AARHUS UNIVERSITY, BUSINESS AND SOCIAL SCIENCES

Joint decision-making roles between spouses in Latvia when buying a family car

VIESTURS UZULEŅŠ VU84938 CLASS U35 BSC IN BUSINESS SUPERVISOR: LAURA KAZBARE

APRIL 2012

Abstract The family has traditionally been considered as an important decision-making unit regarding large quantity of products and services which are purchased by households on a daily basis. One of the most important one-time family purchases is a family car. In the first eight months of year 2011 passenger car sales in Latvia increased by 80% and because family in one of the main consumers of this product it is important for marketers to understand this economic unit. In this research paper author attempts to examine family decision-making in general and, additionally, offers a practical perspective at joint decision-making roles between Latvian spouses. First part of the paper sets up a theoretical framework of family decision-making process and factors that affect it. This is done by studies of theory and previous researches. Second part takes an empirical approach and researches how decision-making roles are divided in Latvian families when they buy a car. A consumer survey is used in order to gather data. Theoretical framework discusses how the individual and family factors determine if purchase decision is joint or dominated by one of the spouses. These factors affect the decision-making process which normally contains three stages – problem recognition, information search and final decision. Empirical part uses data gathered from 69 Latvian families. Tests performed on the basis of this data show that car purchase decision process in Latvia is male dominated, especially in the stage of information search which is autonomous by him. Eventually, this research concludes that Latvia is no exception when compared to other countries and car purchase decision in family is male dominated with some aspects being joint. However marketing strategies should be considered carefully because spousal roles change between stages and sub-decisions.

Table of Contents Abstract List of figures List of tables 1 Introduction .................................................................................................................................... 1 1.1 Introduction to the subject ....................................................................................................... 1 1.2 Problem statement and research questions ............................................................................. 2 1.3 Delimitations............................................................................................................................ 3 1.4 Structure of the thesis .............................................................................................................. 3 2 Theoretical framework.................................................................................................................... 5 2.1 Factors of joint decision-making ............................................................................................... 8 2.1.1 Individual factors in joint decision-making ........................................................................ 8 2.1.1.1

Individual factors when entering decision-making ........................................................ 8

2.1.1.2

Individual attitudes towards products ......................................................................... 12

2.1.2

Family factors of joint decision-making ........................................................................... 13

2.1.2.1

Family roles ................................................................................................................ 13

2.1.2.2

Family types ............................................................................................................... 14

2.1.2.3

Family life cycle .......................................................................................................... 18

2.1.3

Other factors affecting family decision-making ............................................................... 19

2.2 Decision-making process ........................................................................................................ 21 2.2.1 Roles, stages and decisions in purchasing process........................................................... 22 2.2.2

Disagreement and handling of it ..................................................................................... 24

2.3 Decision-making regarding cars .............................................................................................. 26 2.3.1 Application of the previously reviewed theories to car decision making .......................... 26 2.3.2

Previous research on family car purchase ....................................................................... 27

2.3.3

Attributes of automobile ................................................................................................ 32

2.4 Sub-conclusion ....................................................................................................................... 32 3 Empirical research methodology and analysis .............................................................................. 34 3.1 Research methodology........................................................................................................... 34 3.1.1 Questionnaire design...................................................................................................... 34 3.1.2

Answering scales ............................................................................................................ 37

3.1.3

Translation ..................................................................................................................... 37

3.1.4

Sampling technique ........................................................................................................ 37

3.2 Data analysis .......................................................................................................................... 38 3.2.1 Sample characteristics .................................................................................................... 39

3.2.2 3.2.2.1

Assessment of 3 decision-making role determination methods ................................... 42

3.2.2.2

Roles in 3 decision-making stages ............................................................................... 42

3.2.2.3

Roles in 4 sub-decisions .............................................................................................. 43

3.2.2.4

Roles in 8 sub-decisions .............................................................................................. 45

3.2.3

Importance of car’s attributes ........................................................................................ 45

3.2.4

Factors that affect the extent of joint decision-making ................................................... 47

3.2.4.1

Car related variables ................................................................................................... 48

3.2.4.2

Family related variables .............................................................................................. 49

3.2.5 4 5

Decision-making roles from perspective of the three methods ....................................... 40

Other data relevant for marketers .................................................................................. 49

Results and discussion .................................................................................................................. 51 Final assessment ........................................................................................................................... 57 5.1 Conclusion ............................................................................................................................. 57 5.2 Practical implications for marketers in Latvia.......................................................................... 58 5.3 Limitations ............................................................................................................................. 58 5.4 Suggestions for future research .............................................................................................. 59 References ............................................................................................................................................ 60

List of figures Figure 1: Thesis structure ........................................................................................................................ 4 Figure 2: Structure of the theoretical framework ..................................................................................... 6 Figure 3: Family type differences ........................................................................................................... 18 Figure 4: Structure of the questionnaire ................................................................................................ 35 Figure 5: Importance of car’s characteristics to men and women........................................................... 46 Figure 6: Automobiles purchased by Latvian families by class ................................................................ 50 Figure 7: Decision-making patterns by income (frequencies) ................................................................. 55

List of tables Table 1: Classification of family types .................................................................................................... 15 Table 2: Decision-making factors and their effect on decision-making type............................................ 21 Table 3: Car purchase in different countries [1] ..................................................................................... 29 Table 4: Car purchase in different countries [2] ..................................................................................... 30 Table 5: Car purchase in different countries [3] ..................................................................................... 31 Table 6: Roles in automobile purchase decision-making process ............................................................ 41 Table 7: Decision power intervals and corresponding category .............................................................. 41 Table 8: Test of 3 methods .................................................................................................................... 42 Table 9: Comparison of 3 stages ............................................................................................................ 43 Table 10: Comparison of problem recognition and mean result ............................................................. 43 Table 11: Roles in 3 decision-making stages ........................................................................................... 43 Table 12: Roles in 4 sub-decisions .......................................................................................................... 44 Table 13: Comparison of 4 sub-decisions and mean result ..................................................................... 44 Table 14: Roles in 8 sub-decisions .......................................................................................................... 45 Table 15: Comparison of attributes ........................................................................................................ 46 Table 16: Decision-making roles of particular attributes and most important attributes ........................ 47 Table 17: Effect of car related variables on extent of joint decision-making in family ............................. 48 Table 18: Effect of family related variables on extent of joint decision-making in family ........................ 49

1

1 Introduction The introduction chapter will introduce the reader with the subject and why it is a relevant topic. It also contains problem statement with research questions that this paper aims to answer in consecutive sections. It further contains the problem statement and consecutively the research questions. Some boundaries are acknowledged in delimitations sub-sections and lastly the structure of the thesis is presented.

1.1 Introduction to the subject In the age of product diversity, with many companies offering even more brands, it has become essential for companies to position their products, choose market segments and look at market niches more carefully than ever before. Companies have to make purchase decisionmaking an effortless task for consumer. But a question might arise – what is the target segment when the choice of the product is a joint decision? It is easy to market the product when its target segment is clear, e.g. sports cars for wealthy men, lipstick for women and toys for kids (although this is ambiguous). But when decisions are made in a group of two or more persons, it becomes less clear - who makes the buying decision and therefore is the target of marketing effort. The family has traditionally been considered as an important decision-making unit with respect to the large quantity of products and services which are purchased by households on a daily basis (Martinez and Polo, 1999). Marketers could try to sell their products to the spouses as a whole, therefore combining the marketing strategies they would use for husband and wife individually and come up with a strategy that more or less approaches both gender groups. But extracting the middle way strategy or settling for “more or less” strategy is not optimal, if there is a way to find out the patterns of decision-making between spouses regarding various products. Knowing which product attributes each gender values more or which of two has more decision power in particular decision-making stages is a big step further in producing a successful and highly targeted marketing campaign. Authors’ attention was captured by data from European Automobiles Manufacturers Association (ACEA) – Latvia’s passenger car sales in first 8 months of 2011 have increased by impressive 80.15% (ACEA, 2011) comparing to the same period a year before. This means that the market of automobiles is recovering fast after steep decline during recent global recession and therefore in order to draw consumers’ attention (and money) to particular car brand precise advertising must be done. As no prior research has been carried out regarding joint decision-making in Latvia and especially regarding car purchase, decision-making roles of spouses in Latvia, when buying a family car is a so far untouched research topic. Decision to buy a new car is mostly made when

2

some financial stability is acquired and this state often overlaps with the time period in one’s life when he or she is also married. As Latvia’s disposable income per capita is still lagging behind that of European Union (EU) average, people in strong majority of cases own only one car per family. For these reasons married couples can be considered to be one of the major consumer segments in passenger car market and a subject worth researching. Even though the majority of research reports car buying as husband dominated decision in many countries (Harcar et al., 2005; Mohan, 1995; Davis, 1970), Harcar et al. (2005) indicates that in USA it is a joint decision. With the socialistic history of Latvia, where husband and wife shared the decision-making (Zickel, 1989), and Scandinavian gender equality influence nowadays that comes with business, geographical and political proximity, there is a reason to question if a car buying decision is male dominated in Latvia.

1.2 Problem statement and research questions The main aim of this research paper is to find out how joint decision-making process between spouses happens when the purchase object is an automobile. Knowing how joint decision-making process happens and which members of the family have more influence when choosing between particular automobile brands can help companies to aim marketing efforts of a company more directly towards the real decision maker. This way the product awareness will be increased in the mind of the actual buyer instead of scattering marketing effort, and eventually money could be saved by avoiding an attempt to hit multiple birds with one stone. This thesis project would be useful for marketing specialists in media and creative agencies and other companies to construct highly targeted car advertising campaigns. The question about the “who” in the typical household makes decisions about the economic affairs and purchases of the family is constantly asked by advertisers who want to select the most effective media (Wolgast, 1958). This paper intends to look at joint decision-making between spouses and how it might translate into behavioral patterns when buying an automobile. On the practical side this paper aims to provide information to decide on advertising techniques, media and advertisement content more precisely. In the second part this paper will look at the theories and previously publicized articles in order to understand decision-making process between spouses and aspects that influence it. Among things reviewed in this part will be decision-making stages, family roles, gender roles, family types and personal dispositions. In the third part author will use an empirical approach to car purchase decision-making and, by using online questionnaire, research decision-making dynamics and roles of spouses in Latvian market. Five research questions have been defined: Question 1: What are the factors that shape the decision-making process and how these factors affect the extent of joint decision-making? Question 2: What is the joint decision-making process and how the roles are divided in it?

3

Question 3: What are decision-making roles between Latvian spouses when they purchase a family car? Question 4: Which spouse (if any) and how should be targeted by marketers in Latvia? Question 5: Which are the determinants that shift car purchase decision-making towards more autonomy? First two questions will be answered from a theoretical perspective in the theoretical framework chapter. Last three questions are directly related to Latvian families and will be answered in the empirical research chapter.

1.3 Delimitations Most of the resources in theoretical framework concern USA and are subjects that require cautiousness because social and family structures cannot be generalized to other countries to full extent. However nowadays it is acknowledged that “western society” has more common than different. Research regarding families in Europe and North America treats them as experiencing the same phenomena (Cliquet, 2003; Goldthorpe, 1987). By being a member of EU and NATO, Latvia can also be considered as Western society. Before we continue with the main body of this paper, it is crucial to explain what the main object is. In our case it is spouses that alone or with kids form a family. Couple is considered to have made a family unit only if it has left their parents and lives independently (Sheth, 1974). Throughout this paper terms husband and wife will be used disregarding if the couple is bound by legal marriage. In the empirical research part it is important to include couples who have formed a family but are not in a registered marriage, as the ratio of families with kids in Latvia where spouses are not married has increased multiple times in recent decades (Eurostat, 2012; ). There is not one definition for “family” in Latvia but Latvian Constitutional Court has used European Court’s of Human Rights interpretation of “family” which includes also de facto “families” with unmarried spouses (Bitāns et al., 2008). During the last decade the amount of studies on joint decision-making in families has decreased, therefore it is possible that some findings in the theoretical framework part do not have as strong ground as they should. Because the main object of research is spouses, the term “family” in this paper includes only husband and wife and excludes children. Therefore, concepts like “joint decision-making”, “family decision-making” and “decision-making between spouses” will be used interchangeably.

1.4 Structure of the thesis In Figure 1 you can see the structure of the thesis which starts with the current - chapter 1. Chapter 2 consists of theoretical framework. It discusses matters from a perspective of what areas one could look into to understand how family decision-making is shaped, if degree of

4

joint decision-making in family regarding car purchase is a result of underlying factors and processes. The empirical part consists of two parts. In chapter 3 consumer survey will be used to focus particularly on car purchase and answer research questions regarding Latvian automobile market. Results from the data gathered about Latvian families will be presented and discussed in chapter 4. The end part of this paper will present the conclusion, discuss the practical use of this research by marketers, give thoughts on future research and draw the lines of this paper in the shape of limitations. Chapter 1 Introduction

Introduction

Problem statement and research questions

Delimitations

Thesis structure

Figure 1: Thesis structure

Chapter 2 Theoretical background

Literature review

Chapter 3 Empirical research 1

Chapter 4 Empirical research 2

Chapter 5 Closing assessment

Research design

Results

Conclusion

Data analysis

Discussion

Practival implications

Future research

Limitations

5

2 Theoretical framework In this chapter we will look at previous research and theories that might provide some background knowledge about decision-making process in families, factors that influence this process and roles of husband and wife in it. Roles of purchasing process are the central element of this chapter. Theoretical framework will look at reasons that determine the extent of joint purchase decision-making in family. Further theory will be reviewed that explains how particular members of family (husband and wife) act in joint decision-making and what is their influence on purchase decision. All this will be done with an automobile, as a product of interest, in mind. Finally, theoretical background regarding cars will be studied in order to learn what discoveries other researchers have found on the purchasing roles between spouses. Theoretical framework section consists of 3 main parts. First two are more generic and deal with decision-making aspects in general. 2.1 looks at joint decision-making factors that influence the process and 2.2 discusses joint decision-making process itself. Part 2.3 then takes a turn to apply previously discussed things to car purchase and discovers additional aspects regarding family car buying. Figure 2 shows the structure of theoretical background section.

6

2 Joint decision-making

2.1.2 Family factors

2.1.1 Individual factors 2.1.1.1 Individual factors when entering decisionmaking

2.1.1.2 Individual attitudes towards products

2.1.2.1 Family roles

2.3 Decisionmaking regarding cars

2.2 Joint decision-making process

2.1 Factors of joint decision-making

2.1.2.2 Family types

Figure 2: Structure of the theoretical framework

2.1.3 Other factors

2.1.2.3 Family life cycle

2.2.1 Roles, stages and decisions in purchasing process

2.2.2 Disagreement and handling of it

2.3.1 Applications of theories to car decisionmaking

2.3.2 Previous research on family car purchase

2.4 Subconclusion

2.3.3 Attributes of an automobile

7

“Family decision-making is the process by which decisions that directly or indirectly involve two or more family members are made” (Harcar et al. 2005). These decisions are considered to be of two types: autonomous by a single member or joint by both spouses. Of course, there are many transitional states where decisions can be slightly more dominated by one of the spouses. Large numbers of buying decisions are likely to be joint if their consumption, as often is the case with a family car, is also joint (Sheth, 1974). According to Sheth (1974) family’s social class, role orientation and life cycle as well as the relative importance of purchase, perceived risk and time pressure impinging upon a buying decision determine whether a buying decision will be joint or autonomous. The theoretical background part will mainly use determinants pointed out by Sheth (1974) together with internal power distribution theories described by Yavas et al. (1994) to build a theoretical framework. Yavas et al. (1994) identifies four theories that have been advanced to explain how tasks and power are allocated within families: cultural role expectations, comparative resources, the least interested partner, and the relative investment theories. The main principle of the cultural role expectations theory is that power (defined as the potential ability of one partner to influence the other’s behavior) is prescribed into the roles of husband and wife according to existing social norms. Consequently, a spouses power is external to the family and resides in the position rather than in the person (Davis, 1976). Societal characteristics and type of society (patriarchal or equalitarian) have a substantial influence on the type of the family and further on decision-making manner in family. Even though this is of high importance, the thesis will not provide a thorough investigation at how society influences joint decision-making because religion and societal development is an extensive subject itself. In addition cultural roles expectation theory contradicts internal theories that in turn supplement each other. However, topic will be slightly introduced in the section about family types. One of the internal theories employed in this paper is the comparative resources theory (Blood and Wolfe, 1960) that states that the allocation of tasks and authority within a marriage is determined totally within the family and not externally. Comparative resources theory along with the least interested partner theory and relative investment theory will appear in section 2.1.1 where individual factors will be examined. The least interested partner theory focuses on the resources contributed by each spouse in terms of their relative value outside the marriage (Heer, 1963). The relative investment theory (Davis, 1976) is rooted in the relative involvement and interest that each spouse has in a particular choice decision. A comparison of these four theories reveals that the emphasis accorded to the external versus internal nature of the sources of power is the primary factor that differentiates them (Yavas et al., 1994).

8

2.1 Factors of joint decision-making This section will seek to find the factors that determine the extent of joint decision-making in a family and how the increase or decrease of the presence of this factor in the family can result in more autonomous or more joint decision-making. It will be assumed that general decision-making patterns can be applied to purchase decision-making as it is probably one of the most important aspect of family’s everyday life, involving communication and need for decisions to be made. Theory of family decision-making can be seen as consisting of two parts. First one is the individual factor in the shape of predispositions and attitudes brought in the family by husband and wife. It will be explored in section 2.1.1. Second part is family factors and it’s characteristics as a whole and will be reviewed in section 2.1.2. Together with other factors reviewed in section 2.1.3 they shape join decision-making factors that eventually shape purchase decision-making process, roles in it and how family behaves when they want to purchase a product.

2.1.1 Individual factors in joint decision-making Each family member enters joint decision-making with his/her own attitudes, predispositions and “subjective reality”. According to Sheth (1974) there are three main factors that determine the cognitive world of husband/wife. Those are firstly exogenous factors like personality, lifestyle, member’s perception of social class and role orientation in family. Second thing affecting member’s world view is availability of information from number of sources like media, word-of-mouth and others. Third determinant is family itself and how family members over lifetime affect each other. Different “subjective realities” where each spouse perceives various situations and sets of facts differently, according to their own needs, beliefs and values can lead to disagreements between spouses (Safilios-Rothchild, 1969). Most important of individual factors will be reviewed in section 2.1.1.1. Additional individual aspect is attitudes towards products. Affected by his or her cognitive world, different utilities and values are assigned to products where the same product can be viewed differently and satisfy different needs of each spouse. Section 2.1.1.2 will contain information about this “product aspect” of individual attitudes. 2.1.1.1 Individual factors when entering decision-making This section will discuss individual factors that affect spouse’s behavior in joint decisionmaking situation. It can be seen as spouse’s baggage that he or she brings along to the marriage. Attitude towards gender roles Gender role orientation is most probably one of the most important individual factors that shape the understanding of how interaction between men and women should look like. Gender roles are those behaviors and attitudes prescribed and assigned to males and females by the

9

broader culture solely on the basis of gender (Bartley et al., 2005). In the case of assigned gender roles, spouse's authority is based on the belief that he or she should make a decision or carry out a task irrespective of the actual skills or interest that may be present. The source of a spouse's power is thus external to the family: Power resides in the position rather than in the person (Davis, 1970). Even though in recent decades there has been a shift to a more equalitarian family model in the Western world, decision-making in contemporary marriages has been divided along traditional gender lines, with wives making decisions concerning day-to-day details of family life and husbands making the major decisions, such as those concerning career choices and resource allocation (Steil and Weltman, 1991, as quoted in Bartley et al., 2005). In general, if husband’s personal world view is traditional and he supports strict division of gender roles, then he will bring this attitude to the family and implement it there. Because couples with similar values and world views make families and get married, it is unlikely that despite his/her traditional attitudes the family will turn out to be equalitarian. Information The next individual factor that affects spouse’s world view is information availability from the environment. Firstly information is provided by sellers through advertisement, in-store displays and brochures. Consumers draw information from their previous experience as well as by assessing the price of product and type of store where the product is sold. Other sources of information include friends, family, salespeople as well as specialty magazines and product rating web-sites. In general information sources can be divided into two categories – external environmental and information available in consumer’s memory. This distinction between two major information source categories also affects how consumers make decisions (Bettman, 1991). Information that is at the disposal of a spouse could determine his involvement in the information search stage of decision-making where, depending on product, one spouse might be the provider of information. Knowing the information search patterns and sources of final decision-maker is of high importance to marketers. Marketers are interested to reach the spouse that searches for information. Additionally the information sources of final decisionmaker are important when marketers decide on media placement. The way how the individual gathers and perceives information and sources she relies on can affect the information search stage of decision-making which will be discussed later in section 1.2.1. Education There is not common agreement between researchers about whether education of spouses affects level of joint decision-making or not. Duxbury et al. (2007) and Davis (1976) write that spouses with lower education levels that have blue collar jobs possess more traditional attitudes towards family roles where less joint decision-making takes place. But Jenkins (1980) reports that there are no significant differences between the degree of decisionmaking power of educated and uneducated husbands, or high occupational status and low

10

occupational status husbands. Lu (1952) approaches this issue from different perspective and claims that aside from consanguine family factors equalitarianism is also affected by educational difference of spouses. It can be predicted that equal education leads to more equal roles but, what is interesting, more educated wife will not have dominance contrary of more educated husband who will. Slama and Taschian (1985) write that education of a spouse is positively related to search behavior in individuals: With higher levels of education producing greater propensity to search for information and alternatives before making final purchases. Higher levels of education can also affect the way a person perceives a product and its attributes by looking at quality more carefully, for example. Family Personality and the way a spouse sees family roles and exerts dominance can come from both consanguine and conjugal families. Family roles are affected by multiple factors from spouses’ childhood families. These factors are conflict and attachment with parents, discipline at home, birth order in family and reaction to authority. Cases where there was no conflict with parents or attachment to them generates more equal roles in conjugal family whereas conflicts with parents move dominance towards husband or wife (Lu, 1952). Multiple authors suggest that a person’s power to make decisions comes from the ability to provide for the fulfillment of his/her marriage partner’s needs. Therefore, the more a husband fulfills his duty to provide for the family, the more the wife will allow the husband to define the norms of decision-making (Davis, 1976; Scanzoni, 1972). Interest and investment According to relative investment theory, the importance associated with various decision outcomes determines how members within the family interact in a decision situation. If decision outcomes are considerably more important to one spouse relative to the other, then that spouse attempts to dominate. For a different decision, however, the other spouse may be dominant. Investment defines the motivation of a family member to exert influence (Davis, 1976) over a product purchase decision that he finds important and interesting to him/herself. Therefore he or she is willing to invest more effort to get a positive result. Personal interest of each spouse in a particular product affects how involved he/she is in decision-making. The greater is the interest of one of the spouses in a product, the greater is the influence that he/she exercises (Martinez and Polo, 1999). Coleman (1966) proposed that when individuals face a sequence of decisions, it is possible for them to give up control over those of little interest for more control over those of greater interest. For example, the husband can have little interest in house furnishings and even though it is a financially big purchase he lets the wife to decide autonomously. But in the next purchase occasion wife might pay back and let the

11

husband have his own way with the new car purchase because he has significantly more interest here. Comparative resources The comparative resources theory proposes that the spouse that contributes more resources (e.g., income, competence, personal attractiveness, better performance as a homemaker) to the household unit is more powerful. Comparative resources define the potential to exert influence in product purchase decisionmaking. Research of Bartley et al. (2005) shows that wives in dual-earner couples perceive themselves as exerting somewhat greater influence in the day-to-day interactions than are husbands. This mostly can be because of the role assigned to female as the head of household who runs daily home life. Decision-making influence has been greater for husbands in our society. However it has been softened by more equal influence in decision-making among dualearner couples (Bartley et al., 2005), where the spouse who contributes more to family budget has more decision power (Blood and Wolfe, 1960; Rodman, 1972). In USA men lost employment at higher rates than women during most recent recession (Bardenheier et al., 2011) and as a result of this, women’s power in family might be changing right now. Even though greater contribution by female to a family budget means more equalitarian power distribution, variations in occupational status and income according to Jaris (1999) appear to have relatively little impact on marital power. Many of status-reversal (wife earns more than husband) wives back away from whatever power they might derive from their income and status. There is a material component to equality in marriage, but it is neither income nor status that translates into power. Also Commuri and Gentry (2005) point out - spouses tend to adjust to situation when wife earns more than husband by using multiple financial pools instead of one family budget. This way they keep equity in relationship and wife does not gain more power. In general, if wife contributes significantly to family income it does not lead to her dominance in family but rather more equalitarian relationship. This has been acknowledged also by Martinez and Polo (1999). Least interested partner This theory essentially expands upon the comparative resources theory and introduces marriage alternatives as yet another source of power. The spouse, who can most easily find another spouse, as desirable as the present one, has another source of power by having potentially acceptable alternatives (Yavas et al., 1994). The least-interested-partner theory claims that it is not the value of the resources contributed by each partner to family budget, but the value of these resources outside the marriage. For example, the greater the difference between the value to the wife of the resources she might earn outside the marriage, the greater the influence of the husband in family decision-making in case where wife earns more in marriage. The least-interested partner theory has been more powerful and found more support as it explains as much variation in family role structures as the relative contributions

12

theory and, in addition, it can accommodate the changing patterns of family member interaction that occur over the life cycle (Jenkins, 1980). All these personal factors determine if the spouse will exert power in family or be more submissive. If the influential spouse and his or her influence strategies can be identified in a decision, marketers may be able to target the decision-maker in the subsequent purchases (Su et al., 2003). 2.1.1.2 Individual attitudes towards products Every family member has attitudes or predispositions towards products. These predispositions are presumed to be different in many instances because of difference in motives and perceptions of individual members. For example, in buying an automobile, the husband may prefer a medium-priced full size sedan but the wife may prefer low-priced full size station wagon. Within each of these types, the members may perceive the benefit of specific cars differently (Sheth, 1974). Beliefs that affect evaluation of product alternatives may vary from one member to the other not only because their buying motives are different but also they perceive the same alternative differently in its extent of being a perceived instrument to satisfy those motives (Sheth, 1974). From an economists’ perspective, where consumers are perfectly rational beings, consumers obtain information on the alternatives, make trade-offs that allow them to compute utilities for every alternative, and select the alternative that maximizes utility (Bettman at al., 1991). Even though the assumption about perfect rationality does not sustain in real life, utilities theory can be applied to situations where spouses are assessing attributes of the product. Each of the spouses can perceive the same product differently because they assign certain utility to the attributes and qualities of it. When these attributes are summed up, husband and wife can see different value in the good and in the end prefer product different from one that other spouse has chosen (Menasco and Curry, 1989). Preferences represent a more exact specification of priorities within a product set and are likely to be important later in the decision process (Buss and Schaninger, 1983). Also, Ward (2006) found that gender preference intensities play a significant role in the final decisions made by spouses in joint product decisions but it depended also on product category. Preferences arise from assigning weight to product characteristics. In their study Menasco and Curry (1989) used a function where utility for husband and wife was determined by the weight each of them assigns to particular product characteristics. Couples tend to compromise, seeking balance and equity in outcomes so that harmony in family is kept. The forces that stimulate equitable choices are grounded in principle, conflict avoidance, and empathy. Spouses are more likely to seek joint choices that equalize gains or losses in their

13

separate utilities, rather than choices reflecting strict compromises in attribute weights. Husbands and wives show empathy for each others' positions by altering their individual utility functions to agree more with those expressed by their spouses (Menasco and Curry, 1989). Menasco and Curry (1989) also report that outcomes were shifted toward the dominant spouse's position when an external message reinforced that position. It means that if advertising or a word-of-mouth message reaches the more dominant person, his/her weight on particular characteristic, and eventually utility, increases. This might lead to possible capitulation by the other, non-dominant, spouse. When utility functions are opposing, spouses are aware of the potential conflict and, consequently, external messages have some influence. In the case of high agreement the effect of external message can be zero. Initial disagreement and handling of it will be discussed in section 2.2.2. Research of Menasco and Curry (1989) concludes that attempts to position products in a way that appeals uniquely to one or the other gender based on separate models for husband and wife is inappropriate. This implies that multi-attribute product that is to any extent a subject of joint decision-making has to be positioned somewhere between male and female individual and selfish preferences.

2.1.2 Family factors of joint decision-making Besides personal factors there are also family factors that influence joint decision-making. Joint decision-making in family is affected by social class, family roles, communication patterns, family type and other factors. In general middle class families tend to have more joint decisionmaking than lower and higher class families (Sheth, 1974; Komarovsky, 1961 as quoted in Jenkins, 1980). This is because higher social classes, due to higher education level, tend to have more equality in family. But due to scarcer resources middle class engages in more joint decision-making. Next determinant is family role orientation, where families with more pronounced roles tend to have more autonomous members that engage less in joint decision-making. The life cycle of a family also has a role. Next sections will look at most important family factors that form the joint decision-making process. 2.1.2.1 Family roles Family roles have strong ties with individual characteristics of spouses. Some couples accept democratic companionship while others stick to traditional, matriarchal model of the family and some even have a model with the wife in the dominant position. The model of the family depends mainly on the individual’s childhood family and the psychogenic and social backgrounds of the couple (Lu, 1952).

14

In general, marital role influence will differ by product class, the stage of decision process and among families (Davis, 1976). Different roles in decision-making process come also from two distinctive family roles – instrumental role and expressive role. Instrumental role (task specialist) in a family is usually played by male and expressive role (socio-emotional specialist) by female. Because family is an economic and a social unit, specialized roles that complement each other are expected to evolve to preserve the integrity and increase the efficiency of the family (Crano and Aronoff, 1978). "[The] more instrumental role in the subsystem is taken by the husband, the more expressive the wife. ... [The] husband has the primary adaptive responsibilities, relative to the outside situation ... whereas the wife is primarily the giver of love" (Parsons and Bales, 1955, as quoted in Crano and Aronoff, 1978). The instrumental role is to mediate between the family and the outside world as the money earner, while the expressive role is to maintain internal family relationships and the home (Fitzpatrick and Indvik, 1982). If these roles are quite clear in family then in cases of high disagreement wife might adjust to husband’s opinion in order to maintain harmony and peace. Stronger instrumental and expressive role division in family can be mostly assigned to traditional family type that will be discussed in next section. Implications on marketers from family roles’ perspective are that, because of the different roles of the family members in the decision making process, they should not consider the family as a whole. Rather, marketers should make use of the role specializations within the family and of the knowledge that the structure and the dynamics of the family are determining factors in the family buying decision process (Holdert and Antonides, 1997). Ferber and Lee (1974) claim that a clearer understanding of family decision making would come only from a detailed evaluation of the relative roles played by spouses in the purchase decision process. 2.1.2.2 Family types To continue discussion of family roles this section looks at the family types (modern vs. traditional and weakly vs. strongly cohesive) regarding conjugal family and change in family types over time because in order to know where we are going, we have to know where we come from. One of the most popular ways of looking at family types is by assessing family on two structural dimensions - power and cohesion. “Potential power is the ability of a person to change the attitudes, opinions or behavior of other people. Influence is the consequence of the active or passive exercise of power. In a relationship, the division of power has been described as traditional versus modern and considered to be a continuum. The distribution of power in the relationship affects the way decisions are made” (Holdert and Antonides, p50, 1997). The other dimension, cohesion, is indicated by the degree of harmony in a family, the degree of interest in each other and the coalition dynamics. In general, cohesion includes the

15

(degree of) emotional bonds between family members (Ndubisi and Koo, 2006). Division of classic family types and basic characteristics are presented in Table 1. Family type

Traditional Power dimension Modern

Weakly cohesive Cohesion dimension Strongly cohesive

Characteristics Strong, traditional family role differentiation Autonomic decision-making Coalition formation Equal division of power between partners Short power distance between parents and child Joint decision-making Low interdependence Disharmony Egoistic, individually driven Strong interdependence Harmony Altruistic, consideration for others

Source: Holdert and Antonides, 1997)

Table 1: Classification of family types Power dimension The responsibility of the husband in a traditional relationship is earning the money whereas the wife is responsible for housekeeping and child care. Davis (1976) describes that there exist large differences in authority between husband and wife. These differences may frequently take the form of a hierarchical structure. A strictly hierarchical family has a patriarchal structure, where the husband and father is considered the head of the family (Hagenaars and Wunderink-Van Veen, 1990, as quoted in Holdert and Antonides, 1997). Because family roles are clearly divided, there are specific tasks performed by husband and wife that lead to more autonomous decision-making as they are solely responsible for a particular decision area. In households with a more traditional orientation, the husband would be expected to have the greatest influence in the decision-making process, while the wife’s role would be to support her husband’s decisions (Qualls, 1987 as quoted in Henthorne et al., 1997). Modern family on the other hand is characterized by high degree of joint participation in carrying out tasks and taking decisions (Davis, 1976). Influence in decision-making is equal and family roles do not have strict boundaries. Because there is less role segmentation many decisions become joint as tasks aren’t assigned to one particular spouse. Because decisions are made together conflicts arise due to different opinions (Holdert and Antonides, 1997).

16

Modern families usually are the ones where both spouses are active in the labor market. When looking at all states in the world, these families usually come from industrialized countries (Duxbury et al., 2007). In their research, Duxbury et al. (2007) reshape the terms “modern” and “traditional” to match Western standards where women have high labor market participation rate. This new model has two dimensions - gender and work type (“career” job or “earner” job). Career job is one that is highly paid and requires more commitment, whereas earner job is solely for financial reasons. In this model there are 4 family types – dual-career, dual-earner, new traditional and status reversal family. Here a “new traditional” family is one where husband has a career job and wife an earning job. Because of possibly lower education of husband (more traditional views) in a dual-earner family, it could also be considered to possess traditional characteristics. Social exchange theory suggests that dual-career family holds the greatest potential for equality in work and non-work roles. In a status reversal family where wife has a career and husband an earner job, both members are expected to be more egalitarian in their values. It has to be noted that Duxbury’s model has not been tested empirically and some aspects of it are ambiguous. This division into job types is linked to the comparative resources theory – wives decision-power increases as she “switches” from earner to career job. It has been found that with increasing age of spouses, and increased length of marriage, cases of joint decisions decline; and one member or another is increasingly likely to decide alone (Wolgast, 1958; Jenkins, 1980; Martinez and Polo, 1999). This might point to older couples representing more traditional families, whereas couples who get married nowadays have more equalitarian way of thinking. Extent of joint decision-making in connection with age and years in marriage will be discussed in section 2.1.2.3 about family life cycle. The trend over the past several decades show increased preference for equitable and interchangeable gender-role allocation and decreased preference for traditional or differentiated gender-role behavior (Allan and Crow, 2001; Beaujot, 2000, as quoted in Duxbury et al., 2007). From data of female dominance in typical family product purchase Holdert and Antonides (1997) conclude that the modernization of the Dutch family is only beginning. Bardenheier et al. (2011) report that 33% of fathers in USA say they take on the role of a traditional mother. This trend emerges together with economic pressures and blended family models that have redefined individual roles within the family – skill sets have replaced gender, and 62% of mothers and 54% of fathers feel that parenting roles will be redefined away from the traditional “mom and dad” roles of the past. This creates a new opportunity for marketers to think in terms of skill set versus gender, opening up the entire family as a target. The power of the family budget is equally divided among spouses. Everyday shopping is not anymore a task assigned only to females, so marketers will need to change their lexicon accordingly.

17

Cohesion dimension Traditionally families are divided into strongly and weakly cohesive families. A strong cohesive family makes more joint decisions when choosing products than a weak cohesive family (Ndubisi and Koo, 2006). In the cohesion dimension Fitzpatrick and Indvik (1982) divide relationships into 3 categories instead of 2, like Holdert and Antonides (1997), first one being traditional where relationships are very interdependent with great amount of companionship, sharing of time and space and willingness to engage into relational conflicts over important issues. The second type is called “independents” that share relatively a lot with their mates but draw definitive boundaries of time and emotional space. Just like “traditionals”, “independents” do not avoid conflicts and engage in some joint decision-making. Third and most unclear type is “separates”. They seek to avoid conflicts and are lacking interdependence. In the introduced order, these types could also represent decreasing willingness to engage into joint decision-making as more conflicts could arise if both parties seek to make a compromise decision. Traditional couples would engage in joint decision-making even though it might lead to conflict of interests whereas on the other end separates will make more autonomous decisions that do not require use of any problem solving effort. Research has shown that a higher degree of cohesion is associated with a more harmonious family life and less self-centered decision making by the family members (Kirchler, 1989, as quoted in Holdert and Antonides, 1997) and cohesive families relatively often evaluated alternatives jointly, frequently took into consideration another’s desires and ran into conflicts less often (Holdert and Antonides, 1997). To conclude – the extent of joint decision-making can be illustrated as moving opposite directions depending on the power and cohesion of family. Families with equal power distribution (everyone has a say in most of the decisions) and high cohesion (high consideration for others) tend to have more joint decision-making. On the other hand families with unequal power distribution (members have more say in some decisions and less in other) and low cohesion (independence, egoism) tend to have more autonomous decision-making where each member has his own decision area allocated according to his role.

18

Joint decisionmaking

Equal High Modern families Strongly cohesive families

Power Cohesion

Unequal Low

Autonomous decisionmaking

Traditional families Weakly cohesive families

Figure 3: Family type differences

2.1.2.3 Family life cycle The point in family life cycle can affect the level of joint decision-making and agreement between husband and wife as well as the type of it (Jenkins, 1980). Family life cycle goes through the phases where four main variables change – marital status, career advancement, age of spouses and age of youngest child. Theories usually indicate an impact of two of these variables, most important being the age of spouses and youngest child. Presence of young children in the family has a powerful impact on the way how couples organize their marital roles. This demonstrates the importance of taking the stage of the family life cycle into account in the explanation of conjugal role segregation where more segregated roles mean more strict division of tasks and autonomous decision-making (Hill, 1988). Wives have been found to lose influence vis-a-vis their husbands during the child-rearing stage of the family life cycle (Blood and Wolfe, 1960). This means that her role of a mother becomes more important but her other roles and influence on other family life areas might decline. When a child is small, wives invest heavily in rearing and have less time to be involved in decision making and less to contribute in terms of financial resources and information. Areas of decision-making that involve the child (like a vacation of whole family) are more female dominated in life cycle stages where children are present. This is because the mother is more responsible for kids and therefore holds more decision power (Fodness, 1992). According to Davis( 1976) sociologists have found traditional role ideologies more common within families in later stages of the life cycle. It might mean that they do not adapt to changing gender roles and preserve their traditional world views with more gender segregation. Newly wedded couples tend to have less autonomous decision-making style than senior couples (Sheth, 1974).

19

2.1.3 Other factors affecting family decision-making There are factors beyond those of family and those of individuals that affect decisionmaking process. Additionally decision-making is affected by importance of the purchase, time pressure and perceived risk of making a wrong decision. Importance of purchase The greater the cost of the product or service considered and the lower the family income, the greater the tendency for two or more family members to be involved in the decision process (Jenkins, 1980). Important, one-time purchases that imply high expenditure are likely to be handled with more joint decision-making (Martinez and Polo, 1999). In contrast to nondurables, purchases of durable goods are often preceded by a progression of interrelated decisions and activities through time. Husbands, wives and children have more opportunities to become involved at one or more steps in the process. Besides, it can be assumed that family members are also more motivated to participate, since the purchase of an automobile, for example, often delays other purchases, given families' budget constraints (Davis, 1976). Time pressure The more a family is pressed for time, the more there will be autonomous decision-making. One of the main factors causing increased time pressure on the family is the rise in the number of working wives (Sheth, 1974). Since 1950 USA female labor force participation rate has climbed from 33.6% to59.3% in 2005 (USDL, 2012). This is interesting because, as seen previously, there is more equalitarianism and joint decision-making between spouses if both of them are employed and if wife contributes significantly to the family budget. At the same time high involvement in the labor market implies restrictions on time available for family matters. This might mean that decision-power of a woman with a career increases in family but this power is exercised in autonomous decisions instead of joint ones. Time pressure can also affect decision-making in the family life cycle when spouses have a small child. As mother is more involved with kids she loses relative influence not only through less contribution to family budget (if work hours are reduced), but also she has less time to invest in decision-making. Risk perception Risk perception goes inseparably with importance of the purchase and the price of it. If purchase implies high expenditure there is high perceived risk that family budget might be considerably affected if product fails and it has to be replaced. Just like the importance, the risk is also positively correlated to the extent of joint decision-making. Higher risk results in greater joint decision-making among members primarily because the negative consequences affect the whole family either directly or indirectly, therefore they seek guidance and support from one another (Sheth, 1974). Also, Putman and Davidson (1987) found greater reliance on autonomic decision making for “less risky” purchases.

20

As a conclusion of factors that influence the extent of joint decision-making in a family, a table can be made that shows individual, family and other factors that shift the distribution of decision-power towards either to the equal or disproportional / autonomous distribution. The left side presents the characteristics of a factor that shifts the decision-making to more equal type and the right side show the ones that will make decision-making more autonomous by individual spouses.

Individual factors

No conflict in Conflict with parents consanguine family Interest and investment Equal interest by both Low/high interest by spouses one of spouses Comparative resources Equal contribution to Husband contributes budget / Wife more than wife contributes more Least interested partner Equal value outside marriage

Husband / wife has higher value outside marriage

Family factors Family roles Weakly defined, equal Highly divided, pronounced Family types

Autonomous decisions

Joint decision-making

Attitude towards gender roles Flexible, equal roles Strict, divided, traditional roles Education High level education Low level education, blue collar job Low difference of High difference of education levels education levels between spouses between spouses Family

21

Strongly cohesive Weakly cohesive family, family, individualistic interdependent members members Modern family with Traditional family with equal division of role differentiation power Life cycle Young couple Older couple No kids; grown-up kids Small children

Other factors Importance of purchase High Low Time pressure Non-existent Existent Risk High Low Table 2: Decision-making factors and their effect on decision-making type

2.2 Decision-making process The factors of joint decision-making discussed in previous section shape the decision making process, if more precisely – roles in it. This section will discuss decision-making process in terms of the stages and the roles in them as well as the sub-decisions that are made during purchasing process in the family. Normally, the five stage model of product decision-making include problem recognition, information search, evaluation of alternatives, purchase decision and post-purchase behavior stages (Kotler et al., 2009) but compared to an individual’s buying decisions, family’s buying decision tend to be inherently more complex because of the variety of predispositions and the underlying cognitive worlds of members of the family (Sheth, 1974). In each of the stages family decision-making can be seen as joint, wife-dominated, husband dominated or individualized (Harcar et al., 2005). In general, husband dominated decisions usually regard purchases like cars, liquor and lawn mowers while the wife dominates decisions regarding house appliances, food and childcare products. In previous research, differences were also reported in search, decision and evaluation processes for products and services (e.g. Turley and LeBlanc, 1993; Zeithaml, 1981, as quoted in Yang et al., 2006). In addition, it was found that the decision process could differ for various types of service (Hill and Neeley, 1988, as quoted in Yang et al., 2006).

22

Next sub-chapter 2.2.1 will look at joint purchase decision-making stages and roles associated with them. Afterwards, in section 2.2.2 handling of conflict will be discussed as it can be a major consequence if there is more than one person involved in purchasing process.

2.2.1 Roles, stages and decisions in purchasing process Two approaches exist in this area. First one is linked to the well-known product decisionmaking process and includes roles like the initiator, influencer, decider, buyer and user. Davis and Rigaux (1974) identify only 3 phases (problem recognition, information search, and purchase decision) and omit evaluation of alternatives as it is very closely related to information search. The other approach is less connected with the purchasing sequence and includes sub-decision roles – who decides upon brand, color and model, when and where to buy, how much to spend (Green and Cunningham, 1980). First we will look at decision roles that follow decision-making stages. Problem recognition (role of initiator) First stage according to Kotler et al. (2009) is problem recognition. In the terms of purchasing roles this stage represents the person who initiates the purchase (for example, the individual who recognizes the problem or need for the item). This role is particularly important to marketers as they would target direct advertising appeals that intensify the initiators buying motives. Sheth (1974) notes that the initiator of buying determined by the life style and roleorientation of the family. Usually person with the most dominant position in family initiates purchases, but if family is more equalitarian then initiation will not be centralized on any member. Information search (role of information provider) Second stage in purchase process is information search which is performed by the person who has a role of information provider and to some extent – the role of the influencer. Information search can be related with product options as well as purchase places. Information provider usually will be the person who has the most expertise in various aspects of a specific decision. Usually husband is concerned with technical and economic aspects and wife with decorative and ornamental ones (Sheth, 1974). It is essential to know who of the spouses looks for information and what are his/her sources as it will help marketers to create and deliver information about product in the most appropriate way. Evaluation of alternatives Third stage in conventional decision-making is evaluation of alternatives. However researchers do not assign a role to this stage during joint decision-making as it is closely related to the information search stage. During this stage an important matter is conflict and the handling of it. This topic will be addressed in the next section.

23

Purchase decision (role of decider) During the phase of joint purchasing process of greatest interest from the marketers’ and the researchers’ point of view is the actual purchase decision. Given that the final decision stage may be viewed as the culmination of the decision-making process, it is logical to see this stage as possessing the most importance; thus, resulting in the earlier stages of the process being viewed as less important (Ford, et al., 1995; LaTour et al., 1992, as quoted in Henthorne et al., 1997). It is also possible that a person who plays the role of a decider is not the most important person in the decision-making process. She might technically perform the decision but her range of product options is heavily changed by influencer or opinion leader in family. For example, husband is the one in family who decides what family car to buy. But the wife exerts influence and convinces him that the car has to ensure highest possible safety because of small children in the family. In this case the husband will still choose the product, but he might purchase a different brand than he would normally do under conditions when it would be his decision alone. Aside from this aspect, marketers are interested in raising the interest about their product in the eyes of the final decision-maker. Even though researchers do not add more weight to this stage, it is the most important and if marketing specialists have to advertise and inform someone from the spouses, it should be the dominant person in this stage. Actual purchase (role of a buyer) Next role in the decision-making process applies to the person who performs the actual shopping and buying. Since the mother is, in most of the cases, the manager of family budget, she does most of the shopping and buying (Sheth, 1974) of everyday products. The buyer in case of more important product can be the one who has the best numerical or price negotiation skills. To sum up, there are 3 main decision roles – initiator, information provider and decider, last one being the most important. Most conflicts arise during the stage of evaluation of alternatives where individual differences between spouses, discussed in 2.1.1, play a major role. Next section will elaborate on disagreement and how it is handled in family. The second approach that is used quite often is assigning relative roles to sub-decisions, most popular being “which product to buy”, “where and when to buy” and “how much to pay”. Here the final decision is divided into different aspects. The process how spouses come to an agreement on these sub-decisions is not researched. Most important here is the role and power distribution in particular sub-decisions. The patterns of how a spouse becomes more dominant have been discussed in previous chapters. Marketers should take into account all decision-making roles. As they can vary from country to country, also the marketing program should be adjusted to fit role positions in that particular society. Products that are sold with the same strategy in all markets will most probably fail, because of the lack of adjustment and lack of knowledge on which of the spouses play particular

24

roles when purchasing a product. For example, in a case of skiing vacation, if the role of information gatherer is played by the wife but marketer assigns it to the husband and advertises only in men magazines, product will have less success than it might have.

2.2.2 Disagreement and handling of it Conflict and its solution, during joint decision-making process, are factors that lead to one of the spouses being more influential than the other. Conflict Sheth (1974) describes that conflict can arise if there are differences in goals and differences about relevant alternatives or goal-objects. Here each family member’s individual cognitive world and buying behavior comes in. Sheth (1974) also writes that “Presence of inter-member conflict in joint buying decisions entails attempts to resolve it prior to choosing specific brands. These attempts are tactically different, and a particular type is more appropriate if conflict has arisen primarily due to differences in evaluative beliefs as opposed to in buying motives”. It means that, according to the type of conflict, resolution can take one of these forms – problem solving, persuasion, bargaining or politics. Conflict solving strategies Su et al. (2003) divide conflict solving strategies into two types – consensual and accommodative. If consensual strategy is prevailing, family members will most likely engage in problem solving and consider alternatives until they find a way to satisfy all the parties involved. Contrary to consensual strategy, accommodative decisions cannot be satisfactory to both spouses. This implies that one family member may use bargaining or persuasion in order to get his/her way, while the other may compromise. If buying motives are the same for both spouses but they differ in thoughts about type or brand that seems most appropriate then problem solving is the most common approach (Sheth, 1974). For example, problem solving is used if spouses want to buy a family car but can’t decide if it should be from the compact class or full size multi-purpose automobile. Persuasion is used if there is disagreement about sub-goals of purchase. In this case no information is gathered and there is more interaction as motives are evaluated. For example, after the wife initiates a purchase of a new car, husband persuades her to postpone it because the old car is still in a good driving condition. Bargaining means that joint decision-making is turned into autonomous after spouses acknowledge disagreement. One spouse steps out of the decision-making and receives some kind of favor in return (Sheth, 1974). If we continue the example with family car, here wife could give up her decision power and husband will make an autonomous decision about which car to buy. In return wife will be able to choose furniture for the new apartment without consulting her husband.

25

Settlement of conflict may extend beyond one decision episode, where one decision serves as the reason or consequence of another decision. This way decisions form a decision dynamic (Davis 1976). Even if decision-making has been accommodative the previous time, spouses tend not to return coercion in a particular purchase decision. Additionally, spouses do not use coercion consistently across purchase decisions, which may be due to spousal equity in family decision making and higher satisfaction from purchase if other spouse’s opinion is also considered (Su et al., 2003). This can be used in advertising by communicating the target person ending up with preferred brand after he has convinced other spouse or reversely – being able to get preferred product after being coerced in previous purchase episode, as a symbol of fairness. In either case it is useful for marketers to have insights into decision-making as it allows them to create natural approach to the target person. As gender roles are changing from "traditional" to egalitarian, the view of family decision making is changing from consensual to accommodative (Bonfield, Kaufman, and Hernandez 1984, as quoted in Su et al., 2003). Level of disagreement Ward (2007), using research of California consumers, indicates that initial disagreement about product choice between spouses is essential in changing decision roles. Women’s preference was prevalent in decisions regarding within-category choices, where the decision involves selecting a specific brand or model. This finding was also true when the decision involved across category choices where different product options were presented, but where the spouses had indicated low levels of disagreement in their initial product evaluations. However, in those situations involving products from across category choices, when spouses had high levels of initial disagreement, the final decision was more likely to favor the males preferred product than the females. If there are many product options, then there is a possibility of higher disagreement. At higher disagreement male opinion prevails. Females are more concerned about sustaining harmony in the family (expressive role), and will often adapt to male opinion in order to avoid conflict situations. Males, in contrary, are more ego-centric and in cases of high disagreement they often try to demonstrate their stance and decisionpower. This means that in product markets where there are many substitutes for a product it is favorable to market the product for men as they have more impact on family decision. In this situation more across category decisions have to be made and there is possibility of high disagreement. In case when there is one product but the choice is between brands, females’ preferred products are bought. Disagreement and different preferences arise from assigning different values to products and their qualities, which was discussed in section 2.1.1.2. Holdert and Antonides (1997) report that in the cases of disagreement, discussion and gathering of information is used most, followed by negotiation as a way to solve the conflict. From this it appears that products that could bring up more disagreements about brand choice

26

should be supported with more information from marketers as families tend to look for additional knowledge about brands when in conflict.

2.3 Decision-making regarding cars Since previous parts of theoretical framework didn’t involve any particular product, last major subtopic in theoretical framework will give an insight into joint decision-making regarding a car as a product. In section 1.3.1 we will try to apply theories discussed before, shape automobile purchase decision according to them and add theoretical material that is more relevant to car buying. Baterman et al. (2008) describes car purchasing behavior from household’s perspective. When households initiate a car purchase, it is assumed that it faces a choice; either buy one particular type of car from the finite set of car varieties available in the market or make no purchase at all. In making that choice it is supposed that households assess the benefit that they would gain from purchasing each different car model. It is assumed that this choice involves weighing the advantages resulting from the purchase of a vehicle with a certain set of attributes against a different purchase and running costs. Just like individuals, households use a utility function to give a score to each option, attributing a higher score to options that provide a greater surplus of advantages over costs. Households are assumed to purchase the particular vehicle type that scores highest, provided that the utility from that option exceeds the option of not buying a car at all. Arguments of the utility function are the physical and the cost attributes of the car as well as the characteristics of the household. Characteristics of the household are financial capabilities and attitudes from both spouses and their interests. According to Baterman et al. (2008) households evaluate and assign utilities to two types of automobile attributes – physical and financial. Physical and financial attributes can be expressed in numerical values and can be subjected to quantitative research. There is also third aspect that cannot be measured – body type and design.

2.3.1 Application of the previously reviewed theories to car decision making From a gender role point of view car purchase is seen as male dominant in many countries, especially in those where a more patriarchal society is predominant. Therefore marketers can start with assessing where a state stands on a continuum with equalitarian societies (Denmark, Sweden) on one end and patriarchal societies (India, Saudi Arabia) on the other. From the relative investment theory’s perspective men might be more motivated to exert influence in the car purchase decision as they in general have much higher interest in cars, motoring, technical matters and car racing than women. If we look at comparative resources theory, males also have potential to exert higher influence in family car purchase as they contribute more to family budget. In 2009 women earned on average 17% less gross hourly than men in the European Union (Eurostat, 2012).

27

Since car purchase entails major costs, initial and upkeep, families with low income will be most probably less represented within customers. Higher income stem from higher education and as discussed previously higher education and double-earner families have more equalitarian decision-making patterns. Therefore more joint car purchase decision-making can be predicted. Many factors that fall into sub-section “other” can differentiate car purchase from purchase of other products. Buying an automobile is a major decision in most families and can be perceived of high importance; consequently it will involve more joint-decision making. Car purchase implies great costs for families and also the perceived risk is high, therefore one spouse will seek support and agreement from the other. As the consequences of buying a car might be greater than those arising from, for example, buying soap, consumer will devote great amount of effort, searching large amounts of information, soliciting advice, and agonizing over difficult trade-offs (Bettman et al., 1991).

2.3.2 Previous research on family car purchase In this chapter previous studies regarding family car purchase in other countries will be reviewed in order to gain an extensive insight and to compare it to the outcome of empirical research on Latvian consumers. Previous researches, when looking at family car purchase, differ in their approach. Mainly three types of data presentation exist in previous studies. First type is measuring how autonomous (husband/wife dominated) or joint are decisions - “what/where/when to buy” and “how much to pay”. Second type of reporting results regarding car purchase is using three main decision-making stages – problem recognition, information search, purchase decision. In both types five different labels were usually assigned –“autonomous by wife”, “wife dominated”, “joint”, “husband dominated” or “autonomous by husband”. Besides these five labels, in cases where husband’s and wife’s answers were different, individual answers were reported. Author tried to standardize results reported even though researchers have used different presentation methods. Findings from different studies can be found in Table 3, 4 and 5. A possible way of showing results of the study is to report data of first or second type by using a grand mean for the car purchase as a whole. Unfortunately two main problems exist with analyzing and comparing previous studies. The first one is that some studies like Green et al. (1983) do not elaborate on decision-making stages or sub-decisions. The second issue is the way the extent of joint/autonomous decision-making is presented. Studies like Yang et al. (2006) describe results with words by reporting one of five possible answers (discrete data) instead of a more precise option used by Webster (1994) where numbers are used (continuous data). Harcar et al. employ another option – they present the extent of joint decision-making and husband/wife dominance graphically. Walsh (2010) says that the transition from being perceived as secondary to male purchasers and being interested primarily in the decorative features of vehicles to becoming buyers in their

28

own right and persons who understood the mechanical functions of their cars was related to fundamental changes in women’s position in American economy and society. As more women gained their driving licenses and as a new generation of females acquired driver education as part of their upbringing, women came to demand more from either the family vehicle or their personal one as more households bought two automobiles.

29 Decision-making stages

Year

1997

2006

Country

Japan

Singapore

Problem recognition

Husband dominated

Husband dominated

1974

Belgium

Husband autonomic

1994

USA (English)

Husband dominated

1994 1994

USA (bilingual) USA (Spanish)

Information search

Husband dominated

Husband dominated / Joint

Husband dominated

Husband dominated Husband dominated / Joint Husband dominated / Joint

Joint

Joint

Table 3: Car purchase in different countries [1]

Purchase decision

Husband dominated

Husband dominated

Joint Husband dominated / Joint Husband dominated Husband dominated

Comments Even though all stages both, wife and husband, regard as husband dominated, stage 3 shows the most joint decision-making. Females see a more pronounced involvement of the males in the search for information on an automobile purchase than the males do themselves. This might mean that males provide information from their memory and do not search for additional and wives basically assign this role to them

Reference

Henthorne et al.

Only decision-making stage where husband and wife had disagreements about extent of domination is information search, which wife regarded as joint but husband as dominated by him.

Yang et al.

The purchase of a car is very varied in its stages, moving from autonomic for problem recognition, to husband dominated for search and evaluation of alternatives, and finally, to joint for the final decision.

Davis and Rigaux

This study was conducted among USA families with Hispanic ethnic background. They were divided into 3 groups according to language preference at home. English speaking and bilingual families showed opposite role development through purchasing stages than Spanish speaking. In families with high ethnic identification final decision is less joint than previous stages comparing to English speaking Hispanic families.

Webster

30

In this table researches on car purchase are presented that divided roles according to sub-decisions – where, when and which car to buy and how much to pay for it. Sub-decisions

Year

Country

2005

Turkey

2005

Guatemala

2005

Vietnam

2005

USA

2005

Canada

1958

1994

USA

Saudi Arabia

Which car? Autonomous by husband Autonomous by husband Autonomous by husband

When? Autonomous by husband Autonomous by husband Autonomous by husband

Joint Husband dominated

Joint Joint

Where? Autonomous by husband Autonomous by husband Autonomous by husband Husband dominated Husband dominated

This study presents only one sub-decision - when to buy a car. Both spouses presented very high agreement. In 51% if the cases this decision was regarded as autonomous by husband and in 28% as joint, while all other options received under 10% of responses.

Autonomous by Husband/joint

Autonomous by husband

How much Comments to pay? Autonomous by husband Autonomous by husband It is very obvious that Western societies are opposed to Autonomous developing countries regarding extent of joint decision-making. by husband Above that, USA stands out with very high level of reported joint decision-making when it comes to choosing time of car purchase. Joint Husband dominated

Autonomous by husband

Table 4: Car purchase in different countries [2]

Study excludes sub-decision on which particular product a family will choose. Because of Islamic traditions, patriarchal culture and strict gender roles, car purchase is solely assigned to husband. Autonomous Saudi Arabia is the only country in the world where women aren't allowed to drive a car. by husband

Reference

Harcar et al.

Wolgast

Yavas et al.

31

And lastly, in Table 5 we present findings from studies that reported results in an aggregate state without dividing them into decision-making stages or sub-decisions.

Year

Country

Who decides

1980

USA

1980

Guatemala Husband Autonomous Venezuela by husband Husband USA dominated Husband France dominated Husband Holland dominated Husband Gabon dominated

1983 1983 1983 1983 1983

Joint

Comments This study researched following decisions regarding automobiles - which make, which model, what color, when to buy, where to buy, how much to spend. But it presents only mean result from all these decisions.

Results were presented in a figure with no numerical indicators. It is hard to see precisely to what extent car purchase is husband dominated or autonomous by him. USA again demonstrates that a car purchase, even though husband dominated, is close to joint, whereas in Venezuela this decision seems almost 100% autonomous by husband.

Table 5: Car purchase in different countries [3]

Reference Green; Cunningham

Green et al.

32

It is complicated to compare results from different studies, as they have used different methods of measuring the extent of joint decision-making in automobile purchase. Not only have they used different methods but also different levels of joint decision-making. Some studies employ only three levels – male dominated, joint and female dominated while some of them have more detail and have five levels.

2.3.3 Attributes of automobile In automobile industry, different attitudes towards car mean that the same car can satisfy slightly different needs of husbands than those of wives. If different needs are satisfied then different attributes in the car are searched for by each of the spouses. This section will shortly look at how customers look at car’s attributes. This will also shape the empirical research part of this paper. “Global diversity in consumer behavior in family decision-making and purchasing may lead to alternatives in marketing strategy in order to appeal to a specific region and country. Marketers should adjust their strategy to accommodate global consumer differences” (Harcar et al., 2005). Even though findings in Western societies can often be generalized towards other countries that are similar in their societal characteristics, it is fruitful to conduct a research on a particular market. For this reason parts 3 and 4 follow that will investigate joint decision-making between spouses in Latvia. By employing an utility regression model on consumer preferences Baterman et al. (2008) observed negative coefficients on such car cost attributes like purchase price, fixed costs and variable costs since these represent losses in wealth and a positive coefficient on resale price since this represents money back in the bank. They also find that each of the costs parameters is highly significant. Clearly, households consider each of the four cost attributes in making their purchasing decisions, but the different absolute values of the four parameters indicates that those various costs are treated with different weights. From this UK data it appears to be important to have low fixed yearly costs. Almost all of the UK research coefficients estimated on the physical attributes are also significant. Physical attributes tested were engine size, CO2 emissions, gearbox type, number of gears, size of the car, count of doors, power, acceleration, no. of air bags, air conditioning, alloy wheel rims, ABS system. The notable exception is provided by the coefficient on anti-lock braking system which remains stubbornly and anomalously negative.

2.4 Sub-conclusion The sub-conclusion of theoretical framework part will answer first two research questions. The theoretical framework section of this paper discussed two main aspects – decision-making factors and the process. At the end of section 2.1 (factors of joint decision-making) a two sided summary table was presented. Left side represents states of factors that drive the decision-making

33

towards more joint type. While the other side of the table shows the opposite states of factors that pulled the decision-making towards more autonomous type with pronounced roles in family. There are two main categories of factors – individual and family. Besides these two, other factors like importance of the purchase, time pressure and perceived risk exist. Table 2is an answer to the first research question. Factors discussed in section 2.1 and presented in Table 2 affect spousal roles in joint purchase decision-making process. Decision-making process goes through particular stages. In joint decision-making mostly 3 stage model is used. In three decision-making stages (problem recognition, information search and final decision) a different role can be assigned to each stage. These roles can be played autonomously by an individual or jointly by both spouses. Researchers usually measure the extent of involvement in these stages in a scale from 1 to 5 where “1” represents autonomy by the wife, “3” an equal involvement and “5” autonomy by the husband. In Western societies which are mostly transitional equalitarian or equalitarian, this rating rarely reaches any of the poles. There are two other methods of assessing, how joint or autonomous is the decision-making. They are not related to a process but have distinctive sub-decisions that have to be made in order to buy a particular product. Most popular sub-decisions are where, when, which car to buy and how much to pay. Not very popular method is to use product’s characteristics as sub-decisions. Characteristics can be, for example, size of the car and fuel efficiency. The same 5 point system, like in the case of 3 stage method, is applied to particular sub-decisions. Table 3, 4 and 5 present data on decision roles / power from different countries. For a car purchase, ratings range from 3 (joint decision) to 5 (autonomous by husband), depending on decision-making stage or sub-decision. There has not been an observation of female dominance in any decision-making stage or sub-decision. Only one study indicated that across automobile subdecisions wives were more involved than husbands when selecting an automobile color (Davis, 1970). In USA, Canada and Belgium car purchase is a joint decision in general or in particular stages or sub-decisions. Rest of the countries (Japan, Singapore, Turkey, Guatemala, Saudi Arabia, Venezuela, France, Holland and Gabon) showed male dominance or autonomous decision-making by husband. Interesting that in USA car purchase has been observed as being joint for many decades without much change. This point in the paper marks the end of theoretical framework part. The most important points that will be continued in the empirical research are family purchasing roles and division of decision power because they translate into economic actions that marketers are interested in.

34

3 Empirical research methodology and analysis This practical part of the thesis will concentrate on decision-making roles. It aims to answer research questions and bring a practical perspective to this research paper. To understand the complexity of family purchase decision making more fully, it has been stated that the relative role and influence of the husband and the wife should be examined (Strodtbeck, 1951 as quoted in Henthorne, 1997). It is important to know which spouse in the household has the dominant influence in decision-making so the marketing and promotion strategy can be tailored or oriented accordingly (Menasco, Curry, 1989). Theoretical framework gives us a good basis for the empirical part which will consist of quantitative consumer survey and analysis of data. Part 3.1 will introduce general issues regarding questionnaire design followed by section 3.2 where data, gathered from the survey, will be presented and analyzed.

3.1 Research methodology Empirical part has a deductive research approach where results will either support or disconfirm findings from the previously researched theoretical background. The main aim of this quantitative research is to use the consumer survey in order to gain an understanding of how Latvian families conduct a car purchase and how spouses divide roles in the purchasing process. To reach this aim, descriptive research with single cross-sectional design is used. The major objective of descriptive research is to describe something, usually market characteristics according to prior formulation of research questions. In single cross-sectional designs, only one sample of respondents is drawn from the target population, and information is obtained from this sample only once (Malhotra, Birks, 2007). Chapter 3.1 contains sub-sections that will discuss questionnaire design, answering scales, translation of the survey and sampling technique in more detail.

3.1.1 Questionnaire design The questionnaire was designed as a self-administered online questionnaire with the program SurveyXact. The survey can be seen in Appendix 1. Online survey method was used because of the distance limitations between the author and the respondents. Online survey also implies the smallest costs and with the widespread internet access respondents can be reached more easily. Advantages of an online survey comparing to faceto-face and mail surveys include faster gathering of answers, removed interviewer bias and higher quality of data as validity checks can be performed. Some of the disadvantages are possible exclusion of respondents that do not have internet access and researcher cannot check if a respondent represents the target population (Malhotra, Birks, 2007).

35

Questionnaire is divided into 5 sections (see Figure 4). After a language choice and introduction, a respondent had to tell more about the car and its purchasing conditions. Next, respondent had to answer questions about roles in decision-making stages and sub-decisions followed by importance of car’s characteristics and personal information.

Language choice and introduction

Information about car and buying conditions

Decisionmaking stages and subdecisions

Importance of car's attributes

Personal information

Figure 4: Structure of the questionnaire Questions were developed by the author himself but answering options to some questions were made according to those used in previous researches. Further paragraphs will provide more detailed explanation on how each of the five sections was developed. Language choice and introduction First, a respondent was able to choose a language – namely Latvian or English. Even though the survey was intended for Latvian consumers few of them chose to fill it out in English. Introduction shortly stated the intention – bachelor thesis research, and the target population – spouses who have bought a family car in recent years. Several times it was stressed that the product of interest is a “family car” as validity can be significantly raised if more specific product categories are used, for example the "family" car, or the "second car" as opposed to simply "car" (Davis and Rigaux, 1974). Respondents were encouraged by stating that the survey is anonymous and answers will be used only in an aggregate state. Information about the car and its buying conditions Questions v_1 through v_7 were asked to identify certain aspects of the car like class, age, price and if it was new at the moment of purchase, as well as information about how many years have passed since car purchase, how often the respondent uses the car comparing to his/her spouse and how many cars the family owns. In the question v_1 about which automobile classification class their car falls under, qualification system developed by the Latvian Authorized Automobile Dealers Association was used. In the first section of the survey, questions were asked to identify car purchase patterns more precisely as answers to these questions might explain differences in decision-making, as well as to gain an overall picture of families that have bought a car. Decision-making stages and sub-decisions The survey asked questions v_8_1 to v_9_8 to find out decision-making roles in Latvian families and who is dominating in particular car purchasing stages, sub-decisions and the process as

36

a whole. Three approaches to car purchase decision-making were used. Questions v_8_1 to v_8_3 use three decision-making stages to assess the extent of joint decision-making. This method has been used by Yang et al. (2006) and Webster (1994), for example. The use of large number of stages can result in reducing respondents' capability to distinguish many different stages in their assessments of their family purchase decision-making processes. Hence, the general tendency in research has been to examine three or four stages in the decision process to avoid lengthy and complicated questionnaires (Yang et al., 2006). The second method employed was based on Harcar’s et al. (2005) study with 4 subdecisions. These 4 sub-decisions were addressed in questions v_8_4 to v_8_7 where a respondent was asked to report who decided on when, where and which car to buy and how much to spend. Last questions in this section of the survey, v_9_1 to v_9_8, also employed sub-decisions but they were associated with 8 characteristics of a car. Characteristics used in this survey were inspired by those employed by Baterman et al. (2008). Slight changes were made where some characteristics were merged and least important were left out. Major objective of this sub-decision method is to compare it with other two methods and possibly provide data that can be used in advertising. Given a choice between a global measure of purchase influence and questions about influence in specific decisions, the latter is clearly preferable (Davis, 1971). Section 3.2.2 will compare all 3 methods of assessing decision-making roles. Importance of car’s attributes Questions 10_1 to 10_10 assess importance of particular attributes of a car to a respondent. Characteristics themselves were based on those used in previous questions v_9_1 to v_9_8. These questions are asked to understand – if a spouse is interested in a particular characteristic, does it translates into more decision-power when couple decides upon this attribute in one of the 8 subdecisions. Knowing which attributes are more important to consumers can help advertisers to decide from what perspective an automobile should be advertised and what distinctive qualities should be emphasized. Additionally, it is useful to see if there is any difference in perceptions between the genders Personal information To be able to group respondents, gain insights into composition of the sample and assess decision-making dynamics regarding different respondent groups, the survey asks for personal information. Respondent had to answer questions v_11 to v_15 regarding gender, age, income of family, presence of kids and marital status. Income groups were made so that the middle one contained the “statistically average couple” where both spouses earn the average Latvian salary. At the end of the survey respondent was able to leave comments on matters regarding family car purchase.

37

3.1.2 Answering scales After discussing the composition of questions, now we will examine answering options. Traditionally family decision-making has been categorized as husband dominate, wifedominate, joint, or individualized (autonomous) by one of the spouses (Harcar et al., 2005). Traditional categories also affect scales that are used to assess decision-making dynamics during particular stages and sub-decisions. This translates into, undoubtedly, most common measure - a 5point Likert scale ranging from "husband decided" to "wife decided" (Davis, 1976). Likert scale was used for all three methods of purchase decision-making. This scale is often used by researchers, Webster (1994) to mention one. Additionally Likert scale is easy to administer and respondents readily understand how to use the scale, making it suitable for online survey (Malhotra, 2007). Likert scale, which is an itemized rating scale, was also used for assessing importance of car’s characteristics in questions v_10_1 to v_10_10. If bipolar scale was used in assessing the decisionmaking, then unipolar 5-point Likert scale with “not important” and “important” on opposite ends was used in these questions. All other single item questions regarding car purchase and personal information used single or multiple answer possibilities.

3.1.3 Translation Translation from English to Latvian was done by author himself. Later three native Latvian speakers with good English knowledge and understanding of what survey is about reviewed it, suggested improvements in translation and checked understandability of questions and ease-ofchoice of answers. After the review some changes were made regarding order of questions and wording of questions. Finally the adjusted Latvian version was translated back to English to account for changes that were made.

3.1.4 Sampling technique The population of this empirical research is couples, both married and non-married, to which research results will be generalized upon. During introduction of the survey it was stressed that it is aimed at couples who have bought a family car, this way defining a sampling frame. Because of the limited resources and performing quantitative research from distance, online survey is used. Sampling technique fitted to online surveys is non-probability convenience sampling (Malhotra, Birks, 2007). More precisely convenience and snowball sampling was used, where at the end of survey respondent was encouraged to pass on the survey to other couples he/she knows. Even though non-probability sampling has limitations regarding generalizability on population, it implies less costs and time consumption which was of high importance. Questionnaire was distributed through social networks facebook.com and draugiem.lv, car and family forums and through private e-mail list.

38

Previous researches on family decision-making have been conducted by gathering data from both spouses (Wolgast, 1958; Yang et al., 2006) and also from one of them (Green; Cunningham, 1980; Green et al., 1983). Researches that used answers from both spouses indicate that, in general, husbands and wives reflect one another's judgments almost perfectly (Wolgast, 1958) or show high levels of agreement (Yang et al., 2006; Martinez and Polo, 1999) but there are also exceptions. Davis (1971) reports a low correlation between husbands' and wives' scores in his research and the conclusions reached in a study, which relies only on wives as respondents, are questionable. Because of resource and time restrictions author chose to distribute the survey by asking only one of the spouses to fill it out without restricting gender. This way author hoped to increase the sample size. Gaining a response from both spouses would be less possible if an online survey is used. And because data gathering method does not allow pairing the couples but gives only aggregate data by gender it is less crucial to have answers from both spouses. There are multiple determinants when deciding on the sample size. Most essential ones are the importance of decision, nature of research, number of variables, nature of analysis, sample size used in similar studies, incidence rates and resource constraints (Malhotra; Birks, 2007). Sufficient sample size has to be reached so that results can be generalized to wider population. Sample size in previous quantitative family studies vary from 60 (Bonfield, 1978) to 279 (Harcar et al., 2005). To improve response rate, follow-up messages in social sites and forums were posted. This resulted in 71% completion rate from those respondents who answered at least one question. In total 91 filled out surveys were gathered. Consecutively, time frame of 4 years since car purchase was implemented to improve the quality of data. Respondents might be able to recall interaction in the family better if shorter time period has passed. Even though other studies have not showed any concern regarding time frame, author decided to implement it. Even though car purchase is an important purchase, shorter time frame would increase ability to remember it. Even shorter time frame was not possible mostly because of the fall in purchasing power during the global crisis and following economic difficulties of Latvian households. In 2011 new car sales were still 29.11% of those in 2007 (LPAA, 2012).

3.2 Data analysis After the research methodology has been clarified, data gathered from the survey can be analyzed. In order to understand car purchase decision-making roles and what variables affect these roles, data analysis section will mostly concentrate on three assessment methods. Data analysis and consecutive discussion of results will try to answer these questions: 1) How are the decision-making roles between husband and wife divided in three decisionmaking stages – purchase initiation, information search and final decision? 2) How the decision power is divided in four sub-decisions – where, when, what car to buy and how much to pay for it?

39

3) How the decision power is divided in eight sub-decisions regarding different product characteristics? 4) Is there a difference in results between three methods mentioned above? 5) Which family factors affect the extent of joint decision-making? 6) Which car related factors affect the extent of joint decision-making? 7) Are there any differences in decision power distribution between particular decisionmaking stages and sub-decisions? 8) What attributes of a car should be advertised to both spouses and which to men/women individually? These eight questions are an extension of research questions 3, 4 and 5. Answering these more specific questions will give answers to the main research questions. For statistical calculations SPSS 19 was used and MS Excel was used for making figures, tables and calculating descriptive statistics. But first we will start by viewing the sample. Questions mentioned above will be answered in sub-sections 3.2.1 to 3.2.4 where 3.2.1 will begin with describing the sample characteristics. 3.2.2 presents the data from 3 decision-making role determination methods and analysis it. Data gathered in questions v_10_1 to v_10_10 about importance of car’s characteristics will be analyzed in connection with the 8 sub-decision method in section 3.2.3. Data gathered on car purchase conditions and personal information will be employed in sub-section 3.2.4 to find out what factors might affect the extent of how joint is the car purchasing process in Latvian families. And last, other possibly relevant data for marketers will be analyzed in sub-section 3.2.5.

3.2.1 Sample characteristics First, the sample was screened and cleared from responses that would reduce quality of data and representativeness. Criteria for deleting were - if car purchase is outside the time range of 4 years and responses that reported buying a very cheap car (under LVL 1000). Few respondents were removed on basis of answers that implied him/her being outside the target population, e.g. unmarried male with no kids and completely autonomous decisions, who gave the same answers over multiple sub-decisions. After the selection there were 69 cases left from 91 initially. This number still complies with minimum sample size requirement of 60. Majority of cars (43.5%) belong to the middle class and only 15.9% of all cars were new when purchased. Even though this is a small number for further analysis and implications author will assume that car buying patterns are the same for new and used cars. On average price for a car was LVL 8144.93 and it was 6.96 years old. Age of respondents ranges from 20 to 62 years, average being 34. 50.7% of respondents are females and 49.3% males. In regards to gender, this sample has high representability of population and acceptable representability regarding marital status as 39.1% of respondents aren’t married. 52.2% of the spouses have kids that live with them. Respondents are divided into five income

40

groups that author will call lowest, lower, middle, higher and highest. Sample is rather evenly spread across 4 higher income groups. This is understandable because car purchase implies rather high costs and families with lowest income cannot afford it.

3.2.2 Decision-making roles from perspective of the three methods This section will analyze data on decision making roles, compare three methods (3 stages, 4 sub-decisions and 8 sub-decisions) and consequently elaborate on the survey results from particular methods. Sub-chapter 3.2.2.1 compares results from 3 methods and chapters 3.2.2.2 to 3.2.2.4 look at particular methods and the data they produced. There are two main methods of assessing the reported levels of decision-making. Davis (1970) reports frequencies of specific answers. In our case it would mean reporting how many couples reported, for example, problem recognition as a female-dominated stage. The second method, used by Webster (1994), is to report the mean response across different decision-making stages and sub-decisions. Author did not find any arguments for or against any of these options and decided to employ the one used by Webster (1994) because of interest in the “bigger picture”. Frequencies of reported decision-making roles can be seen in the Appendix 3. Analysis will be continued by using mean results. Answers from questions v_8_1 to v_9_8 regarding the decision power in different decisionmaking stages and sub-decisions are presented in Table 6 with an average result given at the end. All values above “3” indicate more dominance by the husband. Not a single decision-making stage or sub-decision leans towards wife’s dominance when it comes to automobile purchase.

41

3 stages Problem Information recognition search 3.51

Final decision

Mean

3.90

3.90

4.30

4 sub-decisions Where? How much? 4.13 3.67

When? 3.68

Which? 3.83

Mean 3.83

8 sub-decisions

Gearbox 3.69

Equipment 3.96

Fuel Safety Space/capacity type/economy equipment 3.75 4.04 4.13

Engine size / Power 4.31

Body type / Styling 3.40

Exploitation cost 3.98

Note: Roles were measured on 5 point scale (1 = autonomous by female; 2 = female dominated; 3 = joint; 4 = male dominated; 5 = autonomous by male)

Table 6: Roles in automobile purchase decision-making process In order to assign a category of “autonomous by wife”, “wife dominated”, “joint”, “husband dominated” and “autonomous by husband” to results, five intervals of equal size are made. Intervals and corresponding categories can be seen in Table 7. 1 - 1.8

1.8 - 2.6

2.6 - 3.4

3.4 - 4.2

4.2 - 5

Autonomous Wife Husband Autonomous Joint by wife dominated dominated by husband Table 7: Decision power intervals and corresponding category Mean results from all three methods seem to be very similar – 3.9, 3.83 and 3.89. But in order to verify that, Kruskal-Wallis tests will be performed. Next sections will use Kruskal-Wallis test not only for differences between various methods, but also for differences between different decision-making stages and sub-decisions. This test will be used further in section 3.2.4 to assess if there is significant difference between two types of families based on different categories (presence of kids, marital status, etc.). Two types of families will be divided based on to what extent a car purchase is a joint decision. Because data gathered from decision-making questions are ordinal (from 1 to 5) we can employ Kruskal-Wallis test to compare different methods that assess decision-making power. This test is designed to compare two or more populations based on one independent variable and allows the data to be non-normal (Keller, 2008). Assumptions for this test are randomness of the sample

Mean 3.89

42

and independent observations. In addition, groups need to have similar distributions (Horn, 2008). The use of random sampling technique in data gathering and the use of cases only in one test group ensure meeting randomness and independence assumptions. Also the distribution assumption is met for all the tests (cf. Appendix 2). 3.2.2.1 Assessment of 3 decision-making role determination methods First we will investigate if all three methods presented in 3.2.2 are equally effective at capturing decision-making roles in family. This is done by comparing mean decision-making stage and sub-decision responses from all 69 respondents. Mean result for 3 stage method is 3.90, for 4 sub-decision method - 3.83 and for 8 sub-decision method - 3.89. All three methods show that husbands hold more decision-power in car purchase decision than wives. Hypothesis for this test are as follows: H0: There are no significant differences in reported decision-making roles between 3 methods H1: At least 2 methods show different decision-making roles. Test Statisticsa,b As it can be seen from Table 8, Kruskal-Wallis test computes significance level of 0.81 which is well above α=0.05. Therefore H0 Mean response from cannot be rejected and it means that all three methods provide us 3 methods with the same overall results. Chi-Square df Asymp. Sig.

.422 2 .810

a. Kruskal Wallis Test b. Grouping Variable: 3 methods

Table 8: Test of 3 methods

3.2.2.2 Roles in 3 decision-making stages Analysis is continued with testing if there are any changes in decision-making roles as the process goes through different stages and sub-decisions. First we start by analyzing the method with 3 stages – problem recognition, information search and final decision. Results from stages will not only be compared with each other but also with the mean result. This can determine if it is useful to elaborate research and use different stages as opposed to just asking respondents to indicate the decision-power in car purchasing in general. Here 4 groups are compared to assess if there are differences between decision-making power (decision-making roles) in them. These groups are problem recognition, information search, final decision and a mean of all previous 3. Hypothesis for this test are: H0: There are no significant differences between 3 decision-making stages and the mean result,

43

H1: There is difference at least between 2 of groups. Test Statisticsa,b Results (Table 9) show that there are significant differences between Decision at least 2 decision-making stages or the mean result. At significance power in 3 level of α=0.05, H0 has to be rejected as significance level of 0.001 stages shows high level of differences. To find which of the groups differ we Chi-Square 16.529 employ Mann-Whitney test which is similar non-parametric test to df 3 Kruskal-Wallis test. This test allows comparing 2 groups in cases of a Asymp. Sig. .001 non-normal distribution (Pallant, 2011). From Table 6 it can be seen a. Kruskal Wallis Test that final decision and the mean results are the same. Additionally, b. Grouping Variable: 3 stages Table 9: Comparison of 3 stages problem recognition and information search stages are almost the same “distance” from the mean result. Therefore it is convenient just to test if problem recognition stage is different from the mean result. Hypothesis for this test is as follows: H0: There are no differences between problem recognition stage and mean result, H1: There are differences between these groups. Test Statisticsa The test reports significance level of 0.12 > 0.05, Decision therefore there is no statistically significant difference power in 3 between problem recognition stage and mean result (or final stages decision). It additionally implies that the only considerable Mann-Whitney U 2019.500 difference is between problem recognition and information Wilcoxon W 4434.500 search stages. Problem recognition stage is the most joint of Z -1.557 all 3 and husbands have the most domination in information Asymp. Sig. (2-tailed) .120 search which can be considered autonomous by the husband. a. Grouping Variable: 3 stages In Table 11 categories are assigned to decision-making stages. Table 10: Comparison of problem recognition and mean result

3 stages Problem Information Final decision recognition search 3.51 4.30 3.90 Husband Autonomous Husband dominated by husband dominated Table 11: Roles in 3 decision-making stages

Mean 3.90 Husband dominated

3.2.2.3 Roles in 4 sub-decisions Test are continued with assessing 4 sub-decision method. Table 12 provides with an overview of the role distribution over four car purchasing sub-decisions.

44

When? 3.68 Husband dominated

4 sub-decisions Where? How much? 4.13 3.67 Husband dominated

Husband dominated

Which? 3.83

Mean 3.83

Husband dominated

Husband dominated

Table 12: Roles in 4 sub-decisions Similar to the previous method, Kruskal-Wallis test is performed to determine if there are any differences in decision-making roles between 4 sub-decisions: 1. Who decided when to buy the car; 2. Who decided where to buy the car; 3. Who decided how much to pay for the car; 4. Who decided which brand and model to buy. Hypothesis for the test are: H0: There are no differences in decision-power between 4 sub-decisions or the mean result, H1: There are differences at least between 2 sub-decisions or the mean result. Test Statisticsa,b Significance level of 0.018 < 0.05 means that H0 has to be Decision rejected and there are statistically significant differences in decision power in 4 power of husband and wife between four different sub-decisions. To sub-decisions save time we can use the analogy from 3 stage method instead of Chi-Square 11.887 testing all possible paired combinations to find out where the df 4 difference lies. In the 3 stage method there was no difference Asymp. Sig. .018 between the mean value and the lowest (problem recognition). When a. Kruskal Wallis Test we apply it to the 4 sub-decision method, it can be concluded that the b. Grouping Variable: 4 subdifference most probably is between “how much” (or “when”) decisions Table 13: Comparison of 4 sub- decision and “where to buy” as those are decisions with the most decisions and mean result difference regarding extent of joint decision-making. Even though there are considerable differences between stages, they are all in one category – “husband dominated”. Interesting that the 3 stage method produced mean result (3.9) equal to the final decision stage. The same was detected with the 4 sub-decision method where both mean result and “who decided which car to buy” stage were husband dominated with a result of 3.83. It means that in both methods the whole purchase decision-making process manifests in possibly the most important stage – final decision - and the most important sub-decision – which car to buy.

45

3.2.2.4 Roles in 8 sub-decisions The last method that can be used (but is the least popular among researchers) to understand decision-making roles in family is 8 sub-decisions. These 8 sub-decisions are related to car’s characteristics and are connected with the final decision stage and “which car to buy” sub-decision. 8 sub-decisions

Gearbox 3.69

Equipment 3.96

Space / capacity 3.75

Husband dominated

Husband dominated

Husband dominated

Fuel type /economy 4.04

Safety equipment 4.13

Engine size / Power 4.31

Husband dominated

Husband dominated

Autonomous by husband

Body type / Styling 3.40 Joint

Exploitation cost 3.98

Mean 3.89

Husband dominated

Husband dominated

Table 14: Roles in 8 sub-decisions 8 sub-decision method is not suited for statistical tests for differences between subdecisions. This is due to the nature of the questionnaire which in questions v_9_1 to v_9_8 contained an answering option “did not discuss this attribute”. For this reason there is not equal number of responses to all questions regarding particular attributes. But instead of statistical tests, it is possible to look back to other methods. Most different decision-making stages were problem recognition and information search with results 3.51 and 4.3 and as it was concluded – these two stages produced statistically significant differences. In case of 8 sub-decision method largest difference is between body type/styling and engine size/power with reported joint decision-making levels of 3.4 and 4.31 respectively. The gap between these two sub-decisions is larger than between two decision-making stages mentioned before. Therefore it can be assumed that there are statistically significant differences. From 8 sub-decisions only one is autonomous by husband (engine size / power) and one joint (body type/styling). From all 3 methods body type/styling is the only aspect of decision-making process that can be considered as joint. Result of 3.4 suggests that it is slightly joint and almost male dominated.

3.2.3 Importance of car’s attributes Survey questions v_10_1 to v_10_10 were asked to assess how important are particular car related characteristics to the respondent personally. Answers produced results that are shown in the Figure 5. Rating “1” indicates that there is no importance to the respondent and “5” indicates that this characteristic is important to him/her. There are no values lower than “2” therefore starting point is set to “2” in order to better show the differences between importance levels.

46

Importance of car's characteristics Warranty Recognition of the brand

2.89

Equipment

3.44 3.40 3.24 3.26

3.85 4.35 4.23 4.29 4.34 4.06 4.20 3.91 3.94 4.38 3.71 4.29 4.06 4.53 4.46

Body type / Styling Exploitation cost Safety equipment Engine size / Power Gearbox type Space / Capacity Fuel type/economy 2.00

2.50

3.00

3.50

4.00

4.50

Men

Women

5.00

Figure 5: Importance of car’s characteristics to men and women To find out if it is more useful to use some particular attributes of a car in advertising, again Kruskal-Wallis test is used. It will compare if consumers see some particular attributes of a car to be more important than others. Hypothesis for the test are: H0: There are no differences in importance between attributes of a car, H1: There are differences between at least 2 attributes. Test Statisticsa,b Test results (Table 15) show that there are very significant Importance_a differences between importance of particular attributes and H0 has ttr been rejected due to significance level of 0 < 0.05. This test did not Chi-Square 93.515 distinguish genders but results can be generalized to the population df 9 because the sample consists of 50.7% females and 49.3% males. Asymp. Sig. .000 By knowing which attributes are more important to each a. Kruskal Wallis Test gender it is possible to determine which characteristics should be b. Grouping Variable: Attributes advertised to both spouses together and which to male audience. Table 15: Comparison of Characteristics that are decided jointly and are important to both attributes genders can form an advertising strategy. The same is with male dominated attributes from 8 sub-decision method, which should be analyzed together with which attributes are important to men. As not a single one of eight sub-decisions is female dominated and assuming that marketing budget of a company is limited, we will not look at advertising to females. Table 16 shows three car’s attributes that are most husband-dominated and three that are most

47

jointly decided. Additionally the table shows three most important attributes to males and three for both spouses. As it can be observed, fuel type/economy is the only characteristic that is in both lists for males. Even though body type/styling is rather important to men (4.35), it is the most jointlydecided attribute.

Rank Male dominated decisions

1 2

4.31 4.13

3

4.04

Rank More joint decisions

Decisionmaking roles

1 2 3

Decisionmaking roles 3.4 3.69 3.75

Attribute Engine size / Power Safety equipment Fuel type/economy

Attribute Body type / Styling Gearbox type Space / Capacity

Rank

Importance to males

1 2

4.53 4.38

Attribute Fuel type/economy Gearbox type

3

4.35

Body type / Styling

Importance to both Rank genders 1 2 3

4.49 4.32 4.29

Attribute Fuel type/economy Exploitation cost Body type / Styling

Table 16: Decision-making roles of particular attributes and most important attributes

3.2.4 Factors that affect the extent of joint decision-making This section will try to answer question stated at the beginning of data analysis chapter – what factors determine if car purchase is a joint or male dominated process. “Joint” and “husband dominated” families For Kruskal-Wallis test to be valid, number of observations in each group has to be at least 5 (NIST, 2006). In order to be able to generalize the results to larger population, minimum number of observations is set to 25. To attain this number, two groups of 27 observations (families) are made from 4 sub-decision method. It means that 27 families are labeled as “joint families” and the extent of joint decision-making is between 2.6 and 3.5. The other set of families are labeled “husband dominated families” with an interval of 3.5-4.75. Groups have to be of equal sizes for the tests to be applicable, therefore the boundaries of intervals were broadened for “joint” and “husband dominated” categories. In next two sections the two groups established above, joint family and husband dominated family, will be tested for differences when compared across different variables. Section 3.2.4.1 will look at car related variables – price of the car, how many cars are in family, if family bought new or

48

used car and how much the respondent uses the car comparing to the spouse. Further, section 3.2.4.2 analyses if any family factors affect if family decides jointly or car purchase is husband dominated decision. Variables used here are – presence of kids, marital status, age and income. 3.2.4.1 Car related variables Due to having two groups of dependent variables – joint and husband dominated families we can use Mann-Whitney U test. This is a non-parametric alternative of independent samples t-test and has the same assumptions as Kruskal-Wallis test. Hypothesis for all four tests are as follows: H0: Two groups (joint and husband dominated families) do not differ in regards to independent variable, H1: Two groups have statistically significant differences. Test results are reported in Table 17. Conclusion is that three out of four independent variables showed no effect on the extent of joint decision-making in Latvian families. Both groups do not differ regarding the price of the car, number of cars in family and if they bought a new or used car. Independent Significance H0 Direction of variable level hypothesis impact Not Groups do not Price 0.5619 rejected differ Number of Not Groups do not cars in 0.7567 rejected differ family New/used Not Groups do not 0.3032 car rejected differ Joint families Usage of the 0.0004 Rejected report higher car common usage Note: α=0.05

Table 17: Effect of car related variables on extent of joint decision-making in family Only significant difference between joint families and husband dominated families reveals when it comes to the usage of the car. Significance level of 0.0004 indicates very high difference between the two groups. Results show that if spouses use the car more equally then also the car purchase decision is more joint. To perform the test on usage, variables were recoded into binary variables. Values from the survey with higher usage by one person received value of 0 and values that indicated more equal usage received a value of 1. High usage corresponds to answers where respondent indicated 0-25% or 76-100% usage. Equal usage corresponds to values between 26% and 75%.

49

3.2.4.2 Family related variables The same test as in previous section with the same hypothesis now is performed on family related variables. Results can be seen in Table 18. Independent Significance H0 Direction of variable level hypothesis impact Presence of .787 Not Groups do not kids rejected differ Marital status Age

1.000

Income

.561

.358

Not rejected Not rejected Not rejected

Groups do not differ Groups do not differ Groups do not differ

Note: α=0.05

Table 18: Effect of family related variables on extent of joint decision-making in family When using α=0.05, tests indicate significance levels above it. It means that H0 cannot be rejected and two groups, joint and husband dominated families, do not have significant differences in relation to independent variables used in tests. There is a possibility that the small sample size of 54, divided into 2 groups, does not allow to see differences that actually might exist in the population.

3.2.5 Other data relevant for marketers In this final section of data analysis additional data analysis will be conducted that is not directly connected with decision-making roles but can be useful for marketers. In general, Latvian families prefer middle class cars which results in 43.5 % of all cars reported by respondents. Middle class is followed by multi-purpose automobiles and SUV’s with 18.8% and 15.9% respectively. In the Figure 6, SUV and multi-purpose classes contain also small/large SUV’s and multi-purpose cars. No respondents reported buying mini, small or large multi-purpose class cars.

50

Number of cars in each class 35 30

30 25 20 15

13 10

10 5

11

3 1

1

0

Figure 6: Automobiles purchased by Latvian families by class It is possible to observe increase in multi-purpose car buyers. 12.1% of families with no kids reported buying a multi-purpose car whereas for families with kids this number is 25%. One of the respondents supports this observation by commenting: “We bought this car when the child was not planned yet. At the moment we plan to change a car to a bigger one.” Another considerable change is the reduction in SUV’s when going from no kids to having kids in family. 21% of families without kids purchased SUV’s comparing to only 11% of families with kids (cf. Appendix 4). Data also shows that the number of new car purchases increases with income, which is not surprising. There is a steady increase in the proportion of new car purchases as income grows. The new car proportion peaks at the level of 40% for the highest income group (cf. Appendix 5). Marketers of new cars should definitely aim at spouses with income of at least 1200 Ls.

51

4 Results and discussion The following chapter will discuss the results from quantitative consumer survey. These results will be discussed together with the background theory. Results from empirical part will be compared with those reported by other studies. On top of this, reasons standing behind the results will be sought. Research questions that concern the empirical part of this paper will be answered by using more detailed questions presented in the beginning of data analysis chapter. 1) How are the decision-making roles between husband and wife divided in 3 decisionmaking stages – purchase initiation, information search and final decision? Mean result shows that overall car purchase is a husband dominated process. The reason may lie in the fact that men in Latvia have high interest in cars. According to investment theory, males might be motivated to exert more influence because of their interest in the product. Problem recognitions stage is the one where wife engages the most of all 3 stages. In this stage only a problem is acknowledged and car purchase might be a solution. Therefore it is normal that the wife also actively engages in problem solution even though the product is perceived as male dominated. The highest difference in involvement can be seen when decision-making process goes from problem recognition to information search stage. Mann-Whitney test showed that the only significant difference between stages lies here. If problem recognition stage is slightly male dominated with a result of 3.51 then for information search it spikes to 4.3 and is categorized as autonomous by husband. Male interest and competence in cars possibly shows itself to full extent in information search. Also Bettman (1991) acknowledges that the information that is at the disposal of one of the spouses can determine his involvement in the information search stage. Since men in general possess more knowledge about automobiles, they take the role of information provider in the car purchasing process. Final decision stage can be seen as the most important one. A car is an important purchase and a wrong decision implies high risk to the family budget. High risk and high importance usually imply more joint decision-making (Jenkins, 1980; Sheth, 1974). In case of Latvian families final decision in car purchase process still stays male dominated (3.9) but more joint comparing to information search stage. Pattern of movement toward joint final decisions makes intuitive sense, particularly for “higher involvement items” (e.g., family car) (Ford et al., 1995). Comparing to other countries Latvia’s situation through the stages is most similar to results reported from Japan where all 3 stages are male dominated. Latvian data goes against ones observed in USA and Singapore where female involvement is more pronounced in the information search stage comparing to others (cf. Table 3).

52

2) How the decision power is divided in 4 sub-decisions – where, when, what car to buy and how much to pay for it? Car purchasing process is husband dominated also from the perspective of 4 sub-decision method. “When” and “how much” decisions are lightly male dominated and have values of 3.68 and 3.67 respectively. Most husband dominated sub-decision is “where” which is almost autonomous by him and has a significant difference form “when” and “how much” decisions. It is possibly because husband handles more technical matters in car purchase process. As noted before sub-decision “which” (probably the most important) has the same value (3.83) as the average result for all four sub-decisions. Final decision in the 3-stage method also has the same level of male dominance as the whole process. It suggests that these two decisions are the core of decision-making process. For a marketer it can mean that if a consumer survey asks to indicate roles in general, without dividing stages, results will answer questions “who makes the final decision?” and “who decides which car to buy?”. It is useful to note that husbands have more dominance in “where” and “which” sub-decisions and they autonomously gather information about car purchasing. This indicates that husbands interact more with external environment while wives add more to the process in internal (family) matters. This division of environments can be explained by instrumental and expressive roles in the family. In this case wife plays an expressive role and maintains the internal family relationships and delegates the husband to play the instrumental role and mediate between family and the outside world (Fitzpatrick and Indvik, 1982). The attained results support the statement that important products are decided upon more jointly. 49% of the respondents indicated that a decision “how much to pay” is decided jointly (cf. Appendix 3). This relates to findings of Hopper (1994) that in USA families the financing of large purchases is decided jointly in 60% of the cases. Even though the research did not look at relative contributions to family budget by each of spouses, comparative resources theory could also explain some of the differences between joint and husband dominated families. One of the respondents mentions: “At the moment of a car purchase I was the only employed and earning person [in family] and money was all mine, therefore the purchase was made basically autonomously.” Most similar results to the Latvian sample are from Canada where all decisions except “when” are husband dominated. Sub-decision “when” is the only one that is decided jointly. It can be concluded that Latvian families show similar patterns with other transitional equalitarian societies like Canada and Japan but are rather different from the USA where joint decision-making holds a strong position. Therefore media and advertising strategies could not be transferred from USA to Latvia without an adjustment. Only 8 sub-decision method shows a decision area where spouses decide jointly. So far our research on Latvian couples shows noticable male dominance. Next question will turn to discuss 8 sub-decision method.

53

3) How the decision power is divided in 8 sub-decisions regarding different product characteristics? Method with using car characteristics of this type is not used in other studies but for this paper was inspired by Davis (1970) who added a sub-decision about which color to choose besides 4 sub-decisions mentioned above. Like in previous two methods, 8 sub-decision method also has significant differences between sub-decision roles. Most husband dominated decisions regard engine size/power and safety equipment. “Engine size” falls in “autonomous by husband” category as opposed to “body type/styling” which is the only joint sub-decision in our research. Again, it is visible that husbands are more dominant in deciding on technical matters and wives participate to a greater extent when it comes to more practical or everyday issues. Roles in sub-decisions have most advertising implications comparing to other two methods. Characteristics of a car that should be considered for marketing in media channels used by both genders are body type and space/boot capacity. To see frequencies of responses in 8 sub-decisions refer to Appendix 3. There do not seem to be any characteristics that should be advertised especially to women. On the other hand males show exceptionally strong influence in engine size and fuel type/economy matters. Least useful advertising would be for safety equipment as 30% of families do not discuss this attribute. Advertising of an attribute because it is under his/her authority is not sufficient. If two carmakers advertise their cars using two different attributes that are under husband’s control but one of them is not of high interest to husband, his attention, most probably, will be captured by the competitor’s brand. Therefore later in this section importance of attributes to men and women will be discussed. 4) Is there a difference in results between 3 role determination methods? When average responses were compared, Kruskal-Wallis test proves that there are no significant differences between the results from 3 methods discussed in previous questions. Even though all methods provide the same overall results, it does not definitely imply that any method will provide with the results of best quality. Decision to use any particular method therefore should be based on the specific needs of the study and research aims. If the aim is more related to understanding the decision-making process then 3 stage method is the most appropriate. When marketers are interested in differentiating and tailoring the type of information they provide to the consumer, they could look at decision-making stages. For example, in the problem recognition stage more emotional approach could be used that promotes a situation where the product will solve a problem. In the information search stage, on the other hand, approach should be more practical and detailed.

54

If the marketers are more interested in understanding where the authority regarding particular characteristics of a car lies, method with sub-decisions, that include attributes of their choice, can be used. Sub-decision method is rather flexible and many different sub-decisions can be included, starting from “who decided on the place of purchase” to “who decided on the number of seats”. 5) Which family factors affect the extent of joint decision-making? Factors that were tested were mostly related with the life cycle of a family. Davis (1970) has indicated that husband-wife influence differences would be sensitive to life cycle factors, such as marital status and presence of children. Tests compared two groups – joint families and husband dominated families from 3-stage method on four independent variables – presence of kids, marital status, age and income. After a Mann-Whitney U test conclusion is that none of these variables significantly affect the power distribution patterns in Latvian families. When testing for only one independent variable at a time, families with joint decision-making do not differ from families with male dominated decision-making style. Even though income groups cannot be tested for differences due to small group sizes, middle class families actually are more engaged in joint decision-making as reported previously by Sheth (1974) and Komarovsky (1961). Figure 7 shows that, families in the middle income group (600-900 Ls) report joint decision-making in 39% of the cases and other answering options received less. If we leave out the lowest income group (N=3), income group with the most autonomous decision-making seems to be the highest income group. In this group 49% of the decision-making processes are autonomous by husband and 13% autonomous by wife. This again leads to a conclusion that car purchase is possibly less important purchase for them and therefore there is less consultation with the other spouse. Figure 7 was made by using average frequencies from the 3 stage method.

55

78% 49% 39%

300 - 600 Ls (N=22)

27% 25%

16%

Autonomous by husband

Husband dominated

Wife dominated

Husband dominated

2% Joint

Wife dominated

Autonomous by wife

Autonomous by husband

Husband dominated

4% 6% Joint

20%

13%

Joint

28%

8%

Wife dominated

Autonomous by husband

Husband dominated

3% Joint

Wife dominated

Autonomous by husband

Husband dominated

Joint

Wife dominated

Autonomous by wife

Less than 300 Ls (N=3)

Autonomous by wife

3% 3%

0% 0%

Autonomous by wife

11%11%

22%

Autonomous by wife

26%26%

37%

Autonomous by husband

42%

600 - 900 Ls (N=12) 900 - 1200 Ls (N=17) More than 1200 Ls (N=15)

Figure 7: Decision-making patterns by income (frequencies) 6) Which car related factors affect the extent of joint decision-making? The only car related factor that showed significant change across 2 decision-making groups was the usage of the car. If the car is bought for a joint use then both spouses are more involved in the decision-making process. It is obvious that a spouse would participate considerably less in the purchase process if the car is not for his/her use. At the same time Mann-Whitney U test showed no effect of number of cars on joint decisionmaking. It means that if there is more than one car in the family, it rather implies a different use for the second or third car instead of a situation where each spouse has his/her own car. Two of the respondents commented: “During everyday life I and wife use our family car - Opel Vectra II diesel. The big SUV (Jeep Wrangler Unlimited) is used mostly in winter, when going to countryside, traveling with a lot of baggage or pulling a trailer” and “Last car we bough was for special purpose - to pull a trailer. In general decision was made jointly but husband had few important details (trailer hook, etc.)” 7) Are there any differences in decision power distribution between particular decisionmaking stages or sub-decisions? As discussed previously, tests of all 3 methods show that there are significant differences between decision-making stages and between sub-decisions. It implies that it is useful to elaborate the main research question and use stages or sub-decisions. Elaboration can give deeper insight into

56

the patterns of spousal decision-making. If there were no differences between, for example, decision-making stages, then one research question “who dominated the car purchasing process?” would be sufficient. By “difference” it is meant - varying decision-making roles. Since there are different roles involved, “one size fits all” approach will not give the best result. When information search role is autonomously performed by the husband, different media and different message has to be used to target male in the exact information search stage comparing to both spouses in the problem recognition stage. 8) What attributes of a car should be advertised to both spouses and which to men/women individually? Attributes that are under the authority of males turn ou not to be the most important to them. The only important characteristic that men also have high influence over is fuel type/economy. Additionally, two of the most important attributes for males (fuel type/economy and body type/styling) are the same as those that both genders see as important and they are decided more jointly. This implies that marketing strategies used for males and for both spouses can be similar when it comes from what perspective the car should be shown. From Table 16 it can be seen that importance of particular attributes to males does not result in higher influence in corresponding sub-decisions. It means that even though the husbands dominate car purchase process in general they take wives opinion into account. Men and women have similar preferences regarding the car’s characteristics. Two out of male’s top three characteristics can be found in female’s top three characteristics. Namely, fuel type/economy and body type/styling. The common interests and joint interaction also becomes evident by the comment from a male respondent: “What was important - so that both of us liked the car regarding style and functionality. Before the final decision, car was showed to the wife and try-out drives were made. Even though I have a good intuition, women have it very strong. We searched for a family car, therefore wanted a wagon that would be dynamic. Choice fell on SAAB AERO 9-5 wagon. Our wishes and choice coincided.” Men hold 57% of all category B drivers licenses in Latvia. The trend of newly issued driving licenses shows that female participation in road traffic has increased significantly. If in year 2000 only 33.3% of all category B driving license holders were women, then statistics of year 2012 show 43%. Data from first 3 months of year 2012 show slight female dominance in getting drivers licenses (CSDD, 2012). Increased female participation in road traffic most probably will increase their influence in automobile purchase decisions.

57

5 Final assessment The last chapter will present final conclusions as well as some practical implications for marketers. This research paper will conclude with the notation of the most important limitations and discuss what interesting areas further research could go into.

5.1 Conclusion There were two main goals of this research. First one was to investigate joint decision-making process in family from a theoretical perspective. Secondly, it aimed to research spousal purchasing roles in Latvian families when buying a family car. First aim was reached by building a theoretical framework on factors that affect the decision-making process and the process itself. Empirical research was done by employing a consumer survey in Latvia. Factors that shape decision-making process can be divided into individual, family and other factors. More joint decision-making is expected if individuals have equalitarian view at gender roles, have high or similar levels of education. On the other end, more autonomous decisions are possible if there has been a conflict with parents, one partner is more interested in the product or contributes considerably more to the family budget. Family characteristics like equal family roles and interdependent members with equal division of power move the family decision-making towards the joint type. More autonomous decision-making is reported if spouses are older or have small kids in family (Fodness, 1992; Davis, 1976). Family decision-making process traditionally goes through 3 main stages – problem recognition, information search and final decision. Roles in this process can be divided or played more autonomously by each spouse. Decision-making process can be analyzed also without dividing stages but rather by looking at who decides on particular sub-decisions. Most popular sub-decisions employed in studies like Harcar et al. (2006), are when to buy, where to buy, which product to buy and how much to pay for it. Both these two methods, along with a third - 8 sub-decision method, were used in the empirical part of the research. Results of the Latvian consumer survey showed similar decision-making patterns to such transitional equalitarian countries like Japan and Canada. Major role in the car purchasing process is played by the husband. All 3 methods showed that, overall, car buying process is male dominated. Husbands have most domination in the information search stage which is autonomous by them. High overall influence can be partially explained by relative investment theory, where the interest in product results into higher motivation to exert influence (Davis, 1976). Autonomous role in information search stage seems to be very connected with husband’s instrumental role in family where husband mediates between the family and the outside world but wife maintains internal family relationships (Fitzpatrick and Indvik, 1982). Husband is clearly the spouse that should be the main marketing target. But marketing effort should be balanced between males and both spouses as 49% of families report that decision how

58

much to pay for the car was made jointly. Also the styling and body type of the car is a joint decision. Therefore particular stages and sub-decisions have to be evaluated by marketers in order to find what the target segment is and what channels to use to forward the message. Statistical tests show that if a car is used by both spouses equally then the decision-making tends to be more joint. Purchase of a second car does not seem to change purchase patterns because it is still used by both spouses, only in different circumstances. Even though statistically unproven, data suggests that middle class families engage in more joint decision-making.

5.2 Practical implications for marketers in Latvia The findings of the study indicate that husbands, particularly those who are “heavy users” of the car comparing to his spouse, should be considered in the development of targeted marketing strategy by automobile sellers. Marketers should take into account husbands’ high relative influence in car purchase process when developing the optimum target market identification. Attributes like fuel type/economy and body type/styling can be marketed equally successfully to the male audience and to families. As information search is autonomous by the husband, all practical information should be provided to him. One of the practical matters that should be definitely included in the message to both spouses is the price due to the fact that in 49% of the cases decision, how much to pay, is handled jointly. More emotional approach can be used to both husbands and families as problem recognition and final decision stages are less autonomous by male than the information search. The proportion of female car drivers increases every year. Therefore consumers should be surveyed regularly to acknowledge changes in the family car purchasing patterns. Increase of female drivers in last 12 years by 22.6% must be considered by marketers to account for their possibly increasing decision power.

5.3 Limitations First of all, limitation regarding data collection must be noted. One of the weaknesses of online survey is the researcher’s inability to check if the respondent really belongs to the target population. Also couples that do not use internet are excluded. Source of error in answers could result if questions about relative influence were not actually relevant. For example, the choice of equipment for an automobile may have been based on what was available rather than a conscious choice, or the breakdown of the old car may have predetermined when a new car would be purchased instead of being a decision initiated by one of the spouses. One spouse might answer such questions in terms of socially accepted role definitions, the other on the basis of how this decision might have been made (Davis, 1971). Purchase process might not go through the normal stages for some respondents; therefore data can be distorted due to different purchasing process. A comment from a respondent highlights

59

this possibility: “We changed the car for a newer one, because unexpectedly got an offer for a car in a great condition and good equipment.” In terms of sample size this empirical research only marginally satisfied minimum requirement. When performing analysis on two groups (joint and husband dominated families) sample size was reduced even further from 69 to 54. This problem implies generalizability issues and research results should be treated cautiously. In addition, the results should be seen as showing general patterns rather than a precise source of information. Choice of answering scale could have been developed more carefully for car attributes’ questions since almost all attributes scored above “3” (moderately important) on a five point scale. Interpretation of data from these questions should be critically assessed.

5.4 Suggestions for future research Topic of this research paper could be continued with a larger and more representative sample. Some factors that could possibly explain the difference between joint and husband dominated families would be important to include in future researches. These factors could be education and relative contribution to family budget. Comparative studies where the same methods would be used across different markets could inform marketers if media and advertising strategies can be adapted to a new market without much change. During theoretical framework phase of this paper the author encountered a fact that in recent decade the studies on joint decision-making have lost some interest from researchers. It might be because the foundations have already been set and future studies will not bring any radical revelations. New research on family decision-making is needed for two reasons. One is the opportunity to compare results in the longitudinal dimension. Second reason is to keep information and theories updated because recent decades have brought in significant change regarding gender roles, where females have more equal position. More equality between genders in the society also affects family life. As we saw in section 2.3.2 there are various research approaches that aim to determine decision-making roles. These various approaches and inconsistent way of communicating results, where some researchers show the results in full detail and some show only mean results, where all stages or sub-decisions are aggregated, does not give the opportunity to compare the countries. Second point about inconsistency is that it is difficult to look at purchasing role development in one country on a longitudinal basis. It would be a good idea to standardize research approaches and unify the way of presenting findings so that the research paper would serve as an addition to the general understanding of purchase decisions in family and would not be a separate source of knowledge.

60

References Bardenheier, George, Nan McCann, Susan Blumenthal, Jennifer Chrisler, Ellie Kay, Tom Matlack, Manny Ruiz, and Kimberly Allers Seals. 2011. Marketing to the modern family. Edelman. Bartley, Sharon J., Priscilla W. Blanton, and Jennifer L. Gilliard. 2005. Husbands and wives in dualearner marriages: Decision-making, gender role attitudes, division of household labor, and equity. Marriage & Family Review 37 (4): pp. 69-94. Bateman, Ian, David Hensher, Kenneth Train, and Gerard Whelan. 2008. Demand for cars and their attributes. London: Department for Transport, . Bettman, James R., Eric J. Johnson, and John W. Payne. 1991. Consumer decision making. In Handbook of consumer behavior. Vol. 6, pp. 50-84, Prentice Hall. Bitāns, Agris, Jarkina, Viktorija, Unbedahte, Anete and Logins, Māris. Par vīrieša un sievietes nereģistrētu partnerattiecību tiesisko regulējumu eiropā un latvijā. in Ministry of Justice of the Republic of Latvia [database online]. Riga, 2008 [cited 03/30 2012]. Available from http://www.tm.gov.lv/lv/documents/petijumi/Viriesa_un_sievietes_nereg_partneratt.doc. Blood, R., Jr., and D. Wolfe. 1960. Husbands and wives. New York: The Free Press. Buss, W. Christian, and Charles M. Schaninger. 1983. THE INFLUENCE OF SEX ROLES ON FAMILY DECISION PROCESSES AND OUTCOMES. Advances in Consumer Research 10: pp. 439-444. Cliquet, R. 2003. Major trends affecting families in the new millennium: Western Europe and North America. New York: United Nations, Major Trends Affecting Families. Coleman, J. S. 1966. Foundations for a theory of collective decisions. American Journal of Sociology 71 : pp. 615-627. Commuri, Suraj, and James W. Gentry. 2005. Resource allocation in households with women as chief wage earners. Journal of Consumer Research 32 (2): pp. 185-195. Crano, William D., and Joel Aronoff. 1978. A cross-cultural study of expressive and instrumental role complementarity in the family. American Sociological Review 43 (4): pp. 463. Davis, H. L., and B. P. Rigaux. 1974. Perception of marital roles in decision processes. Journal of Consumer Research 1: pp. 51-62. Davis, H. L. 1976. Decision making within the household. Journal of Consumer Research 2 (March): pp. 241-260.

61

Davis, Harry L. 1970. Dimensions of marital roles in consumer decision making. Journal of Marketing Research 7 (2): pp. 168. Duxbury, Linda, Sean Lyons, and Christopher Higgins. 2007. Dual-income families in the new millennium: Reconceptualizing family type. 9 : pp. 472. European Automobile Manufacturers' Association. NEW VEHICLE REGISTRATIONS - BY COUNTRY. [cited 11/05 2011]. Available from http://www.acea.be/news/news_detail/new_vehicle_registrations_by_country/. Eurostat. Gender pay gap statistics. [cited 04/15 2012]. Available from http://epp.eurostat.ec.europa.eu/statistics_explained/index.php/Gender_pay_gap_statistics. ———. Live births by mother's age at last birthday and legal marital status. [cited 03/16 2012]. Available from http://epp.eurostat.ec.europa.eu/portal/page/portal/population/data/database. Ferber, R., and L. C. Lee. 1974. Husband-wife influence in family purchasing behavior. Journal of Consumer Research 1: pp. 43-50. Field, A. 2009. Discovering statistics. 3rd ed. London: Sage Publications Ltd. FITZPATRICK, MARY ANNE, and JULIE INDVIK. 1982. THE INSTRUMENTAL AND EXPRESSIVE DOMAINS OF MARITAL COMMUNICATION. Human Communication Research 8 (3): pp. 195-213. Flick, Uwe. 2010. An introduction to qualitative research. 4th ed. London: Sage Publications Ltd. Fodness, Dale. 1992. The impact of family life cycle on the vacation decision-making process. Journal of Travel Research: pp. 8-13. Ford, John B., Michael S. LaTour, and Tony L. Henthorne. 1995. Perception of marital roles in purchase decision processes: A cross-cultural study. Journal of the Academy of Marketing Science 23 : pp. 120. Goldthorpe, J. E. 1987. Family life in western societies: A historical sociology of family relationships in Britain and North America . Cambridge: Cambridge University Press. Green, Robert T., and Isabella C. M. Cunningham. 1980. Family purchasing roles in two countries. Journal of International Business Studies 11 (1): pp. 92. Green, Robert T., and Isabella C. M. Cunningham. 1980. Family purchasing roles in two countries. Journal of International Business Studies 11 (1): pp. 92-97.

62

Green, Robert T., Jean-Paul Leonardi, Jean-Louis Chandon, Isabella Cunningham, Bronis Verhage, and Alain Strazzieri. 1983. Societal development and family purchasing roles: A cross-national study. The University of Chicago Press 9 (4): pp. 436-442. Harcar, Talha, John E. Spillan, and Orsay Kucukemiroglu. 2005. A multi-national study of family decision-making. Multinational Business Review 13 (2): pp.3. Harris, R. J. 1985. A primer of multivariate statistics. 2nd ed. New York: Academic Press. Heer, David M. 1963. The measurement and bases of family power: An overview. Marriage and Family Living 25 : pp. 133-139. Henthorne, Tony L., Michael S. LaTour , and Tim W. Hudson. 1997. Japanese couples’ marital roles in stages of product purchase decision making. International Marketing Review 14 (1): pp. 39-58. Hill, Malcolm D. 1988. Class, kinship density, and conjugal role segregation, vol. 50. Journal of Marriage and Family 50 (3): pp. 731-741. Holdert, Frederique, and Gerrit Antonides. 1997. FAMILY TYPE EFFECTS ON HOUSEHOLD MEMBERS’ DECISION MAKING. Advances in Consumer Research 24: pp. 48-54. Hopper, JoAnne, Stilley. 1995. Family financial decision making: Implications for marketing strategy. Journal of Services Marketing 9 (1): pp. 24-32. Horn, Robert A. KRUSKAL-WALLIS TEST. in Northern Arizona University [database online]. 2008 [cited 04/27 2012]. Available from http://oak.ucc.nau.edu/rh232/courses/EPS625/Handouts/Nonparametric/The%20KruskalWallis%20Test.pdf. Jaris, Tichenor, Veronica. 1999. Status and income as gendered resources: The case of marital power. Journal of Marriage and Family 61 (3): pp. 638-650. Jenkins, Roger L. 1980. CONTRIBUTIONS OF THEORY TO THE STUDY OF FAMILY DECISION-MAKING. Advances in Consumer Research 7: Pages: 207-211. Keller, Gerald. 2008. Managerial statistics. 8th ed. ed. Mason: South-Western Cengage Learning. Kotler, P., K. L. Keller, M. Brady, M. Goodman, and T. Hansen. 2009. Marketing management. Essex: Pearson education limited. Latvian Authorized Automobile Dealers Association. Automašīnu klasifikators LV 2012.g. Riga, 2012 [cited 03/31 2012]. Available from http://www.lpaa.lv/files/automasinu_klasifikators/lpaa_am_klasifikators_lv_2012_29_02_2012 .pdf.

63

———. Latvijas jaunu pasažieru a/m un vieglā komerctransporta prognoze 2000.-2015. gadam. Riga, [cited 04/20 2012]. Available from http://www.lpaa.lv/lv/statistika/latvija. Lu, Yi-Chuang. 1952. Predicting roles in marriage. American Journal of Sociology 58: pp. 51. Malhotra, Naresh, and David Birks. 2007. Marketing research: An applied approach. 3rd ed. Essex: Pearson education ltd. Martínez, Eva, and Yolanda Polo. 1999. Determining factors in family purchasing behaviour: An empirical investigation. Journal of Consumer Marketing 16 (5): pp. 461-481. Menasco, Michael B., and David J. Curry. 1989. Utility and choice: An empirical study of Wife/Husband decision making. Journal of Consumer Research 16 (1): pp. 87-97. Mohan, Mark. 1995. The influence of marital roles in consumer decision-making. Irish Marketing Review (8): 97. National institute of standards and technology. KRUSKAL WALLIS. [cited 04/26 2012]. Available from http://www.itl.nist.gov/div898/software/dataplot/refman1/auxillar/kruskwal.htm. Ndubisi, Nelson, Oly, and Jenny Koo. 2006. Family structure and joint purchase decisions: Two products analysis. Management Research News 29 (1): pp. 53-64. Pallant, Julie. 2011. SPSS survival manual. 4th ed. ed. Crows Nest: Allen & Unwin. Putman, M., and W. R. Davidson. 1987. Family roles by product category, columbus, OH. In Family purchasing behavior II. Columbus, OH: Management Horizons. Road Traffic Safety Directorate. Vadītāju apliecību skaits no 2000. gada. *cited 04/08 2012+. Available from http://www.csdd.lv/lat/noderiga_informacija/statistika/vaditaja_apliecibas/. Rodman, Hyman. 1972. Marital power and the theory of resources in cross-cultural context. Journal of Comparative Family Studies 1: pp. 50-67. Safilios-Rothchild, C. 1969. Family sociology or wives’ family sociology? A cross-cultural examination of decision making. Journal of Marriage and the Family 29 (May): pp. 290-301. Scanzioni, J. H. 1972. Sexual bargaining: Power politics in american marrriage. Englewood Cliffs, NJ: Prentice-Hall. Sheth, J. N. 1974. Models of buyer behavior. New York: Harper & Row. Slama, Eand, and A. Taschian. 1985. Selected socio-economic and demographic characteristics associated with purchasing involvement. Journal of Marketing 49: pp. 72-80.

64

Su, Chenting, Edward F. Fern, and Keying Ye. 2003. A temporal dynamic model of spousal family purchase-decision behavior. Journal of Marketing Research 40 (3): pp. 268-281. Tabachnick, B. G., and L. S. Fidell. 2007. Using multivariate statistics. 5th ed. ed. Boston: Pearson Education. United States department of labor. Changes in men’s and women’s labor force participation rates. Washington, January 10, 2007 [cited 04/21 2012]. Available from http://www.bls.gov/opub/ted/2007/jan/wk2/art03.htm. VanVoorhis, Carmen R., Wilson, and Betsy L. Morgan. 2007. Understanding power and rules of thumb for determining sample sizes Tutorials in Quantitative Methods for Psychology 3 (2): pp. 43-50. Webster, Cynthia. 1994. Effects of Hispanic ethnic identification on marital roles in the purchase decision process. Journal of Consumer Research 21 (2): pp. 319-331. Wolgast, Elizabeth H. 1958. Do husbands or wives make the purchasing decisions? The Journal of Marketing 23 (2): pp. 151-158. Yang, Xia, U. Ahmed Zafar, Ghingold Morry, and Hwa Ng Kuan. 2006. Spousal influence in singaporean family purchase decision-making process. Asia Pacific Journal of Marketing and Logistics 18 (3): pp. 201-222. Yavas, U., E. Babakus, and N. Delener. 1994. Family purchasing roles in Saudi Arabia: Perspective from Saudi wives. Journal of Business Research 31: pp. 75-86. Zickel, Raymond E. 1989. Soviet Union; A country study. Federal Research Division.

Appendix Appendix 1: Consumer survey Hello! My name is Viesturs Uzuleņš and I am currently writing bachelor thesis at Aarhus University. I would be grateful if you filled out this survey regarding family car purchase. The survey is short and will take only 5 minutes to complete. This survey is intended to be filled out by only one of the spouses of a family that has bought a family car in recent years. Aim of this research is to discover how joint decision-making happens when family buys a car. All data gathered will be anonymous and used only in a summarized state. Thank you for your help! V_1 Which automobile class your car falls under? * examples of cars that fall under this class are given in brackets

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)

            

Mini (Citroen C1, Fiat 500) Small (Toyota Yaris, Peugeot 207) Compact (VW Golf, Ford Focus) Middle (Toyota Avensis, Audi A4) Large (BMW 5 Series, Audi A7) Highest (Mercedes-Benz S class, BMW 7 series) Small SUV (Nissan Juke, Mini Countryman) SUV (Toyota RAV4, BMW X3) Large SUV (Mitsubishi Pajero, Land Rover Discovery) Coupe, convertible or roadster (Audi TT, Volkswagen Scirocco) Small multi-purpose (Opel Zafira, Toyota Verso) Multi-purpose (Ford S-Max, Volkswagen Sharan) Large multi-purpose (Ford transit, Peugeot Boxer)

V_2 How often do you use the car, compared to total time your family uses it? * state percentage of total time

__________ V_3 Did you buy new or used car? (1)  New

(2)



Used

V_4 How old was the car when you bought it? * if you bought a new car, type "0"

__________

V_5 How many years have passed since your car purchase? __________ V_6 How much did you spend on the car? __________

V_7 How many cars does your family own? __________

V_8 Please give an answer to questions below by marking the most suitable option. Female predominantly

Both

Male predominantly

Male

V_8_1 Who was the one who first started the discussion about new (1)  car purchase?

(2) 

(3) 

(4) 

(5) 

V_8_2 Who searched for information regarding cars, their characteristics prices and other important matters?

(1) 

(2) 

(3) 

(4) 

(5) 

V_8_3 Who made the final decision (1)  about car purchase?

(2) 

(3) 

(4) 

(5) 

V_8_4 Who decided upon when is (1)  the right time to buy the car?

(2) 

(3) 

(4) 

(5) 

V_8_5 Who of you chose the car dealer where to buy the car?

(1) 

(2) 

(3) 

(4) 

(5) 

V_8_6 Who decided how much you (1)  will spend on the car?

(2) 

(3) 

(4) 

(5) 

V_8_7 Who decided which brand and model to buy?

(2) 

(3) 

(4) 

(5) 

Female

(1) 

V_9 Please mark who in your family decided upon these attributes of the car. Female

Female predominantly

Both

Male predominantly

Male

Did not discuss this attribute

V_9_1 Gearbox type

(1) 

(2) 

(3) 

(4) 

(5) 

(0) 

V_9_2 Equipment (cruise control, audio system etc.)

(1) 

(2) 

(3) 

(4) 

(5) 

(0) 

Female

Female predominantly

Both

Male predominantly

Male

Did not discuss this attribute

V_9_3 Space in the car / Boot capacity

(1) 

(2) 

(3) 

(4) 

(5) 

(0) 

V_9_4 Fuel type / Fuel economy

(1) 

(2) 

(3) 

(4) 

(5) 

(0) 

V_9_5 Safety equipment (airbags, (1)  ABS)

(2) 

(3) 

(4) 

(5) 

(0) 

V_9_6 Engine size / Power

(1) 

(2) 

(3) 

(4) 

(5) 

(0) 

V_9_7 Body type / Styling

(1) 

(2) 

(3) 

(4) 

(5) 

(0) 

V_9_8 Exploitation cost (insurance, (1)  upkeep etc.)

(2) 

(3) 

(4) 

(5) 

(0) 

V_10 Please rate how important are these car aspects to you.

1 Not important

2

3 Moderately 4 important

5 Important

V_10_1 Fuel type / Fuel economy

(1) 

(2) 

(3) 

(4) 

(5) 

V_10_2 Space in the car / Boot capacity

(1) 

(2) 

(3) 

(4) 

(5) 

V_10_3 Gearbox type

(1) 

(2) 

(3) 

(4) 

(5) 

V_10_4 Engine size / Power

(1) 

(2) 

(3) 

(4) 

(5) 

V_10_5 Safety equipment (airbags, (1)  ABS)

(2) 

(3) 

(4) 

(5) 

V_10_6 Exploitation cost (insurance, upkeep)

(1) 

(2) 

(3) 

(4) 

(5) 

V_10_7 Body type / Styling

(1) 

(2) 

(3) 

(4) 

(5) 

V_10_8 Equipment (air conditioning, audio system etc.)

(1) 

(2) 

(3) 

(4) 

(5) 

V_10_9 Recognition of the brand

(1) 

(2) 

(3) 

(4) 

(5) 

V_10_10 Warranty

(1) 

(2) 

(3) 

(4) 

(5) 

Personal information V_11 What is your gender? (1)  Female

(2)



Male

V_12 How old are you? __________ V_13 What is monthly after tax income of your family? (1)  Less than 300 Ls

(2) (3) (4) (5)

   

300 - 600 Ls

(2)



No

600 - 900 Ls 900 - 1200 Ls

More than 1200 Ls V_14 Are you married? (1)  Yes

V_15 Do you have children who still live in your household? (1)  Yes

(2)



No

V_16 Comments on things regarding family car purchase that weren't discussed in previous questions. ________________________________________________________________________________ ________________________________________________________________________________ Thank you for your participation! Click "X" to finalize the survey. Please encourage also other couples you know to participate in this survey.

Appendix 2: Assumptions about similar distributions Comparison of 3 methods

Comparison of 3 decision-making stages

Comparison of 4 sub-decisions

For the sake of not using too much space of distribution comparisons, 8 sub-decision distributions will not be presented. But they produce rather similar distributions which is sufficient to be able to perform the Kruskal-Wallis test.

Comparison of “joint family” and “husband dominated family” groups

Appendix 3: Frequencies of reported decision-making roles Patterns of influence Did not discuss Who initiated the car purchase? Who searched for information? Who made the final decision?

Wife 12% 0% 4%

Wife predominantly 7% 3% 3%

Both 28% 20% 32%

Who decided: When to buy a car? Where to buy a car? How much to pay for a car? What make of car to buy? Gearbox type Equipment Space/ Boot capacity Fuel type/economy Safety equipment

3% 3% 3% 3% 10% 3% 3% 4% 3%

4% 3% 3% 7% 4% 3% 3% 3% 3%

42% 20% 49% 33% 25% 23% 41% 23% 16%

12% 23% 12% 3% 30%

Husband predominantly Husband 26% 28% 20% 57% 20% 39%

23% 26% 14% 17% 13% 13% 9% 20% 9%

26% 48% 30% 39% 36% 35% 33% 46% 39%

Engine size / Power Body type / Styling Exploitation cost Note: N=69

6% 3% 10%

3% 6% 3%

1% 6% 3%

19% 52% 28%

Appendix 4: Cars by class and presence of kids in family

Percentage of cars in each class 60%

48%

50%

39%

40% 30%

15% 14%

20% 6%3%

10%

25%

21%

12%

11% 3%0%

0%

3%0%

With kids No kids

Appendix 5: Income and purchase of a new or used car New 100.0%

100.0%

Used

91.7% 76.5% 60.0% 40.0% 23.5%

0.0% Less than 300 Ls (N=3)

0.0% 300 - 600 Ls (N=22)

8.3%

600 - 900 Ls (N=12)

900 - 1200 Ls More than 1200 (N=17) Ls (N=15)

12% 10% 16%

59% 23% 41%