This article was downloaded by: [Northern Kentucky University], [Ryan Salzman] On: 05 September 2013, At: 06:19 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Mass Communication and Society Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/hmcs20
News or Noticias: A Social Identity Approach to Understanding Latinos’ Preferred Language for News Consumption in the United States Ryan Salzman
a
a
Department of Political Science, Criminal Justice, and Organizational Leadership , Northern Kentucky University Published online: 28 Aug 2013.
To cite this article: Mass Communication and Society (2013): News or Noticias: A Social Identity Approach to Understanding Latinos’ Preferred Language for News Consumption in the United States, Mass Communication and Society, DOI: 10.1080/15205436.2013.782048 To link to this article: http://dx.doi.org/10.1080/15205436.2013.782048
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions
Mass Communication and Society, 0:1–20, 2013
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
Copyright # Mass Communication & Society Division of the Association for Education in Journalism and Mass Communication ISSN: 1520-5436 print=1532-7825 online DOI: 10.1080/15205436.2013.782048
News or Noticias: A Social Identity Approach to Understanding Latinos’ Preferred Language for News Consumption in the United States Ryan Salzman Department of Political Science, Criminal Justice, and Organizational Leadership Northern Kentucky University
For Latinos in the United States, decisions reflecting which language they prefer for news media consumption can have substantial effects on the quality of the content they receive. However, little is known about what influences Latinos’ language preferences for news media consumption. This project looks to fill that void to the immediate appeal of academics and media analysts by asking, What influences Latinos’ language preference for news? This project builds on social psychology theories highlighting the role played by social identity and self-categorization for inspiring culturally consistent behaviors. This social identity approach focuses on the embrace of Latino characteristics and the complementary nature of Spanish-language media outputs relative to those characteristics. Using the 2006 Latino National Survey, the role of identity is explored. The results indicate that language preferences for news consumption are indeed related to various social identity measures among Latinos in the United States.
Ryan Salzman (Ph.D., University of North Texas, 2011) is an Assistant Professor in the Department of Political Science, Criminal Justice, and Organizational Leadership at Northern Kentucky University. His research interests include political communication, Latino attitudes and behaviors, and Latin American political culture. Correspondence should be addressed to Ryan Salzman, Department of Political Science, Criminal Justice, and Organizational Leadership, Northern Kentucky University, Highland Heights, KY 41099. E-mail:
[email protected]
1
2
SALZMAN
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
INTRODUCTION News media productions today are as plentiful as they are diverse in content coverage and quality of reporting. That diversity makes it difficult to study the effects of news media consumption while treating all news media as a single type. However, some news media share commonalities that place them in a similar class and thus make them appropriate to cluster together in academic studies. For one group of Americans, that simple classification can be delineated by language preference for news consumption. Latinos1 in the United States choose to watch, read, or listen to news in either Spanish or English. That language-based decision can influence the quantity and quality of information received because English-language and Spanishlanguage news producers set different agendas and employ different frames for similar stories (Hale, Olson, & Fowler, 2009). That is, media outputs may be systematically related to production language. Before other researchers endeavor to explain how news produced in different languages affects individuals differently, this project first tries to answer the question, What influences Latinos’ language preference for news consumption? This article highlights social identity as a key to revealing what affects Latinos’ language preference for news media consumption. The statistical results indicate that identity is centrally important in shaping language preferences for news consumption. This research employs overt identity measures taken from survey responses along with proxy characteristics that are justifiably related to identity. Numerous additional measures that could shape language preference are also included to ensure that the model is properly specified. The results indicate a consistent relationship between news language preferences and social identity with other characteristics related in ways that may or may not be intrinsically connected to identity themselves. In general, literature addressing what determines media use is growing in all contexts as the availability of media products continues to expand. As a result of this expansion, media researchers concerned with understanding consumer behavior have tailored their research questions to embrace the substantial variation across media (i.e., computers vs. TV), products (i.e., news vs. entertainment), and consumer groups (i.e., ethnics vs. nonethnics). One consistent finding among these various, sometimes disparate pieces of research is that understanding media use requires understanding individual-level
1 The term ‘‘Latino’’ was chosen over the more classic ‘‘Hispanic’’ to denote individuals in the United States of Latin American ancestry. To be clear, both terms represent the same ethnic classification for individuals in the United States.
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
LANGUAGE PREFERENCE OF LATINOS IN THE U.S.
3
attributes of users.2 For Latinos in the United States, identity and ethnicity is paramount to understanding media consumption behavior. A significant body of research on ethnic identification and general media use serves as a foundation on which this research builds. A survey of that research reveals that identity and acculturation influence individuals’ decisions to consume ethnic media (see, e.g., Jeffres, 2000; Rios & Gaines, 1998; Subervi-Valez, 1986). That research is indispensable for understanding how ethnicity influences decisions related to media consumption. This project adds to that body of knowledge in three ways. First, instead of looking at media in general, this study focuses explicitly on news media consumption. As different media products inspire varied media consumption behavior, focusing on one type of media product is important for testing the ethnic identity–media consumption relationship. Although research on Latinos and media use in the United States is prevalent, there is a lack of focus on news media consumption. Second, this study employs numerous measures of identity. Some of the included measures are overtly relatable to ethnic identity (i.e., self-identification), whereas other measures are viewed as reliable proxies for identity (i.e., language skills). The use of many different measures of identity is a substantial advancement over prior research. The third important feature of this research relates to changes in the news media industry. Relative to research conducted in the 1970 s, ’80 s, and ’90 s, the 21st century has seen a proliferation of Spanish-language news media products. Thus, research focusing on Latinos in the United States need worry less about lack of access to ethnic media outlets as a behavior-influencing factor. This systematic change requires retesting of what has already been learned. The advancements of this study are best understood as refining and updating previous research. Uncovering the differential effects of these factors may prove useful to a variety of individuals because news produced in different languages may vary in content and quality of information received by individuals. That variation may affect numerous attitudes and behaviors. Communication scholars will take a particular interest in the results of this research, as it increases the importance of social identity for understanding media consumption behavior. As Spanish-language media products grow, identity will likely be essential to understanding individual media consumption behavior. This study not 2
For example, Roberts and Foehr (2008) examined trends in media use among children in the United States. They found clear differences in use that appear dependent on the presence of media (i.e., computer), the child’s race and variation in intellect measures (i.e., grades) among other attributes that are individual specific. Trevino, Webster, and Stein (2000) considered what affects communication medium choice in business settings. They found individual-level determinants, such as proximity to other individuals, variably influence media use choices by managers.
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
4
SALZMAN
only focuses on cultural identity but also includes other potentially influencing factors that may trump identity (i.e., socioeconomic status), thus giving this study comparative weight that has not existed to this point for communication scholars. Likewise, academics and policymakers who focus on the determinants of Latino behaviors beyond media consumption will find this research to be of great import. Again, this project illuminates the significance of identity relative to other individual-level factors in shaping behavior. Thus, this project may serve as a jumping off point for researchers and policymakers trying to understand behaviors beyond news media consumption. Finally, because the crux of this project centers on understanding how the substantial population of Latinos in the United States chooses their news media outlets, media marketing analysts would do well to take note. Essentially, this project endeavors to understand what kind of Latino prefers English- versus Spanish-language news. That understanding can inform programming selection and advertising sales. Those marketing decisions should be reflective of the actual consumers of a given media outlet, and that is what this research is focused on identifying. THEORY This research project endeavors to understand what influences Latinos’ language preference when consuming news. In the past, language choice was often determined first and foremost by the availability of Spanish-language programming (see Subervi-Valez, 2008). Today, Latinos in the United States find Spanish and English-language news readily available for consumption (Coffey, 2009). As availability grew, the relevant factors for choosing Spanish-language or English-language news became consistently more consumer-specific and less location related.3 Research indicates that as Spanish-language options increased, so too did the tendency of Spanish speakers to prefer Spanish-language programming (DeSipio, 2003). That is true not just of Spanish-only speakers but also of bilingual consumers. A recent poll conducted jointly by the Associated Press and Univision (2010) echoes that sentiment. According to the poll results, Latinos prefer consuming Spanish-language news for socio-cultural reasons. Understanding how culture fits into language preferences for news consumption requires an acceptance that culture matters for many behaviors, especially when those behaviors further foster cultural characteristics. According to social psychology researchers, understanding individual 3
See Subervi-Velez (2008) Chapter 15 for a discussion of the penetration of Spanishlanguage news media prior to 20 years ago.
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
LANGUAGE PREFERENCE OF LATINOS IN THE U.S.
5
behaviors has moved from a strictly individualistic approach to an interactionist approach that values concepts like group orientations for informing social processes (see Postmes & Branscombe, 2010). This emphasis on social identity highlights the inseparability of individuals from society as ‘‘there is a continuous reciprocal interaction and functional interdependence between the psychological processes of individuals and their activity, relations, and products as society’’ (Turner & Oakes, 1986, p. 239). In short, cognition is mediated by society. Thus, the social identity approach includes group memberships as a relevant factor that informs social processes. Further clarifying the social identity approach, social psychology researchers highlight the role played by self-categorization for shaping social processes (see Haslam, 2001). Self-categorization theory explains those processes as consistent among groups whose individual members share similar social self-concepts. Those self-concepts are intrinsically related to comparisons that highlight differences with other groups, the result being one in which social identity experiences increased salience relative to personal identity (Turner, 1984). Although there are many manifestations of selfcategorization, race is one of the simplest and most overt. This constitutes a meta-contrast where individuals are either in the group or out of the group (Turner & Oakes, 1986). Therefore, shared self-categorizations based on racial, ethnic, or cultural traits can condition behavior consistently across the group. Given that ‘‘Latino’’ is a self-categorization, consistent behaviors should be expected. That is, individuals’ perceived social identities factor into their attitudes and behaviors (Barreto & Mun˜oz, 2003) to the extent that the attitude or behavior exemplifies an in-group norm (Turner, 1999). As Spanish-language news products are overtly geared toward one ethnic class, language preference choices are a direct reflection of ones self-categorization, or social identity. Spanish-language news has an obvious connection to the Latino community that transcends language. Spanish-language news focuses on distinctly Latino issues and at times goes beyond information provision to advocate for Latino positions (Hale et al., 2009). Thus, accounting for access to media and other constraints, Latinos who embrace characteristics central to their culture and heritage should prefer news media outlets that foster that self-identification. It is worth noting that embrace of cultural identity can be recognized by the individual, but it can also be reflected in other individual attributes such as language skills and generation of residence. For minority groups in any society, self-categorization as a member of that group is often quite pronounced. That strength of identity can depend on the desire of the individual to maintain their immigrant culture versus assimilating into the American culture. Maintaining that connection to a family’s home country is well documented and seen as the prevailing factor
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
6
SALZMAN
shaping Latino attitudes and behaviors in the United States (Barreto & Mun˜oz, 2003; Basch, Schiller, & Blanc, 1994; Jones-Correa, 1998; SuberviVelez, 2008). This trend is even born out in academic vernacular as ‘‘assimilation,’’ discarding ones culture for another, has been replaced with ‘‘acculturation,’’ integrating a new culture into one’s existing culture.4 ‘‘Even in the process of achieving the ‘American Dream,’ Latino identity is valued and serves more than just as an occasional nominal classification. . . . Thus, they pay special attention to the news that addresses such matters’’ (Subervi-Valez, 2008, p. 54). That is, Spanish-language products promote that connection to Latino identity well beyond any English-language product. Therefore, Latinos who desire to keep Spanish viable as a language in their personal life will derive more gratification from consuming Spanish-language news media. Likewise, individuals who view ‘‘Latino’’ as being distinct from other minority groups will prefer Spanish-language news. For those who perceive speaking English as integral to being American, a greater preference for English news media is expected. H1: Latinos in the United States with stronger Latino identity will prefer Spanish-language news more than those with a weaker Latino identity.
Language abilities clearly affect preferences for news language. An individual who is fluent in Spanish may not know English to the extent that choosing English-language news is a realistic option. However, an examination of the survey data employed in this project reveals that bilingualism is prevalent among Latinos in the United States. Of respondents identified as fluent Spanish speakers, 36% are also fluent in English. That number swells to nearly 50% when individuals who speak English well are included. Among survey respondents who self-identify as fluent English speakers, more than 65% are fluent in Spanish. In addition to being a determinant skill for shaping language preferences for news consumption, language is central to one’s cultural identity. Immigrants and children of immigrants who have stronger language skills in the new country are often better assimilated (Alba & Nee, 1997). Greater acculturation shifts self-categorization to be more in line with Englishspeaking Americans. Thus, although language skills cannot be the standalone determinant that they have been in the past, it is important to recognize that language abilities reflect characteristics of social identity that influence language preference for news consumption. An inability to speak English undermines acculturation and reinforces Latino identity. Likewise, strong English skills capture acculturation effects that have been ongoing and will 4
See Subervi-Valez (2008) page 52 for a brief discussion of assimilation and acculturation.
LANGUAGE PREFERENCE OF LATINOS IN THE U.S.
7
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
continue to reinforce identities more in line with other English-speakers. Thus, language can be considered an identity cue. H2: Latinos in the United States who are able to speak English well will prefer less Spanish-language news, whereas those who do not will prefer more Spanish-language news. H3: Latinos in the United States who are able to speak English well will prefer more English-language news, whereas those who do not will prefer less English-language news.
Another characteristic that is intrinsically related to acculturation and social identity is how removed one is from immigration. Each new generation of Latinos living in the United States should prefer English news more than the previous generation as their identities are more similar to the American public than any specific ethnic group. This is likely because each generation will be more assimilated than the previous generation (Gans, 1992; Lieberson, 1973). That is, when an immigrant Latino has a child, they will be brought up in American society, including having American friends, watching American TV, and participating in the American education system. Even if the child is surrounded by other Latinos, they will be more inclined to drift toward mainstream American society than their parents. That social identity will be reflected in the preferred language of the news for that generation. H4: Latinos in the United States who are further removed from immigrant status will prefer English-language news more than those who are closer to the immigrant generation.
METHOD To test the hypotheses related to Latino news media consumption in the United States, this project employs survey data taken from the 2006 Latino National Survey (LNS). The LNS is a national telephone survey of 8600 Latino residents of the United States that seeks a broad understanding of the qualitative nature of Latino political and social life. Survey length was approximately 40 minutes. The exact length and number of questions per respondent depended on the citizenship, parental status, and state of residence. Interviewing Services of America conducted the surveys in the preferred language of the respondent. Surveys were administered in English, Spanish, or both languages (Fraga et al., 2006). The universe for the survey was all adult Latinos 18 years of age or older. They did not have to be U.S. citizens or voters. The geographic focus of the study was first on states with large Hispanic populations. Additional states
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
8
SALZMAN
with emerging Hispanic populations were also included to further capture Hispanics in the United States. The sample represents approximately 87.5% of the Latino population in the United States.5 The unaccounted Latino populations lie in states not included in the LNS sample. The dependent variables in the analyses identify the preferred language of news consumed. These variables are single item measures obtained from a single question in the LNS survey then coded to measure news language preference as three separate dependent variables. The survey question asks whether the respondent relies more heavily on Spanish-language or Englishlanguage television, radio, and newspapers when seeking information regarding public affairs and politics. The three responses in this question (English more, Spanish more, and both equally) were broken into three separate dichotomous dependent variables with yes=no responses. To this point, there has been no discussion of preferring both languages for news consumption as a viable option. No prior research has addressed the possibility of preferring both languages for news consumption. The lack of attention to what would motivate preferring both languages renders hypothesizing nearly impossible. However, including this additional measure in the model may provide increased illumination of the research question. Thus, the results of the tests with the dependent variable ‘‘Prefer Both News’’ is discussed as it relates to the other two language preference variables. To conduct the analysis of the dependent variables, a standard logistic model is employed. This model was chosen because the language preference dependent variables are dichotomous (Long & Freese, 2006). The results of the statistical models will reveal the direction and the statistical significance of the effect of the independent variables on the dependent variables of interest. This project employs multiple primary independent variables and numerous control variables. The first set of independent variables overtly captures self-categorization. Latino identity (H1) measures the importance of maintaining a distinct Latino culture. This is coded from 0 (not at all) to 1 (important) to 2 (very important). Keep Spanish and learn English to be American are both divided into four levels of importance—1 (not at all), 2 (not very important), 3 (somewhat important), and 4 (very important). The remaining independent variables are considered reliable proxies for identification. Language ability (H2–H3) is captured with two variables. English quality measures the level of English speaking ability, whereas Spanish quality measures the level of Spanish-speaking ability. The measures are scored from 0 (not at all) to 3 (very well). All respondents who opted to
5
Additional information about the LNS can be found at http://www.icpsr.umich.edu/ icpsrweb/ICPSR/studies/20862/detail.
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
LANGUAGE PREFERENCE OF LATINOS IN THE U.S.
9
take the survey in English were given a score of 3 for English quality, whereas respondents who opted for Spanish were given a 3 for the Spanish quality variable. The generational variables (H4) are each single-item dichotomous responses all coded with 0 equaling no and 1 equaling yes. A secondgeneration respondent is one who answered that neither parent was born in the United States. A third-generation respondent is one that identified at least one of their grandparents was born outside of the United states, and fourthgeneration respondents are those that all parents and grandparents were born in the United States. To ensure that the model is properly specified, a number of potential constraints must be included as control variables. The extensive list of controls illustrates the numerous factors that could systematically influence language preference for news media consumption among Latinos. Along with the variable descriptions, a brief theoretical justification is included. In addition, each control factor is discussed along with the primary variable of interest, identity. Salzman (2011) highlighted a 3-point categorization of factors that influence decisions to consume various media types. Those categories describe individuals’ abilities, interests, and expectations toward news media.6 To understand what determines Latinos’ language preference for news media consumption in the United States, those same categories are employed. Although each category motivates language preference for news consumption differently than decisions about whether to consume news media of various kinds, the classifications remain useful for the reasons just stated. Each category is discussed in turn. Immigrant assimilation researchers have demonstrated that immigrants and their progeny of higher socioeconomic classes assimilate to the new society and culture better than those of lower socioeconomic status (Alba & Nee, 1997). Likewise, Latinos with greater resources should be expected to integrate into American society better than those with fewer resources. Also, that integration occurs quicker, thus integrating them more thoroughly over time. Latinos with greater economic success have spent more time and energy in society at large as opposed to more homogenous Latino communities in the United States. To test the respondent’s abilities, the following variables are utilized: employed, own home, and income. Employed is scored 0 to 2 dividing employment into part-time, full-time, and not employed. Own home is a dichotomous variable as well with a 0 given to those who do not own their home and a 1 to homeowners. Income is measured as the total household income
6
The 2011 study by Salzman focuses on individuals living in Latin America.
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
10
SALZMAN
reported divided into seven income ranges from 1 (below $15,000) to 7 (above $65,000). Interest can be an important motivator for media consumption behavior (Norris, 2000). Interest in the United States political system can motivate news consumption in all languages. As that system of interest is distinct, the best way to learn about the entire system, and not just parts that are of interest to certain ethnic groups, is to use news media outlets with the broadest appeal. Hale et al. (2009) demonstrated that when covering elections, Spanish-language news provides more information that may be directly pertinent to Latinos, but English-language news can satisfy individuals’ desire for more general election information. Therefore, variation in interest in the political system may likewise variably affect language preference for news consumption. Interest employs 10 separate variables to test various levels of interest and political actions. Interest in politics tests whether the respondent is not interested (0), somewhat interested (1), or very interested (2) in politics and public affairs. Political knowledge is an additive index of various items intended to capture both formal and informal information. The questions asked the respondent to identify which party controls Congress, who won their state’s electoral votes in 2004, and which party is more conservative. For each correct response, the respondent receives a score of 1. Incorrect responses receive a score of 0. The scores are then added together to create a political knowledge index that ranges from 0 to 3. Education measures the degree of formal education that a person received and are assigned scores between 0 (none) and 7 (graduate or professional degree). Civic engagement and contact government measure the frequency of actions taken to participate in civic group activities and contact a government official. Both variables are divided into three levels of frequency: no actions (0), one action (1), and two or more actions (2). Vote is a dichotomous variable measuring if a respondent has voted either within the United States or in their country of origin. Registered to vote is scored positively (1) only if the person is registered within the United States. If the respondent is not registered to vote in the United States, he or she is assigned a coded value of 0. Democrat and Republican are variables used to identify if people considers themselves a Democrat or Republican. These variables are obtained from a single question within the LNS survey that asks respondents if they self-identify as a Democrat, Republican, Independent, or other. Those respondents who answered ‘‘Democrat’’ were given a value of 1; all other responses were assigned a 0 value for the Democrat variable. The same method was used for the Republican variable assigning those who answered ‘‘Republican’’ a value of 1 and all other responses a value of 0. Finally, citizenship evaluates whether a respondent is a citizen of the United States. This variable is coded dichotomously with 1 being yes and 0 being no.
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
LANGUAGE PREFERENCE OF LATINOS IN THE U.S.
11
As ‘‘media use is selective and motivated by rational self-awareness’’ (Ruggiero, 2000, p. 18), the usefulness of a medium can be influenced by an individual’s expectations (Fry & McCain, 1983). Experiencing different news media outputs can variably shape an individual’s expectations about news media in their life. Thus, the press conditions in the country of origin may affect news language preference. This derives from the recognition that context matters for influencing individual level attitudes (Almond & Verba, 1963). For Latinos from countries with greater freedom of the press, Spanish-language news will be seen as legitimate, and thus preferred when in the United States. For Latinos from countries with less press freedom, Spanish-language news may be naturally viewed as unreliable, subsequently leading to a preference for English-language news. The system-level variable press freedom is taken from Freedom House’s 2004 press freedom index. Countries of origin were assigned a potential value ranging from 0 to 100. The actual values assigned were 4 through 87, with 4 being the country of origin with the lowest level of press freedom and 87 being the county of origin with the ‘‘most free’’ press. Table 1 illustrates the differences in press freedom scores between the different Latin American countries and the United States in 2004, the year prior to the LNS survey being administered. Some additional basic control variables that have been consistently linked to various individual behaviors (see Almond & Verba, 1963) are included to ensure that the correct relationships are identified as well as to increase the robustness of the models. Although seeking news media is not explicitly a political behavior like those identified by Almond and Verba (1963), individual differences may still variably affect that behavior. These include male, age, married, and kids. Gender is important to account for in any model looking at behavior, as men and women behave differently in ways that systematically reflect their gender. For instance, female Latinos may spend the majority of their day in the house, making them more able to watch TV generally and the news specifically. Male measures the sex of the respondent and is coded as either yes (1) or no (0). Older individuals are likely to consume more news than younger individuals for a variety of reasons. Older individuals may find themselves with more free time relative to younger individuals with families. Also, older individuals may be entrenched in habitual news consumption, whereas that habit is not established in younger people. Finally, older individuals may have greater interest in the information covered in news productions. Age is assigned a continuous value in years. The range of ages in this study is from 18 to 97. Individuals may marry non-Latinos, thus systematically affecting which language they prefer for news consumption. Married identifies a respondent’s marital status where ‘‘married’’ is assigned a value of 1 and ‘‘not married’’ is assigned a value
12
SALZMAN
TABLE 1 Press Freedom Scores for Latin America Freedom House Freedom of the Press 2004
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
Country Argentina Bolivia Brazil Chile Colombia Costa Rica Cuba Dominican Rep. Ecuador El Salvador Guatemala Honduras Mexico Nicaragua Panama Paraguay Peru Puerto Ricoa United States Uruguay Venezuela
Score
Rating
35 37 36 23 63 19 96 39 42 42 62 52 36 37 45 54 34 – 13 26 68
Partly Free Partly Free Partly Free Free Not Free Free Not Free Partly Free Partly Free Partly Free Not Free Partly Free Partly Free Partly Free Partly Free Partly Free Partly Free — Free Free Not Free
Note. High scores ¼ less press freedom. Source: Freedom House 2004. a Puerto Rico received no score as it is considered the same as the United States.
of 0. Like marriage, the introduction of children to the family could influence news gathering behavior for various reasons. First, parenting issues may be better addressed in local news outlets that are less available in Spanish. Second, as children age they may desire to consume news in English instead of Spanish which could change the preferences of the parent. Kids measures the number of children within a household from 0 (none) to 4 (four or more). In essence, failure to include control variables can cause a misidentification of what is actually motivating behaviors. To be clear, all possible controls are not included in this model. RESULTS Table 2 presents the results of the statistical analyses examining the determinants of news language preference for Latinos in the United States. The
LANGUAGE PREFERENCE OF LATINOS IN THE U.S.
13
TABLE 2 Standard Logistic Model of News Language Preference
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
Spanish news
Identity Keep Spanish Learn English to be American Latino is distinct Spanish quality English quality 2nd generation 3rd generation 4th generation Controls Employed Own home Income Interested in politics Political knowledge Education Civic engagement Contact government Vote Registered to vote Democrat Republican Citizen Press freedom (birth country) Male Age Married Kids? Constant Pseudo-R2 N
English news
Both news
z Score
sig.
z Score
sig.
z Score
sig.
3.02 1.20 1.96 4.30 20.78 3.43 4.80 2.98
.003 .231 .050 .000 .000 .000 .000 .003
6.86 2.63 3.88 9.71 15.65 4.07 5.54 5.12
.000 .009 .000 .000 .000 .000 .000 .000
3.91 1.52 2.31 9.44 13.39 3.31 4.02 4.39
.000 .127 .021 .000 .000 .001 .000 .000
0.00 0.34 5.20 0.02 3.96 4.98 1.89 1.74 0.80 0.16 0.79 0.14 2.76 2.89 3.38 2.68 0.08 0.76 2.03
.997 .733 .000 .980 .000 .000 .059 .081 .425 .873 .430 .888 .006 .004 .001 .007 .937 .444 .042
0.64 0.18 6.25 0.27 6.62 4.58 1.65 0.16 1.19 1.61 0.67 1.65 1.69 2.26 1.37 4.05 0.68 1.40 3.42
.521 .860 .000 .785 .000 .000 .099 .874 .234 .108 .506 .099 .091 .024 .172 .000 .498 .161 .001
0.45 0.15 3.10 0.15 1.61 0.53 1.30 1.32 0.70 1.91 0.03 1.22 3.46 0.04 2.27 1.37 0.75 0.72 9.33
.655 .878 .002 .883 .108 .595 .193 .185 .485 .056 .980 .223 .001 .965 .023 .172 .453 .471 .000
0.407 5,941
0.445 5,941
0.076 5,941
Note. Source: 2006 Latino National Survey.
models performed very well with high pseudo-R2 values for the Spanish news preference model (0.407) and the English news preference model (0.445). The both news model also performed well with a pseudo-R2 value of 0.076. The models include variables that capture respondents’ identity as well as numerous control variables. The results indicate consistent support for the hypotheses with each promoting a pronounced role for social identity in shaping language preference for news consumption among Latinos.
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
14
SALZMAN
Latino identity, both self-identified and proxy measures, is an important determinant of news language preference. Latinos who assert that using Spanish in the household is important prefer Spanish news more and English news less. Holding the view that keeping Spanish is important to the Latino identity is also positively correlated with preferring news in both languages. This reinforces the identity-based assertions made by DeSipio (2003) where even as English becomes easier and there are more English options for media products, Latinos embrace their heritage by consuming Spanish-language media. Latinos who endorse the idea that learning English is an important aspect of being American are significantly more likely to consume English news. Although the relationship is expected to be negative and significant for the Spanish-language preference because Latinos in the United States who see learning English as important to being American might embrace Englishlanguage news consumption and reject Spanish-language news, no such correlation is evident. There is also no effect for consuming news in both languages. Latinos who perceive ‘‘Latino’’ as a distinct ethnic group follow expectations. Those who endorse the idea of Latino-ness are significantly more likely to prefer Spanish-language news and less likely to prefer Englishlanguage news. The finding for preferring both languages is positive and significant. Like a desire to keep Spanish in the household, perceiving Latino as a distinct identity may be embraced by later generations and thus lead to preferring news in both languages. Substantial support is found for the role of self-reported identity in shaping news language preferences for Latinos in the United States providing clear support for H1. One alternative concept that reasonably captures Latino identity is language skills. Individuals who speak Spanish well maintain a closer connection to their Latino identity and will thus consume more Spanishlanguage news. The opposite should be true for individuals who speak English well. That is, individuals who speak English well have identities that are more similar to Americans of all races and subsequently employ Englishlanguage news for information gathering. The results of the analysis reinforce these suppositions. Latinos who speak Spanish well are more likely to prefer Spanish-language news. The relationship is negative for preferring English news. Including a variable capturing the quality of English per respondent finds significant results that run opposite those for Spanish quality. It is important to point out that those are not mutually exclusive. The results for preferring both languages for news make that clear as Spanish and English qualities are both positively related to Latinos preferring both languages for news. With language skills representing identity-like characteristics, it appears again that identity is related to preferred language for news consumption among Latinos. Thus, strong support is found for H2 and H3.
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
LANGUAGE PREFERENCE OF LATINOS IN THE U.S.
15
Generation of residence is the second proxy concept employed to measure identity. Relative to first generation Latino immigrants, each of the other three generations included in the models prefers less Spanish-language news and prefers more English-language news or news in both languages. This finding supports H4. Like the findings to this point, later generations are more deeply connected to American society. As a result, Latinos’ identities reflect those of non-Latino Americans, thus encouraging English-language news consumption. The results of the analysis of the identity variables make it clear that no matter how it is conceptualized, identity is highly correlated with news language preference. Although causality for studies like this is elusive, it is worth noting the z scores of the significant variables to reinforce the depth of this connection. Scores that exceed 1.96 (negative or positive depending on the direction of the coefficient) are considered statistically significant at least at the 95% level. Although significance levels must be less than 100%, z scores are more open-ended. The z scores of the significant identity variables tend to exceed the significant control variables discussed next. The largest z scores are for language ability (z scores between 9 and 20). Some scores for other significant identity variables are below 3, but the majority range from 3 to 7. Although this does not prove causality, the consistent significance of the identity variables and their relatively high z scores should serve as evidence that identity is crucial to understanding language preference for news consumption. The control variable results demonstrate that identity is not the only determinant of language preference for news consumption although the results are much less consistent than the identity variables. Respondents living in households with higher income prefer less Spanish-language news and less news in both languages. Higher incomes are positively and significantly related to preferring English-language news. Being employed and owning a home are statistically insignificant for each dependent variable. Although the results are inconsistent, Latinos with a greater economic stake in American society appear to prefer news that focuses less intensely on Latinos alone. Variation in interest is expected to differentially affect the preferred language for news consumption by Latinos in the United States. In general, this is believed to be the case because individuals with more interest are thought to seek news sources with a wider scope that encompass all Americans. The first variable, interest in politics, is insignificant. Political knowledge and education, however, perform well and illustrate the role played by interest for compelling news consumption. Individuals with more political knowledge and more education prefer English news more and Spanish news less. Presumably, with increased knowledge and education comes increased acculturation. Individuals more acculturated to American
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
16
SALZMAN
society should naturally seek English-language news that provides broader information. The results for preferring both languages reveal no statistical significance for either political knowledge or education. In this instance, does increased acculturation mean heightened American identity or less Latino identity? Like economic factors, the connection between political knowledge, education, and identity is not obvious. Political participation is another way of capturing interest. The results of the statistical analysis reveal no support for political or civic participation influencing language preferences for news consumption. Political behaviors such as engaging in civic activities, contacting government officials, and simple voting behaviors are statistically unrelated to language preference for news media consumption.7 The party identification variables also provide scant illumination of our understanding of language preference due to statistical insignificance. Citizenship performs as expected with Latinos who are citizens preferring English or both languages for news, and noncitizens preferring Spanish-language news. That result may be capturing the language skills of the respondent as many citizens in the study are second-plus generation Latinos and will naturally be better at speaking English. Therefore, civic engagement and political participation do not appear to be reflective of acculturation in the same way that the identity variables are. The variable for measuring expectation derived from media experiences performed well. The press freedom score of the respondent’s birth country was found to be negatively correlated with preferring Spanish-language news and positively correlated with English news preference.8 Preferring both failed to reach statistical significance. This result, although interesting, most likely finds its explanation in the fact that second-plus generation Latinos are more than likely natural-born citizens. As the United States scores high on press freedom measures, this test may actually be uncovering other effects related to being born and raised in the United States (i.e., language skills).
7 It should be noted that some of the civic engagement and political variables were significant at the .10 level. The strongest of these is the relationship between Civic Engagement and consuming Spanish-language news (p ¼ .059). That relationship was negative, indicating that increased engagement in civic activities is negatively correlated with consuming Spanishlanguage news. The opposite relationship appears weakly present for English-language news consumption. Thus, it should not be surprising for future research to reveal a relationship between community involvement and language preference where no relationship is present for basic political participation like voting. The performance of other political=community variables including Registered to Vote and Republican also appear weakly related news media language preferences. These could be examples illustrating the role played by depth of community involvement. However, this study stops short of making those pronouncements. 8 The results remain the same when the press freedom score of the respondent’s families’ country of ancestry is included.
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
LANGUAGE PREFERENCE OF LATINOS IN THE U.S.
17
The demographic control variables performed well in the models. Male Latinos prefer Spanish-language news less than females and both news languages more than females. The strong desire for Spanish-language news by Latinas could be rooted in their relative roles in and out of the home. Females are less likely to work outside the house than males, thus decreasing their integration into society and increasing subsequent desire for Spanish news. Males and females were indistinguishable for preferring English news. Age is positively and significantly related to preferring English news and negatively related to preferring Spanish news. This is possible because individuals of an older age are more likely to be acculturated than the younger cohort. Age appears unrelated to preferring both languages for news. Being married and having kids reveal insignificant coefficients.
DISCUSSION The intent of this research project was to explain what motivates Latinos’ language preference for news consumption in the United States. Existing research on Latinos media consumption behavior has proliferated in recent years, but research specifically focused on news media consumption among Latinos is absent. Research on Latino attitudes and behavior highlighted the role that identity plays in shaping those attitudes and behaviors. Although no researcher focused exclusively on language preferences when consuming news media, this study provided theoretical support for social identity influencing those preferences. The empirical results of this study reinforce that theoretical supposition, thus encouraging research on Latino media behavior to focus first and foremost on consumers’ degree of Latino identity assuming that acculturation is a variable process that can enhance or diminish their embrace of Latino culture. This advancement simplifies and streamlines what is known about Latino media consumption behavior by providing a single idea that can be captured by many different measures. These results also add to the work done by social psychologists as it encourages media preferences to be included as another social process that is informed by social identity. Using the 2006 Latino National Survey data set, a series of statistical analyses were engaged. The Spanish news and English news preference models performed exceedingly well in their explanatory values. Those two models also provide face validity as the significant coefficients tended to be in opposite directions. That indicates that the two models are capturing the essence of language preference determinants. In accordance with H1, identity was found to have a strong effect on language preference with the control variables exacting at least some effect on the language preference of Latinos for news consumption.
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
18
SALZMAN
Although self-identification is significant for decisions related to language preference, it is appropriate to broaden the concept of identity to include identity proxies such as language skills (H2 and H3) and generation of residence (H4). All three groups of identity measures provide statistical support for the hypotheses. Whether identity is self-attributed or logically presumed, a strong relationship between identity and language preference is present. This finding is especially important as Latinos in the United States increasingly embrace their cultural heritage. When asked which language type Latinos preferred for news consumption respondents had three options: Spanish, English, or both equally. The first two are straightforward in concept. However, understanding decisions to claim both languages for news consumption is not as obvious. By taking a step back, it becomes clear that preferring both language types for news consumption more closely aligns with preferences for Spanish news. All of the identity variables have results that match responses for Spanish news preference in direction and significance with one exception. The results for English quality are similar for English news and both languages. The similarities in variable performance for Spanish news preference and preferring both equally indicates that Latinos who prefer both maintain a Latino identity but likely possess English-language skills that make consuming news in both languages preferable. Theoretical support for including control measures often hinges on identity-related behaviors and attitudes that reflect potential acculturation. Some of these measures are significant where others are not. There is reason to think that acculturation and social identity are closely related. A first-generation Latino is more closely tied to their Latin American culture and thus maintain a distinctly Latino identity, whereas a fourth-generation Latino is naturally more acculturated to American society. However, the nature of the survey questions and=or the lack of blatant theoretical connections between the concepts and identity prompts hesitation in assuming that all were identity related. This should be a primary focus of future research. Are all of these concepts potentially capturing cultural identity, or are they separate in some meaningful way? Spanish-language media is clearly targeted to Latinos. Thus, having a more pronounced Latino identity should heighten preferences for Spanishlanguage media relative to English-language media. The results of the statistical analyses appear to bear this out. Regardless of the performance of the control variables, it is clear that identity matters for shaping attitudes related to that identity. This project has taken a big step forward in shaping the understanding of what determines language preference for Latino news media consumers. Future research should continue to focus on identity when trying to
LANGUAGE PREFERENCE OF LATINOS IN THE U.S.
19
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
understand language preference for news consumption. As news media consumption is often seen as something that exacerbates existing attitudes and behaviors, understanding the connection between identity and media selection can likely determine consumption and subsequent media effects.9 REFERENCES Alba, R. D., & Nee, V. (1997). Rethinking assimilation theory for a new era of immigration. International Migration Review, 31, 826–874. Almond, G. A., & Verba, S. (1963). The civic culture: Political attitudes and democracy in five nations. Boston, MA: Princeton University Press. The Associated Press, & Univision. (2010). The Associated Press–Univision poll. Retrieved from http://surveys.ap.org Barreto, M. A., & Mun˜oz, J. A. (2003). Reexamining the ‘‘politics of in-between’’: Political participation among Mexican immigrants in the United States. Hispanic Journal of Behavioral Sciences, 25, 427–447. Basch, L., Schiller, N. G., & Blanc, C. S. (1994). Nations unbound: Transnational projects, postcolonial predicaments, and deterritorialized nation-states. New York, NY: Routledge. Coffey, A. J. (2009). Growth and trends in Spanish language television in the United States. In A. Albarran (Ed.), The handbook of Spanish language media (pp. 203–217). New York, NY: Routledge. DeSipio, L. (2003). Bilingual television viewers and the language choices they make. Claremont, CA: Tomas Rivera Policy Institute. Fraga, L. R., Garcia, J. A., Hero, R., Jones-Correa, M., Martinez-Ebers, V., & Segura, G. M. (2006). Latino National Survey (LNS). Ann Arbor, MI: Inter-university Consortium for Political and Social Research. Fry, D. L., & McCain, T. A. (1983). Community influentials’ media dependency in dealing with a controversial local issue. Journalism Quarterly, 60, 458–463. Gans, H. (1992). Second generation decline: Scenarios for the economic and ethnic futures of post-1965 American emmigrants. Ethnic and Racial Studies, 15, 173–192. Hale, M., Olsen, T., & Fowler, E. F. (2009). A matter of language or culture: Coverage of the 2004 U.S. elections on Spanish- and English-language television. Mass Communication and Society, 12, 26–51. Haslam, A. S. (2001). Psychology in organizations. London, UK: Sage. Iyengar, S. (1987). Television news and citizens’ explanations of national affairs. The American Political Science Review, 81, 815–832. Iyengar, S., & Kinder, D. R. (1987). News that matters: Television and American opinion. Chicago, IL: University of Chicago Press. Jeffres, L. W. (2000). Ethnicity and ethnic media use: A panel study. Communication Research, 27, 496–535. Jones-Correa, M. (1998). Between two nations: The political predicament of Latinos in New York City. Ithaca, NY: Cornell University Press.
9 Uncovering the impact of media consumption has been a primary focus of political communication for decades (Herman & Chomsky, 1988; Iyengar, 1987: Iyengar & Kinder, 1987; Cappella & Jamieson, 1996; Norris, 2000; Prior, 2007; Salzman & Aloisi, 2009).
Downloaded by [Northern Kentucky University], [Ryan Salzman] at 06:19 05 September 2013
20
SALZMAN
Lieberson, S. (1973). Generational differences among blacks in the north. American Journal of Sociology, 79, 550–565. Long, J. S., & Freese, J. (2006). Regression models for categorical and dependent variables using Stata (2nd ed.). College Station, TX: StataCorp LP. Norris, P. (2000). A virtuous circle: Political communications in postindustrial societies. New York, NY: Cambridge University Press. Postmes, T., & Branscombe, N. (2010). Sources of social identity. In T. Postmes & N. Branscombe (Eds.), Rediscovering social identity: Core sources. New York, NY: Psychology Press. Prior, M. (2007). Post-broadcast democracy: How media choice increases inequality in political involvement and polarizes elections. New York, NY: Cambridge University Press. Rios, D. I., & Gaines, S. O., Jr. (1998). Latino media use for cultural maintenance. Journalism and Mass Communication Quarterly, 75, 746–761. Roberts, D. F., & Foehr, U. G. (2008). Trends in media use. The Future of Children, 18, 11–37. Ruggiero, T. E. (2000). Uses and gratifications theory in the 21st century. Mass Communication & Society, 3, 3–37. Salzman, R. (2011). News media consumption in Latin America: Who does it? Journal of Spanish Language Media, 4, 23–39. Salzman, R., & Aloisi, R. (2009). News media consumption and political participation in Central America: Causation and explanation. Journal of Spanish Language Media, 2, 46–75. Subervi-Velez, F. A. (1986). The mass media and ethnic assimilation and pluralism: A review and research proposal with special focus on Hispanics. Communication Research, 13, 71–96. Subervi-Velez, F. A. (2008). The mass media and Latino politics: Studies of U.S. media content, campaign strategies and survey research: 1984–2004. New York, NY: Routledge. Trevino, L. K., Webster, J., & Stein, E. W. (2000). Making connections: Complementary influences on communication media choices, attitudes, and use. Organization Science, 11, 163–182. Turner, J. C. (1984). Social identification and psychological group formation. In H. Tajfel (Ed.), The social dimension: European developments in social psychology (Vol. 2, pp. 518–538). Cambridge, England: Cambridge University Press. Turner, J. C. (1999). Some current issues in research on social identity and self-categorization theories. In N. Ellemers, R. Spears & B. Doojse (Eds.), Social identity: Context, commitment, content (pp. 6–34). Oxford, UK: Blackwell. Turner, J. C., & Oakes, P. J. (1986). The significance of the social identity concept for social psychology with reference to individualism, interactionism and social influence. British Journal of Social Psychology, 25, 237–252.
