the 52 items from the Gaming Preferences Questionnaire was performed on data collected from 277 participants. The KaiserMeyer-Olkin measure of sampling adequacy indicated that the sample was factorable (KMO = .81). Bartlett’s Test of Sphericity was significant (p = .00), indicating that PCA was appropriate. The PCA yielded 15 components with eigenvalues of one or more. Examination of the scree plot suggested one strong factor and five to seven additional factors with progressively weaker loadings. Based on this observation, five, six, seven, and eight factor solutions were performed and examined for item interpretability. After careful consideration, the eight factor solution was selected because the factors it yielded were the most interpretable of the solutions examined. In total, this eight factor solution accounted for 51.54 percent of the variance. Table 5 illustrates the amount of variance accounted for by each of the eight factors. Several of the components yielded by the eight factor solution were found to correlate with each other. Component 1 was found to be correlated with all other components but four and five. Table 6 illustrates the correlations of the components yielded by the eight factor solution. Table 7 shows the structure coefficients.

50 Table 5 Variance Accounted for by the Eight Factor Solution Yielded by Principal Components Analysis of the 52 items From the Gaming Preferences Questionnaire with a Promax Rotation. Percent of variance

Cumulative percent

Eigenvalue

1

18.0

18.0

9.38

2

9.2

27.2

4.78

3

6.0

33.2

3.12

4

4.7

37.9

2.46

5

4.2

42.1

2.18

6

3.3

45.4

1.70

7

3.2

48.6

1.67

8

2.9

51.5

1.52

Component

51 Table 6 Correlations of the Eight Components Yielded by Principal Components Analysis of the 52 Items from the Gaming Preferences Questionnaire with a Promax Rotation. Component 1 2 3 4 5 6 7 8

1 1.00

2

3

4

5

6

7

8

.31

.32

.12

.11

.26

.25

.26

1.00

-.02

.18

.01

.02

.29

.19

1.00

.02

.27

-.02

.05

.17

1.00

-.11

.07

-.01

.13

1.00

-.05

.13

.13

1.00

.11

.10

1.00

.16 1.00

52 Table 7 Structure Coefficients Based on Principle Components Analysis with a Promax Rotation for the 52 Items from the Video Game Preferences Questionnaire Items

1

42) Story unfolds while you play. .69 37) Enjoy leveling a character. .68 19) Decide evolution of units. .67 17) Conquer, explore, .66 or commercialize. 22) Characters learn abilities .65 38) Manage resources. .62 4) Setting up character stats. .61 49) Enjoy completing quests. .61 32) Games with intelligent life. .59 39) Exploring & establishing. .58 relationships with characters. 13) Big, complex worlds. .57 48) Initial attributes comparably. .56 equal to all players. 5) Fast moving avatars. .49 35) Some events continue. .43 by themselves. 30) Events after turns. .40 11) Allow players to shoot. 7) Guns are extremely important. 28) Use blade weapons. .48 51) Aiming skill is important. 8) Boss at the end of level. .33 16) Mainly kick and punch enemies. 44) Characters' stats have a key .54 role to hit and resist while fighting. 50) Combos for more damage. .39 2) Scary games. 43) Hand-eye coordination. 41) Frequent puzzle solving. 29) Puzzles for their own sake. 14) Only puzzles. 24) Just a few puzzles.

2

Components 3 4 5

6

.31

7

8

.33 .34 .34 .41

.33

.41 .38

.31 .31 .51

.40 .31 .31 .75 .74 .69 .67 .65 .60 .59 .45 .45 .36

.36

.39 .42

.33 .31

.46 .37 .33 .83 .80 .76 .67

.35 .34

53 Structure coefficients based on principle components analysis with a promax rotation for the 52 items from the Video Game Preferences Questionnaire Items

1

2

Components 3 4 5

6

7

8

23) Intellectual challenge. .44 .62 .31 47) Combat is not that relevant. .46 -.31 20) Can be played online. .77 36) Can be played with others. .72 on the internet. 21) Quick decisions. .34 .55 1) Small maps or arenas. .33 .34 .45 52) Music & rhythm. .86 33) Coordinate with music. .85 45) Keep in time with a beat. .35 .82 26) Emulate aspects of reality. .58 40) Control several avatars. .42 .58 .49 25) Make buildings & structures. .37 .40 .51 27) No specific goal. .50 18) Enjoy freedom. .43 .47 10) High score. .54 9) Hints to optimize play. .54 3) Fast paced games. .37 .53 31) Occasional boss. .32 .42 .44 12) Drive or fly something. .42 46) Controlling multiple units. .51 .31 .67 34) Sports games. .58 15) Only one avatar at a time. -.51 6) Move units tactically. .32 .37 .33 .41 ________________________________________________________________________ Note. N = 277. Bold print indicates factor membership. Item wording was abbreviated for concise presentation in this table. For full items please refer to Appendix C. Coefficients of .40 or higher were considered adequate loadings for items to belong to a factor. In some cases items had coefficients of .40 or greater on more than one factor. In these instances, items were considered to belong to all factors on which they had a factor loading no more than .10 less than the greatest factor loading for that item. Item 46 was an exception to this rule because interpretation of factor one yielded a video game type that is commonly characterized by players controlling multiple units.

54 Component 1 consists of 18 items that strongly describe the popular video game genre known as role playing games (RPGs) and was thus labeled “RPG.” RPGs are characterized by their emphasis on intricate storylines (items 42, 32, and 35), players completing quests (item 49), and relatively open worlds that players are free to explore (items 17, 39, 13, and 18). RPGs are one of the few video game types that require players to control more than one character at once (item 46). Many RPGs employ what is known as turn-based battle systems. In these systems characters take turns taking action during battles. When a player character’s turn arrives, the player is usually given an unlimited amount of time to decide what he/she wants the character to do in his/her turn. After the character’s turn is complete, the player waits for the enemy’s turn to be over so that he/she can input commands for the next player character’s turn. Items number 30 and 35 characterize this turn-based battle system. RPGs are also well known for their use of leveling systems. In these systems, players start a game with characters that have attributes such as strength and defense that start at a set point. As the player progresses through the game he/she gains “experience points” through battling enemies and completing quests. Upon gaining a predetermined amount of experience points, characters gain a level, which is paired with increases in their attributes. The various concepts inherent in leveling systems are represented by items 37, 19, 22, 4, 48, and 44. Component 2 consists of 11 items that appear to characterize video games that emphasize fighting and weapons. In consideration of this, Component 2 was labeled “combat.” This category appears to have characteristics of several popular video game

55 genres. Items 11, 7, and 51 characterize shooters, items 28, 8, 44, 50, and 31 characterize some adventure and action/adventure games, and items 16 and 50 characterize fighting games. Component 3 consists of six items that revolve around the solving of puzzles and intellectual challenges. This component was thus labeled “puzzles.” The four items loading on Component 4 appear to relate to online gaming. Items 20 and 36 directly reflect this, while items 21 and 1 refer to aspects of online shooters. Shooter games are commonly played over the internet with multiple players involved in any one game. These games require a great deal of quick decision making (item 21) and players typically engage opponents in small maps or arenas (item one). In consideration of the items loading on it, Component 4 was labeled “online.” Component 5 is composed of three items that clearly describe the popular video game genre called “music”. These games require players keep in time with a beat (item 45) and coordinate their actions with music (item 33). This component was labeled “music.” Component 6 consisted of six items that appear to describe the popular video game genre called “simulation” and was thus labeled “sim.” Simulation games are generally thought of as those that try to emulate aspects of reality (item 26). This emulation of reality can take many forms, from controlling the lives of every person in a small town (item 40) to building an entire civilization (items 40 and 25). Many of the more popular simulation games, such as Sid Meier’s Civilization V (Firaxis, 2010) and The Sims (Maxis, 2000) require players to explore and establish relationships with other

56 characters (item 39). Of all the popular video game genres, simulation games arguably allow the player the greatest degree of freedom over his/her actions (items 27 and 18). Component 7 consists of seven items that appear to describe the popular video game genre of “racing.” Items five and three concern fast paced games and fast moving avatars and item 12 specifically describes driving or flying a craft or vehicle. Racing games, whether played single or multiplayer, are characterized by players competing against opponents of varying difficulty. In single player mode, racing games typically titrate the difficulty level of opponents to match how far the player has progressed in the game (item 31). Many games allow players to gain scores based on their performance but racing games are one of the few where a player’s completion of an end goal is dependent upon getting a high score, which in racing games takes the form of time elapsed from start to finish line (item 10). Because of its similarity to racing games, Component 7 was labeled “racing.” The five items loading on Component 8 are consistent with sports games and hence the component was labeled “sports” (item 34). Sports games typically require players to manage several different characters or teams (items 40, 46, and 15). In these games players are tasked with tactically moving these players around a playing field (item 6). The items belonging to each of the preceding components were compiled into scales measuring preferences for video games belonging to each of the eight categories designated by the PCA. Table 8 shows reliability and descriptive statistics for each of

57 these eight scales. Five of the eight scales had good or acceptable internal consistency. The remaining three scales had questionable or unacceptable internal consistency.

Table 8 Reliability and Descriptive Statistics for the Eight Video Game Preference Scales Resulting from a Principal Components Analysis of the 52 Items from the Gaming Preferences Questionnaire. Number of items

Scale

M (SD)

Alpha

RPG

18

64.39 (12.04)

.88

Combat

10

32.90 (7.85)

.84

Puzzle

6

18.28 (5.17)

.81

Online

4

13.89 (3.75)

.74

Music

3

9.05 (3.49)

.86

Sim

6

19.01 (4.53)

.68

Racing

7

25.77 (4.33)

.61

Sports

5

15.39 (3.34)

.35

Note. N = 277.

Pearson correlations were performed on the eight components to examine the extent to which they are intercorrelated. This analysis showed that the only relationships that were not statistically significant (p < .05) were between music and RPG, puzzle and

58 combat, music and combat, online and puzzle, and online and music. This finding shows that the eight components are largely intercorrelated. Because the eight components were found to be intercorrelated, a second-order principal components analysis with a promax rotation was performed to further examine the factor structure of video game preferences as measured by the 52 items of the Gaming Preferences Questionnaire. The variables used in the second-order PCA were the eight components yielded by the previous PCA. This second-order PCA yielded three components with eigenvalues of approximately one or greater. In total, this three factor solution accounted for 67.2 percent of the variance accounted for by the original eight factor solution. Table 9 shows the variance accounted for by each of the individual components.

Table 9 Variance Accounted for by Each of the Three Components Yielded by a Second-Order Principal Components Analysis (PCA) of the Eight Components Yielded by a PCA Performed on the 52 Items of the Gaming Preferences Questionnaire. Component

Percent of variance

Cumulative percent

Eigenvalue

1

36.2

36.2

2.89

2

18.7

54.9

1.49

3

12.3

67.2

0.99

59 Although Component 3 had an eigenvalue slightly lower than one, it approximated one, and was deemed to be interpretable. Therefore, a three factor solution was chosen instead of a two factor solution. Table 10 shows the correlations of the three components yielded by this three factor solution. Component 2 was correlated with Component 1 (r = .44) and slightly correlated with component 3 (r = .20). Component 3 was not correlated with Component 1 (r = .06). Table 11 shows the structure coefficients.

Table 10 Correlations of the Three Components Yielded by a Second-Order Principal Components Analysis (PCA) of the Eight Components Yielded by a PCA Performed on the 52 Items of the Gaming Preferences Questionnaire. Component 1 2 3

1 1.00

2

3

.44

.06

1.00

.20 1.00

60 Table 11 Structure Coefficients Based on a Second-Order Principle Components Analysis with a Promax Rotation for the 52 Items from the Gaming Preferences Questionnaire. Component

1

2

Combat

.87

.37

Racing

.85

.33

Online

.61

.48

RPG

.58

.83

.30

Sim

.80

.34

Sports

.72

Music Puzzle

3

.78 .35

.76

Note. N = 277. Bold print indicates factor membership. Coefficients of .60 or higher were considered adequate loadings for items to belong to a factor.

Component 1 appears to represent games that are action oriented and was thus labeled “action.” Combat games and racing games are undoubtedly action packed. Recall that the component labeled “online” included four items. Two indicated a preference for online games, and the other two seemed to indicate a preference for online shooters. Because of these items and this component’s second-order factor loading, it may be more appropriate to think of this component as “online shooter” which, in agreement with this second-order component, is highly action oriented.

61 Component 2 is made up of three types of video games that appear to be similar in that they all require a degree of strategy. Simulation games require players to formulate an end goal for their game playing and to strategize their completion of it. Many popular simulation games can easily be considered to be strategy games as well. One such game, Sid Meyer’s Civilization (Firaxis, 2010) requires players to create a civilization and strategize interactions with neighboring civilizations. This game is very comparable to the board game Risk. Despite this, Sid Meyer’s Civilization is often considered to be a simulation game because it allows players to control many aspects of a civilization as if they were a ruling monarch. This includes aspects such as road building, setting tax rates, and determining how much tax revenue to spend on military and education. RPG games often require a great deal of strategy as well. Turn based battle systems, which are common in RPGs, require players to strategically choose their player’s actions and plan ahead in battle, much like a chess player would. Sports games also require a great deal of strategy. Most sports games effectively make players a team manager and an omnipresent entity responsible for deciding what all team players should be doing during game play. Because of the importance of strategy in all these game types, Component 2 was labeled “strategy.” Component 3 appears to be comprised of games that require a great deal of cognitive processing. It is true that games represented by the previous two components require a degree of cognitive processing as well. The distinction to be made here is that Component 3 appears to represent games which are entirely dependent upon cognitive processing. Puzzle games obviously require players to think through complicated puzzles

62 to find solutions. The cognitive nature of music games requires some explanation. Music games require players to follow commands on the screen, and press certain buttons in time with the music. The on screen instructions usually take the form of something that is comparable to music notes. In this way, players are effectively reading music scores. A significant portion of the challenge in music games is that they play in a continuous, unstopping flow. In order to play a music game well, players must be able to “read” and respond to the on screen instructions in a fast pace flow. This requires a great deal of mental focus. New players will often complain that the hardest aspect of learning how to play a music game is being able to read what is being shown on the screen fast enough. One might be able to move one’s fingers and/or feet fast enough to perform the movements required to succeed in a music game, but a significant challenge is presented by being required to read and pair appropriate movements with what is read. Because both puzzle and music games strongly emphasize cognitive processing skills, Component 3 was labeled “cognitive.” The items that comprise the scales belonging to each component yielded by the second-order PCA were compiled into three scales intended to measure preference for action, strategy, and cognitive games. Cronbach’s alpha levels for these three components were good and were much better than the alpha levels for the original eight components. Table 12 shows reliability and descriptive statistics for these three scales.

63 Table 12 Reliability and Descriptive Statistics for the Three Video Game Preference Scales Resulting from a Second-Order Principal Components Analysis (PCA) of the Eight Factors Yielded by a PCA of the 52 Items from the Gaming Preferences Questionnaire. Number of items

Scale

M (SD)

Alpha

Action

21

72.55 (13.00)

.86

Strategy

29

98.80 (17.58)

.90

Cognitive

9

27.33 (7.11)

.82

Note. N = 277.

Canonical Correlations A canonical correlation analysis was used to explore the relationships between personality variables and video game preferences. The dependent variables were preference for action, strategy, and cognitive processing games, as defined by the preceding second-order principal components analysis. The predictor variables were neuroticism, extraversion, openness to experience, conscientiousness, and agreeableness as measured by the NEO-FFI (Costa & McCrae, 1992), and dominance, empathy, intellectual efficiency, self-acceptance, self-control, and socialization as measured by the California Psychological Inventory (Gough, 1987). With 277 cases in the analysis, the relationship between the sets of variables was statistically significant, Wilks’ Lambda = .74, Rc2 = .26, approximate F(22, 775.55) = 2.57, p < .001. The dimension reduction analysis indicated that only the first two

64 functions were statistically significant; hence, only those first two functions were extracted and interpreted. Percentage of variance explained, eigenvalues, and the squared canonical correlations for the two functions are shown in Table 13. The first function accounted for approximately 57.88 percent of the explained variance and the second function added 32.26% to that. These two functions combined accounted for approximately 90 percent of the explained variance. The Cramer-Nicewander (1979) index indicated that 12.73 percent of the variance of the dependent variates was explained by the predictor variates.

Table 13 Cumulative Percentage of Explained Variance, Eigenvalues, and Squared Canonical Correlations for the Two Canonical Functions Function

Eigenvalue

Percent variance explained

Squared canonical Correlation

1

0.19

57.88

.16

2

0.11

32.26

.10

The structure coefficients for the two functions for the predictor and dependent variables are shown in Table 14 and Table 15, respectively. The first predictor function is associated with higher levels of self-control, agreeableness, and openness to experience and lower levels of extraversion; the first dependent function is associated with higher preference for cognitive processing games and lower preference for action games. This

65 first function appears to indicate that being friendly yet emotionally and socially reserved is predictive of preference for cognitive processing games and is negatively correlated with preference for action games. The second predictor function is associated with higher levels of openness, conscientiousness, and extraversion; the second dependent function is associated with higher preference for cognitive processing games and action games. The second function appears to indicate that being open to trying new experiences, extraverted, and motivated to achieve is predictive of preferences for both cognitive processing and action games.

66 Table 14 Structure Coefficients for Predictor Canonical Variates for the Two Functions Predictor variable

Function 1

Function 2

Openness

.37

.55

Conscientiousness

.02

.32

-.26

.35

Agreeableness

.47

.13

Neuroticism

.33

.09

Dominance

-.11

-.01

Empathy

.29

.23

Intellectual Efficiency

.20

-.23

-.08

-.21

Self-Control

.59

-.19

Socialization

.15

-.09

Extraversion

Self-Acceptance

67 Table 15 Structure Coefficients for the Dependent Canonical Variates for the Two Functions. Dependent variable

Function 1

Function 2

Action

-.79

.61

Strategy

-.24

.34

Cognitive processing

.56

.81

Multivariate Analysis of Variance To examine the relationships between genres of stated top three favorite video games and personality, three one-way MANOVAs were performed with genres corresponding to participants’ stated first, second, or third favorite video games as the independent variables and with neuroticism, extraversion, openness to experience, conscientiousness, and agreeableness as measured by the NEO-FFI (Costa & McCrae, 1992), and dominance, empathy, intellectual efficiency, self-acceptance, self-control, and socialization as measured by the CPI (Gough, 1987) as the dependent variables. Several of the genre groups corresponding to participants’ first, second, and third favorite video games were of insufficient sample size to be included in the analyses. As suggested by Hair et al. (2010), a sample size of 20 was determined to be sufficient for inclusion in the analyses. The first analysis was a 5-group one-way between-subjects MANOVA which used genres corresponding to participants’ stated first favorite video games (RPG, shooter, sports, platform and racing) as the independent variable. A total of 170 cases were

68 included in this analysis. Genre groups were distributed as follows: RPG (15.3%), shooter (42.9%), sports (12.9%), platform (17.1%), racing (11.8%). Box’s M test was statistically significant ( p < .05), indicating unequal variance/covariance of the dependent variables across genre groups. This necessitated the use of Pillai’s Trace to determine the multivariate effect. Pillai’s Trace indicated that the dependent variate was significantly affected by genres corresponding to participants’ first favorite video games, Pillai’s Trace = .471, F(44, 632) = 1.92, p < .001, 1 - Wilks’ Lambda = .411. Univariate ANOVAs were conducted on each dependent variable to determine which were significantly affected by genres corresponding to participants’ first favorite video games. All the dependent variables were evaluated against a Bonferroni adjusted alpha level of .0045 (.05 divided by 11). Genres corresponding to participants’ first favorite video games had a significant effect on openness to experience scores, F(4, 165) = 5.15, p = .001,

2

= .11. Univariate

effects for the remaining dependent variables were not statistically significant. Tukey post-hoc comparisons of genres corresponding to participants’ first favorite video games for the openness measure indicated that participants whose first favorite video game was an RPG had significantly higher openness scores than did those whose first favorite game was a racing, shooter, sports, or platform game. Table 16 illustrates these group differences.

69 Table 16 Differences in Openness Scores between Those Who Indicated a First Favorite Video Game that was an RPG and Those Whose First Favorite was a Racing, Shooter, Platform, or Sports Game. Genre group

Mean

Standard deviation

95 percent Significance of difference confidence interval from RPG group

RPG

33.15

6.08

[30.99, 35.32]

Shooter

28.78

5.68

[27.49, 30.07]

.007

Sports

27.18

4.94

[24.83, 29.54]

.003

Platform

27.79

5.76

[25.74, 29.84]

.005

Racing

27.00

4.95

[24.53, 29.47]

.003

Note. N’s for RPG, shooter, sports, platform, and racing were 26, 73, 22, 29, and 20 respectively. All genre groups in this table were compared to the RPG group. The significance levels refer to the differences between the RPG group and comparison groups. The second analysis was a 6-group one-way between-subjects MANOVA which used genres corresponding to participants’ stated second favorite video games (action/adventure, RPG, shooter, sports, platform, and racing) as the independent variable. A total of 195 cases were included in this analysis. Genre groups were distributed as follows: action/adventure (10.8%), platform (15.4), RPG (16.4%), racing (11.3%), shooter (30.7%), and sports (15.4). Box’s M test was not statistically significant (p > .05). Wilks’ Lambda indicated that the dependent variate was significantly affected by genres corresponding to participants’ second favorite video games, Wilks’ Lambda =

70 .648, F(55, 832.14) = 1.49, p = .01. Univariate ANOVAs were conducted on each dependent variable to determine which were significantly affected by genres corresponding to participants’ second favorite video games. All the dependent variables were evaluated against a Bonferroni adjusted alpha level of .0045 (.05 divided by 11). Genres corresponding to participants’ second favorite video games had a significant effect on openness to experience scores F(5, 189) = 5.76, p < .001,

2

= .13. Univariate

effects for the remaining dependent variables were not statistically significant. Tukey post-hoc comparisons of genres corresponding to participants’ second favorite video games for the openness measure indicated that participants whose second favorite video game was an RPG had significantly higher openness scores than those whose second favorite video game was a racing, shooter, or sports game. Table 17 illustrates these differences.

71 Table 17 Differences in Openness Scores among Those Who Indicated a Second Favorite Video Game that was an RPG and Those Whose Second Favorite was a Racing, Shooter, or Sports Game. Genre difference group Mean

Standard

95 percent

deviation

confidence interval

Significance of from RPG group

RPG

33.22

5.71

[31.24, 35.20]

Racing

26.45

4.62

[24.07, 28.84]

.000

Shooter

29.57

5.81

[28.12, 31.01]

.043

Sports

26.43

5.32

[24.45, 28.42]

.000

Note. N’s for RPG, racing, shooter, and sports were 32, 22, 60, and 30 respectively. All genre groups in this table were compared to the RPG group. The significance levels refer to the differences between the RPG group and comparison groups.

The third analysis was a 5-group one-way between-subjects MANOVA which used genres corresponding to participants’ stated third favorite video games (fighting, RPG, shooter, platform, and racing) as the independent variable. A total of 137 cases were included in this analysis. Genre groups were distributed as follows: fighting (14.6%), platform (14.6%), racing (16.1%), RPG (17.5%), and shooter (37.2%). Box’s M test was not statistically significant (p > .05). Wilks’ Lambda indicated that the dependent variate was not significantly affected by genres corresponding to participants’ third favorite video games, Wilks’ Lambda = .66, F(44, 468.70) = 1.21, p > .05.

72 Differences between the Sexes To test the hypothesis that preferences for different types of video games differ between the sexes, three Pearson correlations were performed to assess the relationships between sex and the three components yielded by the second-order principal components analysis of this study (preference for action, strategy, and cognitive games). All three Pearson correlations were statistically significant, indicating that females were more likely than males to prefer cognitive games whereas males were more likely than females to prefer action and strategy games. Table 18 illustrates the coefficients and alpha levels of these relationships.

Table 18 Coefficients and Alpha Levels for Three Pearson rs Performed to Examine the Relationships Between Sex and Preferences for Action, Cognitive, and Strategy Games. Dependent variable

Sex Mean

SD

r

p

10.37

1.84

-.41

.001

Strategy

8.56

1.74

-.18

.002

Cognitive

6.09

1.62

.30

.001

Action

Note. N = 277. Sex was coded as 0 = male, 1 = female.

To further test the hypothesis that preferences for different types of video games differ between the sexes, three Chi-Square tests were performed to examine the relationships between sex of participants and genres corresponding to participants’ first,

73 second, and third favorite video games. Table 19, 20, and 21 show the observed frequencies of the three Chi-Squares. All three tests were statistically significant, indicating that genres corresponding to participants’ first, second, and third favorite video games differed between the sexes. Table 22 illustrates the Chi-Square statistic and alpha level for each Chi-Square test performed.

74 Table 19 Observed Frequencies for the Chi-Square Performed to Examine the Relationships between Participants’ Sex and Genres Corresponding to Participants’ First Favorite Video Games. Sex Genre

Male

Female

Shooter

44

29

73

RPG

14

12

26

Sports

15

7

22

Puzzle

0

4

4

Strategy

3

2

5

Simulation

0

16

16

Action

6

8

14

Platform

3

26

29

Racing

1

19

20

Action/Adventure

6

10

16

Fighting

7

9

16

Music

0

10

10

MMO Action

2

1

3

MMO RPG

5

4

9

Party

1

3

4

Note. N = 267.

Total

75 Table 20 Observed Frequencies for the Chi-Square Performed to Examine the Relationships between Participants’ Sex and Genres Corresponding to Participants’ Second Favorite Video Games. Sex Genre

Male

Female

Shooter

34

26

60

RPG

19

13

32

Sports

15

15

30

Puzzle

0

9

9

Strategy

2

1

3

Simulation

0

8

8

Action

5

10

15

Platform

3

27

30

Racing

6

16

22

Action/Adventure

13

8

21

Fighting

1

7

8

Music

2

8

10

MMO Action

2

0

2

MMO RPG

3

1

4

Party

1

3

4

Note. N = 258.

Total

76 Table 21 Observed Frequencies for the Chi-Square Performed to Examine the Relationships between Participants’ Sex and Genres Corresponding to Participants’ Third Favorite Video Games. Sex Genre

Male

Female

Shooter

29

22

51

RPG

13

11

24

Sports

7

7

14

Puzzle

3

2

5

Strategy

8

0

8

Simulation

0

7

7

Action

8

8

16

Platform

2

18

20

Racing

7

15

22

Action/Adventure

7

12

19

Fighting

10

10

20

Music

1

10

11

MMO Action

3

0

3

MMO RPG

2

0

2

Party

0

3

3

Note. N = 225.

Total

77 Table 22 Chi-Square Statistics and Alpha Levels for Each of Three Chi-Square Tests Performed to Examine the Relationships between Participants’ Sex and Genres Corresponding to Participants’ First, Second, and Third Favorite Video Games. 2 N df p Top1 x sex

267

14

65.84

.001

Top2 x sex

258

14

51.67

.001

Top3 x sex

225

14

46.51

.001

Note. Top1, Top2, and Top3 represent first, second, and third favorite game, respectively.

Despite the fact that these Chi-Square tests yielded statistically significant main effects, many of the group samples were of insufficient size to examine any simple effects. It was determined that groups with fewer than 10 participants in them were of insufficient sample size to examine their related simple effects. As a result, only the genre groups corresponding to participants’ reported first, second and third favorite video games that had ten or more males and ten or more females in them were further analyzed for simple effects. To test whether the observed differences between the frequencies of male and female participants who named first, second, or third favorite video games belonging to one of these genre groups were greater than would be expected by chance, several one-way Chi-Square analyses were conducted. Of the 15 different genre groups corresponding to participants’ identified first favorite video games, only two (shooter and RPG) had a sufficient number of participants to be analyzed via one-way Chi-Square. Among genre groups corresponding to

78 participants’ second stated favorite video games, shooter, sports, and RPG were the only ones with sufficient sample size. The genre groups corresponding to participants’ third favorite video games that had sufficient sample size were shooter, RPG and fighting. The one-way Chi-Squares that were performed on these groups were not found to be statistically significant at an alpha level of .05, indicating that the observed differences between the sexes in genre groups examined were not different enough to rule out that they could have occurred by chance. Table 23 illustrates the observed frequencies of males and females for each of the genre groups examined via one-way Chi-Square analyses.

79 Table 23 Observed Frequencies of Male and Female Participants in Each of the Genre Groups Corresponding to Participants’ First, Second, and Third Favorite Video Games Examined via One-Way Chi-Square Analyses. Male

Female

Total

p

Shooter

44

29

73

n.s.

RPG

14

12

26

n.s.

First favorites

Second favorites Shooter

34

26

60

n.s.

RPG

19

13

32

n.s.

Sports

15

15

30

n.s.

Shooter

29

22

51

n.s.

RPG

13

11

24

n.s.

Fighting

10

10

20

n.s.

Third favorites

Note. n.s. signifies relationships that were not significant at p = .05.

80 Relationships Not Addressed by the Study Hypotheses Additional analyses were conducted to examine relationships that were not stated in this study’s original hypotheses. Several Pearson rs were performed that yielded statistically significant relationships. Table 24 illustrates the correlation coefficients and alpha levels of each of these relationships. Table 25 shows the descriptive statistics for each variable involved in these significant relationships. Results indicated a positive correlation between age and participants reporting that they would spend more time playing video games if more free time was available to them. Male participants indicated more frequent video game play and more hours spent playing video games in a typical week than did females. Female participants reported experiencing motion sickness while playing video games more often than did male participants. Participants who indicated that they only play multiplayer video games reported experiencing motion sickness while playing video games more frequently than did those who did not indicate that they only play multiplayer games. Participants who indicated that they would choose to spend more time playing video games if more free time was available to them indicated significantly more frequent video game play and significantly more hours spent playing in a typical week than did those who would not choose to play more. Reported frequency of video game play was found to be positively correlated with preference for action and strategy games as defined by the second-order PCA of this study. However, reported frequency of video game play was also found to be negatively correlated with preference for cognitive games as defined by the second-order PCA of this study. A positive correlation was also found

81 between reported hours spent playing in a typical week and preference for action and strategy games as defined by the second-order PCA of this study. Lastly, participants who indicated that they would choose to spend more time playing video games if they had more free time and participants who indicated that they only play multiplayer games were more likely to prefer action games as defined by the second-order PCA of this study.

82 Table 24 Correlation Coefficients and Alpha Levels for Several Pearson rs that Yielded Statistically Significant Relationships That Were Not Addressed by the Study’s Hypotheses. Age Age

Sex

Play Weekly frequency hours

Motion sickness

Only More multiplayer

1

Sex

1

Play frequency

-.50 (.00)

Weekly hours

-.47 (.00) .65 (.00)

Motion sickness

.15 (.01)

More

1 1 1

.12 (.04) --------- .43 (.00) .30 (.00)

Only multiplayer

---------

Only Facebook

---------

1 .13 (.03)

Action

-.41 (.00) .43 (.00)

.32 (.00)

Strategy

-.18 (.00) .24 (.00)

.22 (.00)

1

.16 (.01) .16 (.01)

Cognitive .30 (.00) -.13 (.03) Note. Dashes indicate relationships that were examined via Chi-Square. These analyses are reported later in this section. Sex was coded as 0 = male and 1 = female. The caption “play frequency” refers to subjects’ reported frequencies of video game play. “Weekly hours” refers to the number of hours participants reported playing video games in a typical week. “Motion sickness” refers to how often participants reported experiencing motion sickness when playing video games. “More” refers to participants’ responses to the question “If I had more free time I would choose to spend more time playing video games” and was coded as 0 = false and 1 = true. “Only multiplayer” refers to participants’ responses to the question “I only play video games with other people in the same room or over the Internet” and was coded as 0 = false and 1 = true. “Only Facebook” refers to participants’ responses to the question “I only play video games on Facebook or another similar social website” and was coded as 0 = false and 1 = true. “Action”, “strategy”, and “cognitive” refer to preferences for the three video game types yielded by the secondorder principal components analysis of this study.

83 Table 25 Descriptive Statistics for All Variables Involved in Significant Relationships Explored in Addition to Relationships Specifically Addressed in the Hypotheses of This Study. Mean

SD

N

Age

20.96

3.27

277

Sex

0.61

0.49

277

Frequency

3.24

1.51

277

Weekly

5.99

8.30

274

Motion sickness

0.41

0.74

276

More

0.48

0.50

277

Only multiplayer

0.43

0.50

277

Only Facebook

0.09

0.30

277

10.37

1.84

277

Strategy

8.56

1.74

277

Cognitive

6.09

1.62

277

Action

84 To examine the relationships between several dichotomous variables, three ChiSquare analyses were performed. The first Chi-Square test examined the relationship between sex and participants’ indications that they would or would not choose to spend more time playing video games if more free time was available to them. Table 26 illustrates the observed frequencies of this Chi-Square. The relationship was found to be statistically significant, 2 (2, N = 277) = 11.50, p = .001, indicating that males were more likely than females to indicate that they would choose to spend more time playing video games if they had more free time.

Table 26 Observed Frequencies for the Chi-Square Performed to Examine the Relationship between Sex and Participants’ Indications That They Would or Would Not Choose to Spend More Time Playing Video Games if More Free Time Was Available to Them. Would play more

Sex Male Female

Total

False

42

101

143

True

66

68

134

Total

108

169

277

The second Chi-Square test examined the relationship between sex and participants’ indications that they do or do not only play video games on Facebook or another similar social networking site. Table 27 shows the observed frequencies of this

85 Chi-Square analysis. The relationship was found to be statistically significant, 2 (2, N = 277) = 17.42, p < .001, indicating that female participants were more likely than male participants to indicate that they only play video games on Facebook.

Table 27 Observed Frequencies for the Chi-Square Performed to Examine the Relationship between Sex and Participants’ Indications That They Do or Do Not Only Play Video Games on Facebook or another Similar Social Networking Site. Only Facebook

Sex Male Female

Total

False

107

146

253

True

0

23

23

Total

107

169

276

A final Chi-Square test was used to examine the relationships between sex and participants’ indications that they do or do not exclusively play multiplayer video games. The relationship was not found to be statistically significant.

86 Chapter 4 DISCUSSION

The primary goals of this study were to collect empirical evidence supporting the existence of several distinct types of video games, and to explore the relationships between participants' preferences for those video game types and sex, the personality traits openness, conscientiousness, extraversion, agreeableness, and neuroticism as measured by the NEO-FFI (Costa & McCrae, 1992) and dominance, empathy, intellectual efficiency, self-acceptance, self-control, and socialization as measured by the CPI (Gough, 1987). The literature on video game play patterns and personality demonstrates the importance of investigating the relationships between these two variables. Preferences for various types of video games have been linked to several five factor model personality traits (Chory & Goodboy, 2011; Markey & Markey, 2010; Zammitto, 2010). However, little research has been conducted using other personality traits, such as those measured by the CPI (Gough, 1987). Additionally, much of the research has focused specifically on the relationships between violent video game play and sex (Lucas & Sherry, 2004) and aggressive behavior (Anderson et al., 2004; Ferguson, 2007; Ferguson & Kilburn, 2010). Hypothesis One Hypothesis one, that the number of distinct types of video games is fewer than are identified by the current genre classification system, was tested through principle components analysis of participants' preferences for various video game descriptors.

87 Principle components analysis of participants' preference ratings for various descriptors contained in the Gaming Preferences Questionnaire yielded eight different types of video games. These eight were labeled combat, racing, online/online shooter, RPG, simulation, sports, music, and puzzle. An examination of the relationships among these eight components indicated that they were highly intercorrelated. Because of this, a secondorder principle components analysis using these original eight components was performed. Similar to Lucas and Sherry (2004), the second-order principal components analysis of the current study yielded only three distinct video game categories. These three components were labeled “action,” “strategy,” and “cognitive processing” for the unifying qualities of the types of games that composed them. This is different from the factors found by Lucas and Sherry (2004), which were labeled “traditional,” “physical enactment,” and “imagination.” Nevertheless, there are many similarities and differences between the genres that composed the factors found by Lucas and Sherry (2004) and the video game types that comprised the second-order components yielded by the current study. Table 28 illustrates these comparisons.

88 Table 28 A Comparison of the Factors Found by Lucas and Sherry (2004) and the Second-Order Principal Components Found in the Current Study. The current study

Lucas and Sherry (2004)

Action combat racing online shooter

Physical enactment fighter racing/speed shooter sports

Strategy RPG simulation sports

Imagination fantasy/RPG simulation strategy action/adventure

Cognitive processing music puzzle

Traditional card/dice classic board games quiz/trivia puzzle arcade

Lucas and Sherry's physical enactment factor is very similar to the current study's action component. Both of these categories describe games that are action oriented and revolve around players controlling their avatars through the completion of physical challenges. The current study's strategy component compares to Lucas and Sherry's (2004) imagination factor. The only way these two are different is the inclusion of action/adventure games in Lucas and Sherry's (2004) factor. This may be a reflection of the fact that Lucas and Sherry (2004) used several popular conceptual video game genres

89 rather than constructing video game categories from various descriptors as the current study did. If Lucas and Sherry (2004) had employed a similar technique, they might have found a category similar to the current study's “combat” which contains descriptors that characterize action, action/adventure, fighter, and shooter games. Lucas and Sherry's (2004) traditional factor is the most different from the components of the current study. An examination of the genres that composed the traditional factor shows that card, dice, trivia, and board games were included in this category. A review of the descriptors contained in the current study's Gaming Preferences Questionnaire (Appendix C) reveals that the current study failed to include descriptors of these types of games. If the current study had included these descriptors, it is possible that the component “cognitive processing” would have been more similar to Lucas and Sherry's (2004) traditional factor. The primary goal of hypothesis one was to identify several distinct types of video games. The description of the current study's three video game types as distinct should be made with caution. Although the three second-order components were less intercorrelated than were the original eight components, the component “strategy” was significantly correlated with the components “cognitive processing” and “action.” Because of this, hypothesis one was only partially supported by the findings of the current study. It was supported by the fact that fewer types of video games than are currently in popular use were defined, however, it was not supported by the fact that the three categories found were so intercorrelated. Because the current study examined the existence of distinct video game types through the measurement of participants' video game preferences, these relationships may

90 be a reflection of the fact that individuals might tend to enjoy and prefer numerous types of video games. Perhaps the original eight components yielded by this study were intercorrelated not because they share qualities but because individuals who prefer certain types of games may tend to prefer others as well. This suggests that preferences for various video games may not be an adequate means of detecting the existence of distinct video game types. Hypothesis Two The second hypothesis of this study was that video game preferences would differ between participants based on their scores on the NEO-FFI (Costa & McCrae, 1992) and CPI (Gough, 1987) scales used in this study. This hypothesis was tested in two parts: first, the relationships between the personality traits measured in this study and the three video game types yielded by the second-order principle components analysis were explored. In the second part of testing this hypothesis, the relationships between these same personality traits and the genres corresponding to participants' first, second, and third favorite video games were examined. The first part of testing hypothesis two was carried out by conducting a canonical correlation analysis using participants' preferences for the three video game types yielded by second-order PCA as the dependent variables and the personality traits measured by the CPI (Gough, 1987), and the NEO-FFI (Costa & McCrae, 1992) as the predictor variables. Two functions were found to be statistically significant and were extracted for further analysis. However, interpretation of these functions should be made with caution because they both had eigenvalues that were significantly lower than one. Function one

91 had an eigenvalue of 0.19 and function two had an eigenvalue of 0.11, indicating that the variance explained by the functions was significantly lower than the estimated error variance for each. In addition, the Cramer-Nicewander (1979) index indicated that only 12.73 percent of the variance of the dependent variables was explained by the predictor variables. The first predictor function yielded by the canonical correlation analysis was associated with higher levels of self-control, agreeableness, and openness to experience and lower levels of extraversion, while the first dependent function was associated with higher preference for cognitive processing games and lower preference for action games. This appears to indicate that participants who were introverted and controlling of their emotions, yet friendly, caring, unconventional, and creative were likely to prefer cognitive processing games and have low preference scores for action games. This is congruent with Chory and Goodboy (2011) who found agreeableness to be negatively correlated with preference for violent video games which are rich in action content. This function also confirms many of Zammitto's (2010) findings. Zammitto found openness to be negatively related to preference for shooters and sports games and positively related to preference for puzzle games. Contrary to the findings of Zammitto (2010) and the current study, Chory and Goodboy (2011) found openness to be positively related to preference for violent video games, which are high in action content. This first function is also in agreement with Zammitto's (2010) findings that preferences for shooters, action non-shooters, fighting, and sports games, which are all high in action content are negatively related to agreeableness and positively related to extraversion. Similarly,

92 Chory and Goodboy (2011) found extraversion to be positively related to preferences for violent video games and agreeableness to be negatively related to preferences for violent video games. Zammitto (2010) also found agreeableness and openness to experience to be positively related to preferences for adventure games, which are often high in action content. This first function does not support these findings. The second predictor function was associated with higher levels of openness to experience, conscientiousness, and extraversion; the second dependent function was associated with higher preference for cognitive processing games and action games. This suggests that participants who are more extraverted, unconventional, creative, strong willed, and purposeful are more likely than others to prefer both cognitive processing and action games. Like Chory and Goodboy (2011) and Zammitto (2010), this function suggests that extraversion is positively related to preferences for games that are high in action content; however, contrary to both of these studies, this function also suggests that individuals who are high in extraversion also prefer cognitive processing games. Zammitto (2010) found a negative relationship between openness to experience and preferences for video games that are high in action content and a positive relationship between openness to experience and preferences for puzzle and adventure games. Function two is in agreement with both of these findings. However, Chory and Goodboy (2011) found openness to experience to be positively related to preferences for violent video games, which are high in action content. Lastly, Zammitto (2010) found conscientiousness to be positively related to preferences for action non-shooter games and puzzle games. Function two is in agreement with this finding.

93 The second part of testing hypothesis two was accomplished by conducting three one way MANOVAs, each using the personality traits measured in this study as the dependent variables and genres corresponding to participants' first, second, and third favorite video games as the independent variable for the first, second, and third MANOVA respectively. Only the first two MANOVAs yielded statistically significant results. The first MANOVA found that participants whose first favorite video game was an RPG had significantly higher openness to experience scores than those whose first favorite game was a racing, shooter, platform, or sports game. These findings are in partial agreement with Zammitto (2010) who found openness scores to be negatively related to preferences for shooter and sports games. However, contrary to the findings of the current study, Chory and Goodboy (2011) found openness to be positively related to preference for violent video games, which the majority of shooters would be classified as. The second MANOVA indicated that participants whose second favorite video game was an RPG had significantly higher openness scores than those whose second favorite game was a racing, shooter, or sports game. Because sports, racing, and RPG games are not associated with any inherent amount of violent content, and because most research concerning video games has focused on violent video games, there is little research available to compare to these findings. Hypothesis Three Hypothesis three, that there would be observable differences in preferences for different types of video games between the sexes, was tested in two parts. In the first part

94 the relationships between sex and preferences for the three video game types yielded by the second-order PCA of this study were examined via Pearson correlations. In the second part of testing hypothesis three, three Chi-Squares were performed to examine the relationships between sex and genres corresponding to participants' first, second and third favorite video games. The three Pearson rs used to examine the relationships between sex and preferences for the three types of video games yielded by second-order PCA were all found to be statistically significant. These correlations indicated that female participants were more likely than male participants to prefer cognitive processing games whereas male participants were more likely than female participants to prefer action and strategy games. These findings are in partial agreement with Consalvo and Treat (2002) who found that males tended to prefer action/adventure games (which fit into the second-order component “action”), sports, and simulation games (which fit into the second-order component “strategy”) whereas females tended to prefer puzzle, platform, and sports games. These findings are also similar to those of Lucas and Sherry (2004) who found that males preferred fighter, shooter, racing, and action/adventure games (which fit into the second-order component “action”), and sports, RPG, and strategy games (which fit into the second-order component “strategy”) more than did females and that females preferred card/dice games, classic board games, quiz, trivia, and puzzle games more than did males. The research examining the differences in video game preferences between the sexes appears to suggest that males prefer more action and violence oriented video games than do females whereas females tend to prefer more prosocial types of video

95 games (Consalvo & Treat, 2002; Lucas & Sherry, 2004; Terlecki et. al., 2010). The current study appears to support these findings. All three of the chi-square analyses were statistically significant, indicating that there were significant differences in genres corresponding to first, second, and third favorite video games between the sexes. Despite these significant findings, very few of the individual genre groups were of sufficient sample size to conduct further analyses to examine any simple effects. Several one-way chi-square analyses were conducted to examine the simple effects of the groups that contained sufficient numbers of participants but none of them were found to be statistically significant. Despite the fact that the sample sizes of most of the genre groups were too small to examine simple effects, an examination of the observed frequencies of males and females in each group suggests some potential differences that might be found with a larger sample size. Among all three chi-squares, zero male participants named a favorite game that was a simulation whereas a total of 31 games that were simulations were named by female participants. Also, among the three chi-squares only three games were named by males that belonged to the genre “music” whereas a total of 28 were named by females. Interestingly, among the three chi-squares, female participants named 50 favorite games that belonged to the genre “racing” whereas males only named 14. Another interesting trend that appears in the observed frequencies is that female participants named 71 favorite games that belonged to the genre “platform” whereas males only named eight. This is a particularly interesting trend because the most well known video game of all time [Super Mario Brothers (Nintendo R&D4, 1985)] is a

96 platform game. It seems reasonable to suggest that individuals who have relatively little experience playing video games might name Super Mario Brothers as a favorite game if asked to provide one. This would support the findings of previous studies that reported males playing video games more frequently and for longer periods of time than females (Royse, Lee, Undrahbuyan, Hopson, & Consalvo, 2007; Lucas & Sherry, 2004; Terlecki et. al., 2010). Future research using the current genre system to examine sex differences in video game preferences would be well advised to use larger sample sizes than the current study so that relationships such as those suggested here might be adequately analyzed. Relationships Not Addressed by the Study Hypotheses In addition to the relationships specifically addressed by the current study’s hypotheses, several others were investigated and a number were found to be significant. Results showed that older participants were more likely than younger participants to indicate that they would choose to spend more time playing video games if more free time was available to them. Also, male participants were more likely than female participants to indicate that they would choose to spend more time playing video games if more free time was available to them. Similar to many previous studies (Royse, Lee, Undrahbuyan, Hopson, & Consalvo, 2007; Lucas & Sherry, 2004), male participants indicated more frequent video game play and more hours spent playing video games in a typical week than did female participants. Female participants reported experiencing motion sickness more frequently than did male participants. This is possibly a reflection of the finding that females play video games less often and for less time than do males,

97 thereby having less habituation to motion sickness caused by exposure to virtual environments. Participants who indicated that they only play multiplayer video games reported experiencing motion sickness while playing video games more frequently than those who did not indicate that they only play multiplayer games. Two of the most popular forms of multiplayer video games are MMOs and first-person shooters. First-person shooters require players to navigate through three dimensional environments, while MMOs often necessitate frequent shifting in the player’s visual perspective. Both of these aspects of gameplay make MMOs and first-person shooters more likely than other types of video games to induce motion sickness in players. Participants who indicated that they only play multiplayer games were more likely to prefer action games as defined by the secondorder PCA of this study. This reflects the standing of first person shooters (which would fit into the action category) as one of the more common forms of multiplayer video game play. Of the numerous forms of multiplayer games, those played over Facebook or other similar social networking websites are some of the most popular. These games are typically more prosocial than other games and place a significant emphasis on player interdependence. Many Facebook games require players to make friends with other players and trade goods and services with them in order to succeed in the game. The current study found that female participants were significantly more likely than male participants to indicate that they only play video games on Facebook or another similar social networking website. This supports the findings of previous studies that showed

98 that females tend to prefer prosocial video games more than do males (Consalvo & Treat, 2002; Lucas & Sherry, 2003; Terlecki et. al., 2010). Reported frequency of video game play and reported number of hours spent playing video games in a typical week were found to be positively related to participants’ preferences for action and strategy games as defined by the second-order PCA of this study. In contrast to this, preference for cognitive processing games as defined by the second-order PCA was found to be negatively related to reported frequency of video game play. These findings are easily explained by the fact that many games that would fit into the categories of action and strategy games involve intricate and engrossing story lines that can motivate players to dedicate more time to game play. Additionally, cognitive processing games, in contrast to many action and strategy games, set goals for players that take a relatively short amount of time to achieve. In this way, cognitive processing games require much less time commitment. Concordantly, participants who indicated that they would choose to spend more time playing video games if more time was available to them were more likely to prefer action games than those who indicated that they would not spend more time playing video games. Limitations and Implications of this Research The current study yielded several findings that should be interpreted with caution. The three components yielded by the principal components analysis limited the comparison of the current study’s findings to many previous studies that used video game classification systems different from the current study’s classification system. Similar to Lucas and Sherry (2004), the current study suggests that classification of video games

99 can be reduced to three types. However, the three types yielded by the current study were highly intercorrelated. Because the current study attempted to identify distinct video game categories through measuring participants’ video game preferences, these three video game types are dependent upon subjective opinions. Therefore, the intercorrelation of the three video game types of this study may be a reflection of the tendency for individuals to prefer various types of video games simultaneously. Perhaps the three components yielded by the second-order PCA would be better described as “preference groupings.” The canonical correlation analysis yielded minimally interpretable results. However, the analyses of the relationships between personality traits and genres corresponding to first, second, and third favorite video games appeared to yield results that were more interpretable. The variance exhibited in the genres corresponding to participants’ favorite video games suggests that a significant amount of descriptive ability may be lost in reducing video games to three categories. There were numerous limitations in the current study that posed potential confounds. Of the various analyses performed in this study, those involving the genres corresponding to participants' first, second, and third favorite video games yielded the most promising results. However, much of this data was uninterpretable due to insufficient sample size. It would therefore be important for future studies to use larger samples when using the current genre system as a method of classifying video games. Future studies would also be well advised to sample from a population of identified video game players. Differing levels of experience with video games posed a potential

100 confound in the current study. Several participants in the current study stated that they rarely or never play video games. Participants who stated that they have never played video games were omitted from inclusion in data analysis; however, there were many other participants who reported playing very infrequently who were included in data analysis. The Gaming Preferences Questionnaire used in the current study was flawed in that it had an unbalanced number of items representing each of the conceptual video game genres. For example, there were only three items included in the inventory that were expected to correspond to the genre “music” whereas 16 items loaded on the component “RPG” in the principal components analysis. The current study showed that various aspects of video game play are shared by different types of video games. This makes creating such a balanced inventory a difficult undertaking. Another significant issue that must be considered in the current study is the issue of fatigue among participants. Although the majority of participants were able to complete their packets of materials in less than 30 minutes, in total, each participant's packet consisted of 19 pages of materials, including 307 multiple choice items and demographic information. It would not be unreasonable to suspect that a number of participants experienced significant fatigue, which may have affected their item responses. Despite its many flaws, the current study will lend support to any future efforts to measure empirically distinct video game categories. The analyses not involving the components yielded by the second-order PCA showed promising results that should be

101 explored further in future studies. The current study also demonstrated differences in video game preferences between the sexes beyond preferences for violent or prosocial games, which have been the primary focus of a majority of previous studies. This study was hindered by the inability to reliably and accurately measure preferences for distinct video game types. Although the current study attempted to create such a means of measuring video game preferences, it was not entirely successful in doing so. Given that video game play is an ever growing pastime, with a sizeable number of individuals already playing, it is likely that more research examining the relationships between personality and video game preferences will be performed in the future. In order for those studies to yield valid results, a more accurate and valid measure of video game preferences needs to be created. The current study provides support toward the creation of such a scale. However, much work remains to be done.

102 APPENDIX A Descriptions of the Entertainment Software Rating Board (ESRB) Ratings Rating

appropriate ages

possible content__________

Early Childhood (EC)

3 years and older

no material that could be considered inappropriate for children.

Everyone (E)

Everyone 10+ (E10+)

6 and older

10 and older

minimal cartoon, fantasy, or mild violence, Infrequent use of mild language. more cartoon, fantasy, or mild violence than (E) games, mild language, and/or minimally suggestive themes.

Teen (T)

13 and older

Violence, suggestive themes, crude humor, minimal blood, simulated gambling, and/or infrequent use of strong language.

Mature (M)

17 and older

Intense violence, blood and gore, sexual content, and/or strong language.

Adults only (AO)

18 and older

Prolonged scenes of intense violence and/or graphic sexual content and nudity.

Note: From “Game Ratings and Descriptor Guide” by the Entertainment Software Rating Board. Game ratings & descriptor guide. Retrieved February 13, 2012, from http://www.esrb.org/ratings/ratings_guide_print.jsp.htm.

103 APPENDIX B Demographic Sheet Please answer all of the following questions as honestly as you can. Doing so will allow your researchers to compile a more accurate representation of population characteristics using the data obtained from this sample. What is your: 1) Age________ 2) Sex (circle one): Male

Female

3) Ethnicity (please circle your choice): White/Caucasian

Asian or Pacific Islander

Black/African American Native American Other________________ 4) Major(s)___________________ 5) Year in college? (circle one) Freshman Sophomore

Junior

6) Marital status ___________________

Senior

Hispanic/Latino

104 APPENDIX C Gaming Preferences Questionnaire Using the given scale, please rate how much each of the following statements concerning video game preferences pertain to you. VERY LITTLE

LITTLE

NEUTRAL

MUCH

VERY MUCH

1

2

3

4

5

Place the number corresponding to how you feel each statement pertains to you in the empty space provided next to each item. Note that all items in this inventory are concerning video games. All items containing the word “games” are referring specifically to video games. ____1) Prefer games where opponents are engaged in small maps or arenas. ____2) Prefer games that try to scare the player. ____3) Prefer fast paced games. ____4) Enjoy setting up character stats (strength, intelligence, etc.). ____5) Enjoy moving an avatar (player controlled character, object, or vehicle) around really fast. ____6) Enjoy games that require players to move units around tactically. ____7) Prefer games where using guns is extremely important. ____8) Prefer games that have a tougher enemy at the end of the level. ____9) Prefer games that show hints about how to optimize play. ____10) Enjoy trying to get a high score. ____11) Prefer games that allow players to shoot. ____12) Prefer games where players can drive or fly a vehicle, craft, or robot. ____13) Prefer games with big and complex worlds.

105 ____14) Prefer games that only require solving puzzles. ____15) Prefer to control only one avatar (player controlled character, object, or vehicle) at a time. ____16) Prefer games where players mainly kick and punch enemies. ____17) Prefer games where players can conquer, explore, or commercialize. ____18) Enjoy fooling around the game world without any main reason or objective. ____19) Prefer games that allow players to decide evolution paths for their units. ____20) Prefer games that can be played online. ____21) Enjoy games that require players to make quick decisions. ____22) Prefer games where characters can learn abilities. ____23) Prefer games that are an intellectual challenge. ____24) Enjoy resolving just a few puzzles. ____25) Prefer games where players can make buildings and structures. ____26) Prefer games that emulate aspects of the real world. ____27) Prefer games that don't have any specific goal. ____28) Prefer games where characters use blade weapons. ____29) Enjoy resolving puzzles for their own sake. ____30) Prefer games where events happen once a player finishes his/her turn. ____31) Enjoy games that only sometimes require players to engage a character stronger than the average. ____32) Prefer games with intelligent life. ____33) Enjoy games that require players to make certain actions in coordination with music. ____34) Prefer sports games.

106

____35) Prefer games where some events continue by themselves. ____36) Prefer games that can be played with other people on the internet. ____37) Enjoy leveling a character. ____38) Prefer games where players can manage resources. ____39) Enjoy exploring and establishing relationships with other characters. ____40) Prefer games where players are given the chance to control several avatars (player controlled characters, objects, or vehicles) at a time. ____41) Prefer games that require players to resolve puzzles frequently. ____42) Prefer games with a story that unfolds while you play them. ____43) Enjoy being challenged with hand-eye coordination tasks. ____44) Prefer games where characters' stats have a key role to hit and resist while fighting. ____45) Enjoy games that require players to keep in time with a beat. ____46) Enjoy controlling multiple units. ____47) Prefer games in which engaging in combat is not that relevant. ____48) Prefer games that are carefully balanced by setting initial attributes comparably equal to all players. ____49) Enjoy completing quests. ____50) Enjoy doing combo moves for higher damage. ____51) Enjoy games in which having good aiming skill is a must. ____52) Prefer games where music and rhythm are an important part of game play.

107 APPENDIX D Gaming Patterns Questionnaire Please answer the following questions as they pertain to your video game playing patterns. 1) How often do you typically play video games (circle one)? a. every day b. almost every day c. several times a week d. several times a month e. several times a year f. rarely g. I have never played video games 2) How many hours a week on average would you say you spend playing video games? _________Hours

108 3) How old were you when you played your first video game (circle one)? a. 5 years old or younger b. between 6 and 11 years old c. between 12 and 18 years old d. between 19 and 35 years old e. older than 35 years old f. N/A (I have never played video games) 4) How often do you experience motion sickness while playing video games (circle one)? a. never b. rarely c. sometimes d. often e. always f. N/A (I have never played video games)

109 5) Please list up to 3 platforms you own and have played video games on in the last 12 months,ordering them from most used (1st) to least used (3rd). For example: PlayStation 3, Xbox 360, Nintedo DS, iPad, and PC (personal computer). 1St _____________________ 2nd _____________________ 3rd_____________________ 6) From the following options, please place a check mark next to your most preferred way of playing video games. ___Single player alone. ___Single player with other people (passing pads, hot seat) or helping out. ___Competitive multiplayer mode with someone in the same room. ___Cooperative multiplayer mode with someone in the same room. ___Competitive multiplayer mode on the Internet. ___Cooperative multiplayer mode on the Internet. 7) Please list your top 3 favorite video games ever. 1__________________________________ 2__________________________________ 3__________________________________ For the following 3 items, please circle “TRUE” if you feel the statement applies to you and “FALSE” if you feel it does not apply to you. 8) If I had more free time I would choose to spend more time playing video games. TRUE

FALSE

9) I only play video games with other people in the same room or over the Internet. TRUE

FALSE

10) I only play video games on Facebook or another similar social website. TRUE

FALSE

110 APPENDIX E Consent Form I hereby agree to participate in research which will be conducted by Joe Borders, a graduate student in psychology. In this research I will be given a packet of materials including some demographic questions, some questions about my video game play patterns and history, two inventories measuring personality characteristics, and another inventory measuring my preferences in video games. The research will take place in one of the research rooms on the third floor of Amador Hall and will require one half hour of my time. I understand that I will receive one half hour of credit toward satisfying the Psychology Department’s research participation requirement by participating in this study. I understand that this research may lead to further understanding of the relationship between personality and preferences in video game play. I understand that there is a possibility that some questions may make me feel uncomfortable and that I may discontinue my participation at any time without any penalty other than loss of research credit, and that the investigator may discontinue my participation at any time. This information was explained to me by Joe Borders. I understand that he will answer any questions I may have now or later about this research. Joe Borders can be reached at [email protected].

Signature: _____________________________

Date:____________________

111 APPENDIX F

Debriefing Purpose The purpose of this study was to investigate the relationships between several personality factors and video game preferences. A secondary purpose of this study was to provide a foundation for the construction of an empirically supported classification system for video games.

Hypotheses and Supporting Research Very little prior research has been done examining the relationship between personality and video game preferences (Hartmann & Klimmt, 2006). It is likely that this is because the video game classification system, currently in the form of genres, consists of categories that often overlap and are not consistently agreed upon (Apperley, 2006). The results of this study will be used, in part, to test the existing classification categories and to provide the basis for constructing a more empirically supported one. The only study known to have done anything similar to this found empirical support for the existence of only three individual genres which the researchers termed “Traditional”, “Physical Enactment”, and “Imagination” (Lucas & Sherry, 2004). Of the research performed to date on the relationship between personality and video game preferences, a study by Zammitto (2010) is particularly noteworthy. Zammitto found neuroticism to be a significant predictor of preference for shooters, action-no shooting, fighting, and sports games. Extraversion was found to be a significant predictor of preference for these same genres, as well as online play. Openness to experience was found to be significantly negatively related to preference for shooters, sports, and online play and positively related to preference for simulation, adventure, and puzzle games. Agreeableness was found to have a significant negative relationship to shooters, action-non shooters, fighting, sports, and online games while being positively related to preference for adventure games. Lastly, Zammitto's study found a significant positive relationship between conscientiousness and preference for action-non shooters and puzzle games and a negative relationship between conscientiousness and driving games. In her study, Ziammitto analyzed her data using conceptual genres rather than empirically supported video game categories. One of the main purposes of this study is to employ factor analysis in an attempt to provide such empirically supported categories. As such, it is hypothesized that patterns will be found in preferences for specific video game characteristics and that these patterns will be identifiable as distinct video game categories. Similar to the findings by Lucas and Sherry (2004), it is hypothesized that the actual number of distinct categories found will be far fewer than those used in Ziammitto's study and common video game nomenclature. It is further hypothesized that

112 the categories yielded by this analysis will be composites of the ones used in Ziammitto's study. For example, analysis may yield a category best described as “aggressive play” that is composed of games stereotypically thought of as belonging to the shooter, action, fighting, and sports genres. As such, it is hypothesized that the categories found by this study will have a similar relationship to personality as the component genres used in Ziammitto's study. In addition to the NEO Five-Factor Inventory used to assess personality in Ziammitto's study, the current study also employed the use of several scales from the California Psychological Inventory (CPI). Inasmuch as little research has been done on the relationship between personality and video game preferences in the past, it is unclear what relationships will be found between the factors of the CPI and video game preferences. However, it is hypothesized that dominance will be found to be positively related to preference for competitive games, and preference for violent video games will be negatively related to empathy and self-control.

Contact Information The results of this study will be available by May 24, 2012. If you would like further information about the study or have questions regarding the study, please contact Joe Borders at [email protected] at your convenience. Psychological Services If you have experienced any personal distress caused by the content or materials in this research and want to talk to someone, counseling services are available through the Student Health Center free of charge. Please contact Psychological Services at 278-6416 for assistance.

113 References Anderson, C. A., Carnagey, N. L., Flanagan, M., Benjamin, A. J., Eubanks, J., & Valentine, J., C. (2004). Violent video games: Specific effects of violent content on aggressive thoughts and behavior. Advances in Experimental Social Psychology, 36, 199-249. Apperly, T. H. (2006). Genre and game studies: Toward a critical approach to video game genres. Simulation and Gaming, 37(6), 6-23. ArenaNet. (2005). Guild Wars. [Computer software]. Bellavue, WA: NCsoft. Atari Inc. Pong. (1972). [Computer software]. Atari Inc. New York, NY: Atari Inc. Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice Hall. Baranowski, T., Buday, R., Thompson, D. I., & Baranowski, J. (2008). Playing for real: Video games and stories for health-related behavior change. American Journal of Preventive Medicine, 34(1), 74-82. Barlett, C. P., Anderson, C. A., & Swing, E. L. (2009). Video game effects-Confirmed, suspected, and speculative. A review of the evidence. Simulation and Gaming, 40(3), 377-403. Bartholow, B. D., Sestir, M. A., & Davis, E. B. (2005). Correlates and consequences of exposure to video game violence: Hostile personality, empathy, and aggressive behavior. Personality and Social Psychology Bulletin, 31(11), 1573-1586. BioWare. (2007) Mass Effect. [Xbox 360 software]. Redmond, WA: Microsoft.

114 Bioulac, S., Arfi, L., and Bouvard, M. P. (2008). Attention deficit/hyperactivity disorder and video games” a comparative study of hyperactive and control children. European Psychiatry, 23, 134-141. Blizzard Entertainment. (2004). World of Warcraft. [Computer software]. Irvine, CA: Blizzard Entertainment. Bushman, B. J., & Whitaker, J. L. (2010). Like a magnet: Catharsis beliefs attract angry people to violent video games, Psychological Science, 21(6), 790-792). Bruggemann, J. M., & Barry, R. J. (2002). Eysenck's P as a moderator of affective and electrodermal responses to violent and comic film. Personality and Individual Differences, 32, 1029-1048. Bryant, J., & Davies, J. (2006). Selective exposure to video games. In Vorderer, P. & Bryant, J. (Eds.) Playing video games. Motives, responses, and consequences (pp. 115-131). Mahwah, NJ: Lawrence Erlbaum Associates, Publishers. Capcom. (2009). Resident Evil 5. [XBOX 360 software]. San Mateo, CA: Capcom Capcom. (2009). Street Fighter IV. [PlayStation 3 software]. San Mateo, CA: Capcom. Cattell, R. B., Eber, H. W., & Tatsuoka, M. M. (1970). The handbook for the Sixteen Personality Factor Questionnaire. Champaign, IL: Institute for Personality and Ability Testing. Chan, P. A., & Rabinowitz, T. A cross-sectional analysis of video games and attention deficit hyperactivity disorder symptoms in adolescents. Annals of General Psychiatry.

115 Charlton, J. P., & Danforth, I. D. (in press). Distinguishing addiction and high engagement in the context of online game playing. Computers in Human Behavior, 1531-1548. Chory, R. M., & Goodboy, A. K. (2011). Is basic personality related to violent and non-violent video game play and preferences? CyberPsychology, Behavior and Social Networking, 14(4), 191-198. Chuang, Y-C. (2006). Massively multiplayer online role-playing game-induced seizures: A neglected health problem in internet addiction. CyberPsychology and Behavior, 9(4), 451-456. Church, T. A. (1994). Relating the Tellegen and Five-Factor models of personality structure. Journal of Personality and Social Psychology, 67(4), 898-909. Codemasters. (2011) Dirt 3. [XBOX 360 software]. Warks, UK: Codemasters. Cohen, J. (2001). Defining identification: A theoretical look at the identification of audiences with media characters. Mass communication and Society, 4(3), 245-264. Cole, H., & Griffiths, M. D. (2007). Social interactions in massively multiplayer online role-playing games. CyberPsychology and Behavior, 10(4), 575-583. Colwell, J. (2007). Needs met through computer game play among adolescents. Personality and Individual Differences, 43, 2072-2082. Consalvo, M., & Treat, R. (2002). Exploring gameplay: A survey of game players' preferences'. Unpublished manuscript.

116 Costa, P. T., Jr. & McCrae, R. R. (1985). The NEO Personality Inventory manual. Odessa, FL: Psychological Assessment Resources. Costa, P. T., Jr. & McCrae, R. R. (1992). Revised NEO Personality Inventory (NEO-PI-R) and NEO Five-Factor Inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources. Crystal Dynamics. (2012). Tomb Raider. [PlayStation 3 software]. Tokyo, Japan: Square Enix. Engler, B. (2003). Personality theories: An introduction. (C. Hartford & D. Richardson, Eds.)Boston, MA: Houghton Mifflin Company. Entertainment Software Association. (2011). 2011 sales, demographic, and usage data. Essential facts about the computer and video game industry. Retrieved July 4, 2011, from http://www.theesa.com/facts/pdfs/ESA_EF_2011.pdf. Entertainment Software Rating Board. (n.d.) Game ratings & descriptor guide. Retrieved February 13, 2012, from http://www.esrb.org/ratings/ratings_guide_print.jsp.htm. Ferguson, C. J. (2007). Evidence for publication bias in video game violence effects in literature: A meta-analytic review. Aggression and Violent Behavior, 12, 470-482. Ferguson, C. J., & Kilburn, J. (2010). Much ado about nothing: The misestimation and over interpretation of violent video game effects in eastern and western nations: Comment on Anderson et al. (2010). Psychological Bulletin, 136(2), 174-178. Firaxis. (2010). Sid Meier’s Civilization V [Computer software]. Novato, CA: 2k Games. Funk, J. B. (2005). Video games. Adolescent Medicine Clinics, 16(2), 395-411.

117 Funk, J. B., Baldacci, H. B., Paold, T., & Baumgardner, J. (2004). Violence exposure in real-life, video games, television, movies, and the internet: Is there desensitization? Journal of Adolescence, 27, 23-39 Gentile, D. A., Anderson, C. A., Yukawa, S., Ihori, N., Saleem, M., Ming, L. K., Shibuya, A., Liau, A. K., Khoo, A., Bushman, B. J., Huesmann, L. R., & Sakamoto, A. (2009). The effects of prosocial video games on prosocial behaviors: International evidence from correlational, longitudinal, and experimental studies. Personality and Social Psychology Bulletin, 35(6), 752-763. Goldberg, L. R. (1993). The structure of phenotypic personality traits. American Psychologist, 48, 26-34. Gough, H. G. (1957). Manual for the California Psychological Inventory. Mountain View, Ca: CPP, Inc. Gough, H. G. (1987). The California Psychological Inventory administrator's guide. Mountain View, Ca: CPP, Inc. Gough, H. (1996). CPI manual (3rd ed.). Palo Alto, Ca: Consulting Psychologists Press, Inc. Gough, G. H. (2002). Technical Brief for the CPI 260 Instrument. Mountain View, Ca: CPP, Inc. Gough, H., & Bradley, P. (1996) CPI manual. Third edition, California Psychological Inventory: Administrator's guide. Palo Alto, Ca : Consulting Psychologists Press.

118 Greitemeyer, T., & Osswald, S. (2010). Effects of prosocial video games on prosocial behavior. Journal of Personality and Social Psychology, 98(2), 211-221. Greitemeyer, T., Osswald, S., & Brauer, M. (2010). Playing prosocial video games increases empathy and decreases schadenfreude. Emotion, 10(6), 796-802. Griffiths, M. D., Davies, M. N., & Chappell, D. (2004). Online computer gaming: A comparison of adolescent and adult gamers. Journal of Adolescence, 27, 87-96 Groth-Marnat, G. (2009). Handbook of psychological assessment (5th ed.). Hoboken, N.J.: John Wiley & Sons. Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E. (2010). Multivariate data analysis (7th ed.). Upper Saddle River, N.J. : Prentice Hall. Hartmann, T., & Klimmt, C. (2006). The influence of personality factors on computer game choice. In Vorderer, P. & Bryant, J. (Eds.) Playing Video Games. Motives, Responses, and Consequences (pp. 115-131). Mahwah, NJ: Lawrence Erlbaum Associates, Publishers. Hathaway, S. R., & McKinley, J. C. (1943). The Minnesota Multiphasic Personality Inventory. Minneapolis: University of Minnesota Press. Hudson Soft. (2008). Mario Party 8. Redmond, WA: Nintendo. Id Software. (1993). Doom. [Computer software]. Mesquite, TX: Id Software. Id Software. (1994). Wolfenstein 3d. [Computer software]. Mesquite, TX: Id Software. Kestenbaum, G., & Weinstein, L. (1985). Personality, psychopathology and development issues in male adolescent video game use. Journal of the American Academy of Child Psychiatry, 24, 329-333.

119 Kirsh, S. J., & Mounts, J. R. (2007). Violent video game play impacts facial emotion recognition. Aggressive Behavior, 33, 353-358. Klimmt, C., Hefner, D., & Vorderer, P. (2009). The video game experience as “True” identification: A theory of enjoyable alterations of players' self-perception. Communication Theory, 19, 351-373. Konami TYO. (2001). Dance Dance Revolution. [PlayStation software]. Tokyo, Japan: Konami. Konijn, E. A., Bijvank, M. N., & Bushman, B. J. (2007). I wish I were a warrior: The role of wishful identification in the effects of violent video games on aggression in adolescent boys. Developmental Psychology, 43(4), 1038-1044. Laarni, J., Ravaja, N., Saari, T., & Hartmann, T. (2004). Personality-related differences in subjective presence. In M. Alcaniz & B. Rey (Eds.), Proceedings of the seventh annual international workshop presence 2004 (pp. 88–95). Valencia: Ed. UPV. Lachlan, K. A., & Maloney, E. K. (2008). Game player characteristics and interactive content: Exploring the role of personality and telepresence in video game violence. Communication Quarterly, 56(3), 284-302. Lucas, K, & Sherry, J. L. (2004). Sex differences in video game play: A communicationbased explanation. Communication Research, 31, 499-523. Markey, P. M., & Scherer, K. (2009). An examination of psychoticism and motion capture controls as moderators of the effects of violent video games. Computers in Human Behavior, 25, 407-411.

120 Markey, P. M., & Markey, C. N. (2010). Vulnerability to violent video games: A review and integration of personality research. Review of General Psychology, 14(2), 82-91. Maxis. (2000). The Sims. [Computer software]. Redwood City, CA: Electronic Arts. McCrae, R. R., & Costa, P. T. Jr. (2002). Comparison of EPI and psychoticism scales with measures of the five-factor model of personality. Personality and Individual Differences, 6(5), 587-597. McCrae, R. R., & John, O. P. (1992). An introduction to the five-factor model and its applications. Journal of Personality, 60(2), 175-215. Murray, G., Rawlings, D., Allen, N. B., & Trinder, J. (2003). NEO Five-Factor Inventory scores: Psychometric properties in a community sample. Measurement and Evaluation in Counseling and Development, 36, 140-149. Myers, I. B., & McCaulley, M. H. (1985). Manual: A guide to the development and use of the Myers-Briggs Type Indicator. Palo Alto, CA: Consulting Psychologists Press. Newtoy. (2010) Words with Friends. [Computer software]. Dallas, TX: Newtoy. Nintendo. (2003). The Legend of Zelda. The Wind Waker. [Nintendo Game Cube software] Redmond, WA: Nintendo. Nintendo. (2005). Nintendogs: Lab and Friends. [Nintendo DS software] Redmond, WA: Nintendo. Nintendo R&D4. (1985). Super Mario Brothers. [Nintendo Entertainments System software] Kyoto, Japan: Nintendo.

121 O'Connor, B. P. (2002). A quantitative review of the comprehensiveness of the five-factor model in relation to popular personality inventories. Assessment, 9(2), 188-203. Peters, C. S., & Malesky, A. Jr. (2008). Problematic usage among highly-engaged players of massively multiplayer online role playing games. CyberPsychology and Behavior, 11(4), 481-484. PopCap Games. (2005). Bookworm Deluxe. [Computer software] Seattle, WA: PopCap Games. Przybylski, A. K., Ryan, R. M., & Rigby, C. S. (2009). The motivating role of violence in video games. Personality and Social Psychology Bulletin, 35(2), 243-259. Pytlik Xillig, L.M., Hemenover, S. H., & Dienstbier, R. A. (2002). What do we assess when we assess a Big 5 trait? A content analysis of the affective, behavioral, and cognitive processes represented in Big 5 personality inventories. Personality and Social Psychology Bulletin, 28, 847-858. Reinecke, L. (2009). Games and recovery. The use of video and computer games to recuperate from stress and strain. Journal of Media Psychology, 21(3), 126-142. Roe, K., & Muijs, D. (1998). Children and computer games: A profile of the heavy user. European Journal of Communication, 13, 181-200. Rovio. 2009. Angry Birds.[Computer software]. Macclesfield, England: Chillingo. Royse, P., Lee, J., Undrahbuyan, B., Hopson, M., & Consalvo, M. (2007). Women and Games: Technologies of the gendered self. New Media and Society, 9(4), 555-576. Sacau, A., Laarni, J., & Hartmann, T. (2008). Influence of individual factors on presence. Computers in Human Behavior, 24, 2255-2273.

122 Sas, C. (2004). Individual differences in virtual environments. In M. Bubak, G. Dick van Albada, P. Sloot, & J. Dongarra (Eds.), Computational Science – ICCS 2004, Fourth International Conference, Proceedings, Part III. Lecture Notes in Computer Science (vol. 3038, pp. 1017–1024). Springer-Verlag. SCE Studios Japan. (2005). Shadow of the Colossus. [PlayStation 2 software]. Foster City, CA: Sony Computer Entertainment. Sensory Sweep. (2007). My Spanish Coach. [Nintendo DS software]. Larkspur, CA: Ubisoft. SCE Studios Santa Monica. (2010). God of War 3. [PlayStation 3 software]. Forester City, CA: Sony Computer Entertainment. Sharpe, J. P., & Desai, S. (2001). The revised NEO Personality Inventory and the MMPI-2 Psychopathology Five in the prediction of aggression. Personality and Individual Differences, 31, 505-518. Smith, B. P. (2006). The (computer) games people play. In Vorderer, P. & Bryant, J. (Eds.) Playing video games. Motives, responses, and consequences (pp. 43-56). Mahwah, NJ: Lawrence Erlbaum Associates, Publishers. Square. (2001). Final Fantasy X. [PlayStation 2 software]. Tokyo, Japan: Square. Swing, E. L., Gentile, D. A., Anderson, C. A., & Walsh, D. A. (2010). Television and video game exposure and the development of attention problems. Pediatrics, 126(2), 214-221.

123 Tellegen, A., & Atkinson, G. (1974). Openness to absorbing and self-altering experiences (“absorption”), a trait related to hypnotic susceptibility. Journal of Abnormal Psychology, 83, 268-277. Tellegen, A. & Waller, N.G. (1992) Exploring personality through test construction: Development of the Multi-dimensional Personality Questionnaire (MPQ). Unpublished manuscript. Department of Psychology, University of Minnesota. Teng, C. (2008). Personality differences between online game players and non players in a student sample. CyberPsychology and Behavior, 11(2), 232-234. Terlecki, M., Brown, J., Harner-Steciw, L., Irvin-Hannum, J., Marchetto-Ryan, N., Ruhl, L., & Wiggins, J. (2010). Sex differences and similarities in video game experience, preferences, and self-efficacy: Implications for the gaming industry. Current Psychology, 30, 22-33. Tiburon. (2011) Maden NFL 12. [PlayStation 3 software] Redwood City, CA: Electronic Arts. Tolchinsky, A., & Jefferson, S. D. (2011). Problematic video game play in a college sample and its relationship to time management skills and attentiondeficit/hyperactivity disorder symptomology. Cyberpsychology, Behavior, and Social Networking, 14(9), 489-496. Tyson, P., & Tyson R. L. (1990). Psychoanalytic theories of development: An integration. Binghamton, NY: Vail-Ballou Press.

124 Von Salisch, M., Oppl, C., & Kristen, A. (2006). What attracts children? In Vorderer, P. & Bryant, J. (Eds.) Playing video games. Motives, responses, and consequences (pp. 115-131). Mahwah, NJ: Lawrence Erlbaum Associates, Publishers. Vorderer, P., Hartmann, T., & Klimmt, C. (2003). Explaining the enjoyment of playing video games: The role of competition. In D. Marinelli (ED.), Proceedings of the 2nd international Conference on Entertainment Computing (ICEC 2003), Pittsburgh (pp. 1-8). New York: ACM. Weiss, M. D., Baer, S., Allan, B., Saran, K., Schibuk, H. (2011). The screens culture: Impact on ADHD. Attention Deficit and Hyperactivity Disorders, 3(4), 327-334. Williams, R. B., & Clippinger, C. A. (2002). Aggression, competition and computer games: Computer and human components. Computers in Human Behavior, 18, 495-506. Williams Entertainment Inc. (1996) Ms. Pac-Man. [Computer software] Chicago, IL: Midway Games. Wolf, M. J. (2001) Genre and the video game. The medium of the video game. Austin: University of Texas Press. Wolf, M. J. (2008) The video game explosion. A history from Pong to PlayStation® and beyond. Westport, CT: Greenwood Press. Zammitto, V. L. (2010). Gamers’ personality and their gaming preferences. (Master's thesis).