Ni x 1 9&3 lllHllHHlHUW‘WHHLNIUINHWIIIII‘HIII THS 9 00 I. mum?" LIBRARY Michigan State University This is to certify that the thesis entitled IMAGES OF HISPANICS IN MAINSTREAM AND ETHNIC MAGAZINE ADVERTISEMENTS presented by Rebecca F. Bowden has been accepted towards fulfillment of the requirements for the Master of degree in Intercultural Communication Arts I V Major’Professor’s Signature 07W ’3; 01003 Date MSU is en Animative Action/Equal Opportunity Institution PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE rc'“7 1 9 9%DQ .Ix'r C \— "l A v I v 8/01 c:/ClRC/DateDue.p65-p. 15 IMAGES OF HISPANICS IN MAINSTREAM AND ETHNIC MAGAZINE ADVERTISEMENTS By Rebecca F. Bowden A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTERS IN INTERCULTURAL COMMUNICATION Department of Communication 2003 ABSTRACT IMAGES OF HISPANICS IN MAINSTREAM AND ETHNIC MAGAZINE ADVERTISEMENTS By Rebecca P. Bowden A content analysis of People and People en Espanol (PE) advertisements from 2002 is performed to examine advertising portrayals of Hispanics in ethnic minority media, as predicted by dual role theory (DRT) and agenda setting. The issues investigated are the occupational roles of Hispanics, setting distributions of Hispanic models, the “Latin lover” portrayal, Hispanic cultural content cues, the body size of models, context and themes of advertisements across both magazines. Results indicate that PE has one additional occupational portrayal, the setting distribution of advertisements containing Hispanic models was not significantly different, the concept of “Latin lover” among men was enforced in PE but not in People, Hispanic cultural content cues were prevalent in PE but not in People, rating of female body size is significantly larger in PE over People, the context of models and advertisement themes were significantly different. The findings were consistent with the predictions of DRT and agenda setting. TABLE OF CONTENTS List of Tables and Figures iv Introduction ..... 1 Why Study Ethnic Minority media? 3 Dual Role Theory ,,,,,,,, ...5 Agenda Setting Theory__ 7 Methods 11 Sample 1 1 Coding Scheme __________________ 12 Coders 16 Results 13 Discussion 23 Appendix A (Tables One — Ten) 28 Appendix B (Codebook & Operational Definitions) 39 References," 50 iii LIST OF TABLES Table One: List of Variables for this Study Table Two: Calculated Scott’s Pi Intercoder Reliability of Variables Table Three: Occupational Distribution of Hispanics Across Magazines .................. Table Four: Occupational Setting Distribution Table Five: Attractiveness Ratings of Female Models. Table Five: Sexiness Ratings of Female Models Table Six: Attractiveness Ratings of Male Models Table Six: Sexiness Ratings of Male Models Table Seven: Hispanic Cultural Cue Distribution in People en Espanol ,,,,,,,,,,,,,,,,,,,, Table Eight: Context Distribution of Advertisements in People en Espanol ,,,,,,,,,,,,, Table Nine: Primary Theme Distribution Table Ten: Theme Distribution Comparison ....................... iv 29 3O 31 32 33 ”.33 34 34 35 36 37 38 Images of Hispanics in Mainstream and Ethnic Magazine Advertisements As the American population continues to diversify, marketing efforts target non- mainstrearn cultural groups by varying the content and images in advertisements. Academics and practitioners are interested in how variations in social context, minority status and race affect advertisement placement and audience members (Grier & Deshpande, 2001). With the introduction of each new venue including ethnic minority media, the potential for a wider diversity of images, messages and effects substantially increases. Ethnic minority media can be defined as media venues targeting a minority group and through the medium allowing the audience to “perceive themselves as constituting a common culture, ancestry, language, history, religion or custom” (Riggins, 1992, p.1-2). Ethnic minority media have two unique purposes (FullertOn & Kendrick, 2000). First they allow the ethnic group to understand how they should behave in order to ease acculturation into mainstream America’s culture. Second they help maintain ethnic heritage by reinforcing an independent cultural identity. Advertisers understand that minority consumers respond more favorably to advertisements in general compared to ethnic majority consumers, making the minority market enticing and financially rewarding (Grier & Deshpande, 2001). To increase the effectiveness of advertisements toward Hispanic audience members, savvy marketers have begun to include Hispanics and to portray them more favorably in their advertisements. Examining the images in advertisements in People magazine that produces both an English reader and Spanish reader version should uncover the types of portrayals that exist. It should also expose potential differences between Hispanic specific media and mainstream American media. This is the first step in answering the question, how do image and content in Hispanic targeted magazines affect the self-concept or cultural identities of Hispanics? Critics suggest that the portrayals of Hispanics in advertisements may not be accurate depictions of Hispanics in America (Alaniz & Wilkes, 1995). In addition, harmful effects toward a depicted audience can be induced by brief exposures to inaccurate portrayals or limited role portrayals (Lafky, Duffy, Steinmaus & Berkowits, 1996). Further studies suggest that inaccurate portrayals of minority groups can contribute to prejudice and inequality in educational and occupational opportunities or limited role portrayals may develop self-fulfilling prophecies (Taylor, Lee & Stern, 1995). Orbe (1998) suggests this phenomenon may occur due to an imbalance of power leading to a muted group mentality, such that the perspective of minority groups are repressed by the majority opinion. The aim of this study is to determine if images vary in advertisements of an English reader and Spanish reader versions of the same magazine. To answer this question the study begins by explaining why research should be conducted on ethnic minority media and discusses possible effects of misportrayal. Next, dual role theory (DRT) (Subervi- Velez, 1986) will illustrate why ethnic minority media must be examined due to its potential to intensify negative effects. Agenda setting theory (Dearing & Rogers, 1996) will then be utilized to defend why a content analysis is the first step to understanding variation between the magazine versions. Why study Ethnic Minority Media? Hispanic Americans in the United States have a combined buying power of $630 billion a year (AHAA, 2002). This makes Hispanic consumers extremely attractive to marketers and advertisers. In fact Santiago and Valdes (a Hispanic advertising firm) suggest that only eight percent of an advertising budget is needed to optimally reach this practically untouched audience (AHAA, 2002). This is interesting considering that the Hispanic population represents about thirteen percent of the United States population. In essence, marketers are getting an unheard of return on their investment. Companies only need to invest eight percent to glean a thirteen percent market share (AHAA, 2002). According to Grier and Deshpande (2001), advertisements targeted toward a numeric minority are more effective than majority—directed advertisements. In fact, these targeted advertisements are so effective that the minority group responds more favorably to them than a majority group would to a similar advertisement aimed specifically at them. These researchers discovered that when a group is a numeric minority and advertisements include depictions of that minority, the advertisement becomes more effective. As advertising evolved, it began to focus on image-based advertisements (Jhally, 1995). Images are used to promote a product or to create a connection with a person in an advertisement. By using common images, advertisers sell products as audiences form associations in thirty seconds or less. This has become second nature, and as Jhally (1995) said, advertisement images “are ubiquitous, it is the air that we breathe as we live our daily lives” (p. 79). Since images have become a staple in advertisements, the question remains what are these images teaching consumers? A study looking at the portrayals of Hispanics in mainstream America’s media illustrates a possible detrimental effect of the media’s portrayals of Hispanics (Taylor et al., 1995). This study found that Hispanic Americans are usually shown as uneducated blue-collar workers who are unwilling to acculturate into mainstream America’s culture (Taylor et al., 1995). Because of this image, Hispanic children may feel it is socially permissible to drop out of school, which may explain a portion of the high number of students from this ethnic group who drop out of school earlier compared to other Americans (Czepiec & Kelly, 1983). This study also illustrates why academics are concerned that misportrayals of Hispanics may contribute to inequality in occupational opportunities and self-fulfilling prophecies (Taylor et al., 1995). Self-fulfilling prophecies are when people perceive limited opportunities for themselves based on the media’s portrayals of their group, so they limit themselves to these roles and never try to attain anything else (Fullerton & Kendrick, 2000). In essence, the individual seals his or her own fate instead of trying to move beyond an expected role in society. As marketers try to reach multicultural America, group boundaries become less important than the distinctions advertisers make between “us” and “them” (Tharp, 2001). Advertisers rely heavily on images to create quick audience associations by distinguishing one ethnic group from another (Jhally, 1995) thus increasing the effectiveness of their advertisement (Greir & Desphpande, 2001). As image stereotyping becomes more prominent, the chances of misportrayal and negative effects can also increase (Taylor et al., 1995). The nature of ethnic minority media and its potential to increase negative effects associated with misportrayal will be explored next. Dual Role Theory DRT discusses how ethnic minorities use their own media to acculturate into the majority culture (Subervi-Velez, 1986). Acculturation is the process of learning about another culture. Ethnic minority media serve two purposes in aiding acculturation (Subervi-Velez, 1986). First, it allows minorities to acculturate into the dominant group, and second, it maintains pluralism. Pluralism is sustained ethnic differentiation and continued heterogeneity. Ethnic minority media allows individuals to shield themselves from the mainstream culture and preserve their own ethnic heritage. On the other hand, ethnic minority media also teaches individuals the roles they can have, or the behavior they should embrace to be accepted into the mainstream American culture (Subervi- Velez, 1986). Faber, O’Guinn and Meyer (1987) examined how media portrayals of minority groups affect minority acculturation. They found that misperceptions regarding the minority group could inhibit or halt acculturation. In fact, Hispanics in this study believed that the portrayals they saw in mainstream media were biased and unrepresentative of the Hispanic culture. According to DRT, over time these misportrayals may limit the occupational roles Hispanics believe are attainable for them (Subervi-Velez, 1986). Hispanics may believe that only a select number of occupations are available and thus limit their potential and halt acculturation. Hispanics are blocked from learning about available occupational roles in a society when they feel isolated and pi geonholed by that same society. The Faber et al. study (1987) was conducted in the mainstream American media; however, very little research has examined DRT in Hispanic-ethnic minority media. Albarran and Umphrey (1993) similarly found that Hispanics said they watched television because “It helps me learn about myself and others,” and “I can spend time with friends or family.” This study further confirms the media’s role in teaching Hispanic people who they are and should be in mainstream America’s culture, while allowing the viewers to socialize thus preserving their culture. According to DRT, the images within advertisements of People en Espanol should illustrate a diversity of occupational roles, and should include occupational roles beyond mainstream media’s roles. By doing this the Hispanic media’s goal to educate would be met (Subervi-Velez, 1986). Thus DRT informs these two hypotheses. H1: People en Espanol will contain a greater diversity of Hispanic occupational role depictions in its advertisements as compared to Hispanic occupational depictions in People, which is targeted toward mainstream American culture. H2: The occupational setting distribution for advertisements containing Hispanics in People en Espanol will be significantly different from those in People. To determine if the advertisements illustrate additional occupational roles beyond mainstream American media’s typical portrayals, it is important to understand how this group is normally portrayed. Studies have found that Hispanics are usually shown as blue-collar workers and/or migrant workers (Taylor et al, 1995; Fullerton & Kendrick, 2000; Mastro & Atkin, 2002). Although the image of Hispanics as “Latin lovers” is not an occupation, the concept is still prevalent in the mainstream media. Fullerton and Kendrick (2000) found the “Latin lover” concept was common in Hispanic broadcasting on cable television, as did Mastro and Atkin (2002) in their billboard study. Both sets of authors conceptualized the concept to be “very attractive and sexy models.” Due to these findings it would be expected that the “Latin lover” concept should also be found in magazine content. This reasoning informs the third hypothesis. H3: Both male and female Hispanic models in People en Espanol and People will be more attractive and sexy as compared to all other models in this same medium. As shown, DRT can be used to predict that a greater diversity of occupational roles for Hispanics should exist in People en Espanol over People. This would fulfill the ethnic minority media’s first function of educating the vieWers on what roles exist for them. If the media does not illustrate occupational diversity, role limitation or self- fulfilling prophecies could be a serious consequence. Agenda setting theory will reveal each magazine’s advertising content, which can be analyzed to detemrine if advertisements maintain cultural uniqueness, thereby, fulfilling the second prediction of DRT — that ethnic minority media maintains cultural pluralism. Agenda Setting Theory An “agenda” is a set of issues that is communicated in a hierarchy of importance at a given time (Dearing & Rogers, 1996). Agendas result from different issues rising and falling in importance over time. This is a dynamic interplay and agenda setting provides researchers with a snapshot of its fluidity (Dearing & Rogers, 1996). Issues are defined as conflict between two or more groups over the distribution of positions or resources (McCombs & Shaw, 1993). Traditionally agenda setting research measures the salience of issues because changes in salience indicate the importance society and the media place on an issue (Kiousis, Bantimaroudis & Ban, 1999). This theory previously focused almost exclusively on political issues, although now the theory is recognized to incorporate other types of issues. Recently, researchers have paid attention to how the media frames issues (Dearing & Rogers, 1996). Framing refers to the interpretation the media places on an issue, thus allowing the issue to appear positive or negative to viewers (McCombs, Shaw, Weaver & Hamm, 1999). Framing tends to reflect the culture around the media, so an issue should be framed differently depending on the target culture (Kiousis et al., 1999). The way an issue is framed is said to reflect how the media are telling the audience to think about another group (Roessler, 1999). This is considered second level agenda setting. The media is not only telling the audience what to think about, but how to think about topics (McCombs et. al, 1999). Second level agenda setting has not been studied as widely nor has it been applied heavily to minority specific media (McCombs & Shaw, 1993). Nonetheless, the basic principles of agenda setting still apply in inter-group settings (Miller & Wanta, 1996). Content analyses are the main tool of agenda setting (Krippendorf, 1980). By conducting a content analysis, researchers are able to quantify meaning from advertisements (Roessler, 1999). As the first step, studying the content within the media and comparing it to the characteristics of the target audience provides a better understanding of the agenda-setting process. If meaning can be found within a given medium then two questions can be answered: How important does the media think an issue is, and how is the media telling the audience to think about this issue? Although agenda setting cannot be used to explain how salient images are affecting individuals, it remains useful by allowing researchers to examine the issues, images and topics the media thinks are important (McCombs & Shaw, 1993). Agenda setting provides the opportunity to examine how the media is telling its audience to think about issues. The snapshot agenda setting offers becomes exceptionally useful when studying a newer media, such as ethnic minority media. Little research has been done in this area; yet, agenda setting can help predict the kinds of content and issues that will exist If, as agenda setting suggests, the media reflects the culture of which it is a part, then the cultural content should vary within one magazine that is targeting two distinct cultures (Dearing & Rogers, 1996). Because images are so essential in advertising (Jhally, 1995), the advertising images within the magazines should reflect the target culture. This would allow the target culture to make quick associations (Tharp, 2001), thereby increasing the effectiveness of the advertisements (Greir & Desphande, 2001). When examining the advertisements in the two magazines one targeted toward Hispanics one targeted toward Mainstream American culture, cultural distinctions should appear. The cultural content cues more frequently expected in People en Espanol are: Hispanic art, food, flags, colors, maps, textiles, landmarks and plants, family contexts, and a preference for larger body sizes of women (Fullerton & Kendrick, 2000; Mastro & Atkin, 2002). The advertisements in People en Espanol should also contain more depictions of models in family and social settings due to their preference for these types of portrayals (Fullerton & Kendrick, 2000; Taylor et. al’s 1995). This logic informs the fourth, fifth and sixth hypothesis. H4: The advertisements in People en Espanol will have more Hispanic cultural content and cues as compared with People. H5: The average body size for women will be larger in People en Espanol as compared to People. H6: The context distribution of advertisements with models in People en Espanol will be significantly different from People. This current study also examines which issues the media think are important and possibly what the media is telling us to think about the Hispanic culture. This informs the following research question. RQl: What themes about Hispanics are advertisements delivering to the target audience? 10 Methods Sample People en Espanol was selected for analysis because it has one of the highest circulations, 229, 950, among magazines targeting Hispanic consumers (AdAge, 2002). People (its mainstream counterpart) has the highest circulation, 2, 216, 560, among all magazines (AdAge, 2002). These magazines were selected for inclusion in this study based on their wide circulation and comparative demographics. According to People ’3 circulation desk, 65% of their readers are women and 35% are male, over half of their subscribers (56%) are between the ages of 24—49, and the average household income is $59,000. Both magazines are virtually identical on these key components. All advertisements were analyzed only once and repetitions across issues were excluded to avoid violations of the assumption of independence between data points (Neuendorf, 2002). Based on this criterion, one issue each of People and People en Espanol were briefly analyzed to determine how many advertisements would meet this criterion. On average fifty advertisements were found in each issue. Taking into account identical advertisements between issues, about thirty to forty usable advertisements were collected per issue. All the advertisements in each selected issue that met the above criteria were coded. Based on the above numbers, all 2002 issues of the People en Espanol were selected. People en Espanol is published ten times a year, however, only nine were still available on back order. Issues from 2002 were selected because the most recent magazines should depict new trends within advertisements. Next, sampling was 11 conducted using a simple random sampling without replacement technique (Neuendorf, 2002). Issues were selected by placing issue numbers in a hat and drawing seven (Budd, Thorp & Donohew, 1967). The issues selected included February, March, April, August, September, October and November. The remaining two issues (May and December) were reserved for training. Because People puts out more issues a year (on average 52) a modified sampling method had to be undertaken. For consistency, only 2002 issues that corresponded to the months selected from People en Espanol were included. This was done to inhibit any possible confounds of having units drawn from different times in the year thus influencing the results (Nueendorf, 2002). For each month selected from the People en Espanol the four or five (depending on the month) corresponding issues of People were placed into a hat and one issue was selected. Once all issues were selected, the pertinent back issues were purchased from Time Warner Inc. Advertisements to comprise the sample were selected and cross-referenced between issues to avoid duplication. For People en Espanol, 288 advertisements were found while 90 others were excluded because of duplication. Four advertisements were incompletely coded, thus they were discarded decreasing the sample to 284. For People, 364 advertisements were included in the sample and 49 were excluded due to duplication. The total sample size was 648 advertisements. Coding Scheme The coding scheme was based on three elements-- occupational depictions, cultural cues and types of themes present in the advertisements. Coding for occupational 12 depiction of the primary character in the advertisement was based on the categories identified in a previous study conducted by Taylor et al. (1995). The concept of primary person was defined as a character who is very important to the advertising theme or layout, shown in the foreground usually holding or demonstrating the product (Taylor et al., 1995). In addition, children shown with an adult were always listed as the secondary person, even if the child appeared closer in the foreground. This was added because coders had difficultly determining the sexiness and attractiveness of children. Second for this study coders initially had a difficult time determining between Caucasian and Hispanic ethnicity. To solve this problem, the coders and researcher created a working definition. The coders agreed that to code someone as Hispanic the person would have a rounder face, flatter nose, darker complexion and usually darker eyes. Hair color was omitted because of how easy it is to change hair color. The coders had no difficulty coding the remainder of the ethnicities - African American and Asian. The first variable, taken from the Taylor et al. study (1995) is the setting of the advertisement (Refer to Table 1 for a full list of settings). This variable hints at the occupation of the primary person in the advertisement. For example if the setting is a business environment and the primary person in the advertisement is wearing a suit then it seems logical to infer that the primary person in the advertisement has a business occupation. One addition was made to Taylor et al’s (1995) coding scheme. A “technical setting” was added which is defined to include hospitals, clinics, medical labs, computer labs, robotic labs or engineering settings. 13 Next, occupational categories were derived from Taylor et al.’s (1995) settings. Occupations were broken down into seven applicable categories (See Table l) and one “other” category. This will test hypothesis 1, which suggests that People en Espanol should have more varied portrayals of occupational roles for Hispanics as compared to the Hispanics portrayed in People. The occupation variables included those typically used to portray Hispanic people such as blue-collar workers and migrant workers (Faber etal., 1987) under the heading of “blue collar” or “natural” setting. The coding scheme also includes novel settings such as “technical setting.” By including normal roles and novel occupational roles support could potentially be found for DRT. The coding scheme then assessed a range of cultural cues drawn from a variety of researchers. It also included some new categories of cultural depiction created for this study. The coding scheme examined “context” within advertisements based on Taylor et al. (1995) study and secondly, evaluated the relationship of people in the advertisement. (Taylor et al., 1995). This variable suggests that Hispanic people prefer to be shown in family contexts within advertisements (Fullerton & Kendrick, 2000). Thus it may shed light on the cultural cue of family contexts. For example Hispanics prefer to be portrayed in family or social context, so they should appear that way in the People en Espanol because it is targeted toward the Hispanic community. The different contexts studied include family contexts, social contexts, and business contexts. Next more cultural cues were taken from Mastro and Atkin’s (2002) study on Hispanic cultural cues on billboards. Their criteria were drawn from a broad spectrum of research. To capture ethnic minority media’s uniqueness, common cultural cues such as 14 traditional colors, flags, festivals and maps were studied (Mastro & Atkin, 2002). Traditional Hispanic colors are defined to as green, red, yellow and blue. Flags and maps are categorized into American and non-American. If a flag or map should appear it will be cross-referenced with the census bureau’s definitions of Hispanic countries. Festivals are defined as depictions of Cinco de Mayo, Spanish Carnival, Spanish Dancers or Day of the Dead celebrations. The researcher also added some novel categories to this list of cultural cues. Advertisements were analyzed to determine if they contained depictions or references to Hispanic art, food, textiles or plants and landmarks. Also, the written message was analyzed to determine if it is “consistent with the cultural depictions,” if it “offers no cultural cues,” or if it “offers additional cultural cues.” At times the text in an advertisement may offer additional cultural cues, so by analyzing the text these cultural cues can be unearthed. Mastro and Atkin (2002) also looked at gender portrayal. For example it is quite common for women to appear as passive and sexual (Alaniz & Wilkes, 1995). Mastro and Atkin (2002) examined this aspect by making a determination of model sexiness and attractiveness. By replicating this portion of the research, hypothesis two can be studied. In addition, because of the typical portrayal of men as Latin lovers (Fullerton & Kendrick, 2000), this analysis was repeated with men. One additional aspect to be studied is the body size of women. Fullerton and Kendrick (2000) found that the Hispanic culture prefers larger shaped women as compared to mainstream American culture. Thus as consistent with a study assessing 15 body size in the mainstream media (Greenberg, Eastin, Hofshire, Lachlan & Brownell, 2003), coders will be asked to identify the body size of men and women in advertisements to determine if this cultural difference exists. To maintain consistency with the above mentioned study, the same graphics will be used to categorize the men and women’s body size, although one addition was made that suggested when an advertisement is zoomed in so close only the face shows the actor is considered “uncodeable.” The last aspect the coding scheme analyzes the types of themes present in an advertisement. This portion was included to answer research question one. Eight themes were created (see Table l) by briefly analyzing the magazine editions of People and People en Espanol not included in the sample. An “other” category was also included. Coders were asked to determine the primary theme of each advertisement and then determine if another primary theme was co-existing. For example, when analyzing a celebrity in a beauty advertisement, the coder would categorize the advertisement as celebrity and beauty. Thus the coding scheme analyzed occupational roles, cultural cues and themes within advertisements. Although the coding scheme was not exhaustive, it was designed to shed light on the hypothesized relationships of these variables to each respective magazine. Coders Two coders who were functional or fluent in Spanish participated in this project. Both were Hispanic in ethnicity. One was male, with Mexican ethnicity, and one was 16 female, with Colombian ethnicity. They were trained using advertising images from People and People en Espanol (the issues not utilized in the sample). Coders were trained on advertisements similar to those they would actually be coding. Great care was taken to be certain that the training advertisements did not appear in the sample, thus it is unlikely that the coders saw advertisements that appear in the sample. After the coders had been trained they were asked to code a series of pilot advertisements (N = 60) independent of each other. Intrercoder reliability was then established using a Scott’s pi analysis as detailed by Porter and Levine-Donnerstein (1999), and any discrepancies were discussed (Neuendorf, 2002). This process was only done once because an overall average reliability of 90 percent (Pi = .9) was attained (see table 2), within the acceptable range suggested by Krippendorf (1980). Each individual variable passed the first reliability check, and only one was below 80 percent. After this stage was completed, the sample advertisements were divided as equally as possible between the coders. Coders were also kept blind to the hypotheses of the study throughout the training, pilot, and coding process. This was done to ensure that the coders remained as free from bias as possible and to diminish or inhibit demand characteristics (Neuendorf, 2002). The entire training, pilot and coding process was done in a fashion consisted with that of Banerjee, Capazzoli, McSweeney & Sinha (1999) who offered suggested techniques of statistical control and ways to limit the confounding influence of researcher bias. [7 Results The collected data were cleaned and any errors were corrected or recorded as errors before any analysis began. Hypothesis I predicted that People en Espanol would contain a greater diversity of Hispanic occupational role depictions in its advertisements as compared to Hispanic occupational depiction in People. A frequency table was run isolating only advertisements that contained Hispanic models. Then the number of roles that occurred in each medium was compared. Only one additional occupation category was found in People en Espanol as compared to People. This category was “service.” To test if a significant difference between the two distributions existed a chi-square analysis was run (see table 3). The results were not valid due to empty cells or cells with to few entries. Overall People only had 14 depictions of Hispanic models compared to People en Espanol ’s 122, making a chi-square analysis impossible to run. Thus, little support was found for hypothesis 1. Next hypothesis 2, which predicted that the occupational setting distributions of Hispanics in People en Espanol would be significantly different from Peoples, was analyzed. “Professional,” “blue collar” and “medical/technical” categories were all dropped due to cell sizes less than 5. A chi-square analysis was run on the remaining categories. The result was not significant, X2 (2, N = 211) = 2.39, p = .303 (see table 4). Thus there was no support for hypothesis 2. Hypothesis 3 which predicted that both male and female Hispanic models would be more attractive and sexy than all the other ethnicities was then tested. To test this hypothesis a series of ANOVA’s were run. First, the advertisements with females were 18 selected and two 2 x 2 factorial ANOVAs were used to compare attractiveness/ sexiness of female models, magazine and ethnicity. There was no main effect for attractiveness, and magazine, F(1, 357) = .38, p = .54 or ethnicity, F(1, 357) = .029, p = .86 , nor was there a main effect for sexiness and magazine, F(1, 357) = .4, p = .53 or ethnicity, F (1, 357) = .002, p = .96. There was also no significant interaction effect of magazine and ethnicity with attractiveness, F(1, 357) = 2.69, p = .10 or sexiness, F(1, 357) = 2.29, p = .13. Although there was a trend for female Hispanic models to appear more sexy and attractive in People when compared to all other ethnicities and less sexy and attractive in People en Espanol when compared to all other ethnicities; the results were not significant (see table 5). Thus for women the results were not consistent with the hypothesis 3. To complete the test of hypotheses 3 this process then was repeated with men. For the sexiness of male Hispanic models there was no main effect for magazine, F (1, 357) = .005, p = .95 or ethnicity, F(1, 357) = .19, p =.67, however, a significant interaction effect was discovered for magazine and ethnicity, F(1, 357) = 4.850, p < .05. A Hispanic male was rated more sexy (M = .6, SD = 1.01) than all other ethnicities (M = .29, SD = .81) in People en Espanol, and less sexy (M = .23, SD = .83) than all other ethnicities (M = .69, SD = 1.1) in People (see table 6). Next, attractiveness of Hispanic models was analyzed. Once again no significant main effect for magazine, F(1, 357) = .07, p = .8 or ethnicity, F (1, 357) = .25, p =.6, was discovered. Identical to sexiness of male Hispanic models a significant interaction effect was discovered for magazine and ethnicity, F(1, 357) = 4.83, p < .05. A Hispanic male was rated more attractive (M = .6, SD = 1.01) than all other ethnicities (M = .28, SD = .78) in People en Espanol, and less 19 attractive (M = .23, SD = .83) than all other ethnicities (M = .74, SD = 1.3) in People (see table 6). Thus for men the results were consistent with hypothesis 3. Hypothesis 4 was then tested. It predicted that advertisements in People en Espanol would have more Hispanic cultural content and cues as compared to People. An independent samples t-test was used to test the hypothesis. The results showed a significant difference between People en Espanol and People, t(646) = 6.631, p < .05. People en Espanol contained more cultural content and cues (M = .116, SD = .32) than People (M = .003, SD = .05). Therefore, it was concluded that the data were consistent with hypothesis 4. Next an analysis of the types of cultural cues that appeared in the data was conducted. Only one advertisement in People had Hispanic cultural cues and it was a reference to Hispanic food. Of the 36 advertisements in People en Espanol containing cultural cues two advertisements had two cultural cues in one advertisement, one advertisement contained seven cues, and the remainder contained just one cultural cue per advertisement. In People en Espanol, one Hispanic flag, and one Hispanic map appeared. Hispanic festivals, Hispanic textiles and Hispanic Art were referenced three times each (see table 7). Seven references to Hispanic plants/landmarks, and Hispanic food respectively also appeared. The remaining eleven where all coded as “other.” A post hoc analysis was conducted on the advertisements in the other category. The researcher analyzed the open-ended responses and found that four references were made to traditional Hispanic pastimes, like soccer and dominos. Three of the advertisements in 20 the other category were using colors traditional to Hispanics and the remaining three referenced traditional Hispanic sayings. Hypothesis 5 predicted that the average body size for women would be larger in People en Espanol as compared to People. The body size ranges from 1 to 6 (see table 1). An independent samples, one tailed, t-test was used to test the hypothesis. The results showed a significant difference between body size of women in People en Espanol and People, t(213) = 1.9, p < .05. Female models in People en Espanol were rated with a larger body size (M = 2.6, SD = .87) than female models in People (M = 2.4, SD = .8). Therefore, it was concluded that the data were consistent with the hypothesis 5. Hypothesis 6, which predicted that the context distributions of advertisements with models in People en Espanol would be significantly different from People, was analyzed. The result was significant, X2(4, N = 379) = 14.238, p < .05 (see table 8). Models in People en Espanol were least likely to appear in “business settings,” as apposed to People where they are more likely to appear in this setting. Models in People en Espanol were more likely to appear in “Family” and “social” settings, as apposed to People where they were less likely to appear in these setting. Thus it is concluded that context of Models in People en Espanol was significantly different from context of Models in People. Thus the data were consistent with hypothesis 6. The last analyses run were designed to answer research question one which investigated what themes about Hispanics are advertisements delivering to the target audience of People en Espanol. To answer this question an analysis of the frequency of the different themes in People en Espanol was run (see table 9). After completing the 21 analysis 206 advertisements were in the “other category,” thus a post hoc analysis was conducted. Most of the 206 advertisements were placed into four additional codes, ,9 6‘ ’9 66 “drug and wellness, materialism, cigarettes/alcohol,” and “public service.” The advertisements were placed into the new categories or existing categories based on the coder’s open-ended descriptions of all advertisements they placed into the “other” category. A chi-square analysis was then run to see if the themes in People and People en Espanol differed from one another. The categories of “Romance” and “immigration story” were omitted due to small cell size. Eighteen advertisements in the “other category provided no theme information, so they were discarded. The result was significant, X2(9, N = 623) = 46.018, p < .05 (see table 10). Models in People in Espanol were most likely to appear in an advertisement with a ‘beauty” theme and least likely to appear in an advertisement with a “travel and escape” theme. Whereas, models in People were most likely to appear in “domestic/food” themed advertisements and least likely to appear “public service” themed advertisements. 22 Discussion The results of this study become most interesting when each result is interpreted in light of the DRT, the framework of this study. Although support of this theory is mixed the findings of this study do provide a snap shot of advertisement differences across People and People en Espanol. It could be argued that a little evidence was found for the first claim of DRT, which suggests that ethnic minority media teaches its viewers about the occupational roles they can pursue; with only one additional occupation category this study does not strongly support this claim. In additions when the results of hypothesis 2 are examined in conjunction with hypothesis 1 it becomes clear that People en Espanol does not try to teach its audience about the occupational roles they can pursue. After all little difference was found between the two versions of People. It can be suggested that the finding of Hispanic male models fits within the framework of DRT. Previous researchers have found a trend for Hispanic male models to typically appear as very attractive thus enforcing the concept of “Latin Lover” (Fullerton & Kendrick, 2000). A study by Fullerton & Kendrick (2000) found this when analyzing 162 television advertisements from a Spanish-language affiliate in a major US market. This study also found that sex-role portrayals in the TV advertisements were consistent with the traditional norms of the Hispanic culture. Fullerton and Kendrick determined traditional norms by examining the sex-role portrayals reported by Hispanics in a focus group sponsored by the American Marketing Association. In this study the image of men as “Latin lovers” was reinforced, as Hispanic male models appeared significantly more attractive and sexy in People en Espanol as compared to People. Thus 23 as DRT predicts, the male Hispanics model is shown as a “Latin Lover” in People en Espanol, but not in People, thereby, maintaining cultural pluralism and reinforcing a traditional Hispanic norm of sex-role portrayal. Hispanic cultural content within the two magazines is also consistent with DRT and agenda setting (Subervi-Velez, 1986; Dearing, 1996). Both theories predict that a medium should portray the culture of which it is a part. This study only found one Hispanic cultural cue in People and 36 Hispanic cultural cues in People en Espanol. This further suggests that cultural pluralism is being maintained in People en Espanol. Since Hispanic cultural cues were prevalent in a medium that targets Hispanics, their culture is being preserved or sustained through their medium. Whereas, hardly any Hispanic cultural cues were found in People (a non-Hispanic targeted medium), which lends support to the notion that a mainstream medium (not ethnic minority medium) would not be visually portraying Hispanic cultural cues because it is not trying to maintain Hispanic cultural pluralism. The finding for the body size of Hispanic female models also fits within the theoretical framework of DRT. A common belief is that the Hispanic culture prefers women with larger body size (Fullerton & Kendrick, 2000). This study suggests that in People en Espanol as compared to People this belief is held constant. Thus the result suggests, that the Hispanic cultural concept of female body size is endorsed in People en Espanol but not People. The context distribution also fits within the concept of DRT. Models in People en Espanol were more likely to appear in “Family” and “social” settings, both of which are cultural preferences (Gomez & Fassinger, 1994; Fullerton & Kendrick, 2000). Lastly, research question 1 also fit within DRT. Models in People en Espanol are most likely to appear in the “beauty” category followed by “domestic/food,” “celebrity” and “technical/auto” themed advertisements. In addition, models in People en Espanol were unlikely to appear in “immigration story” themed advertisements a role more typically associated with portrayals of Hispanics in mainstream culture (Fullerton & Kendrick, 2000). This may lend support to the first claim of DRT, which predicts that ethnic minority media will show its audience who they can become in the mainstream culture. By showing models in some non-traditional themes (technical/auto) People en Espanol is subtly enforcing the roles Hispanics can have in the mainstream media. As some themes are non-traditional and the cultural content and cues are present in the advertisements, both facets of DRT are met simultaneously. Someone utilizing this study as a foundation for future research should be cautious of its limitations. The study is very narrow in scope and only analyzed one type of magazine on the market If more resources were available, a broader range of magazines and possibly different mediums would have been employed. By analyzing a broader range of magazines and/or mediums one could see if the pattern is repeated across genres and mediums. This would strengthen this limited study and provide a deeper insight into the patterns that were revealed. This study also looked at very few of the cultural cues exploited by the media. To enhance this study a broader range of cultural cues such as jewelry, accent, clothing, and 25 dialect could be incorporated, depending on the medium. This would broaden the established cultural cues used in this study and provide a clearer snap shot of the types of cues used by the media. Thus offering a better representation of how the media is hoping Hispanic viewers form associations with their products in thirty seconds or less. Lastly, the themes portion of this study also represents a limitation. Coders only used latent cues to determine the types of themes in an advertisement. To establish more accurate evidence of what themes are apparent in advertisements one could analyze the attitude, stereotypes and implied themes. This would provide a more in-depth analysis of what type of messages are being conveyed to the Hispanic audience. To broaden this study an analysis of how the images in the advertisements affect Hispanics could be conducted. Participants could be asked whether or not the images they see are stereotypes. This would help researchers determine if Hispanic audience members even recognize images as representing stereotypes. This would also help researchers take the preverbal temperature on how a portrayed ethnic minority feels about “chronic misportrayal.” Another study could ask Hispanic participants to self-report their self-identity before viewing the advertisements and again after viewing them to determine if the advertisements affected their self-concept. This would allow researchers to capture a small piece to the puzzle of how advertisements showing Hispanic’s affect their self- concept. The possibilities are endless for effects studies and the current study merely serves as a springboard for further research. 26 Overall, this study lends strong support to the theoretical framework of DRT. Evidence was found for both of its claims, although support was mixed on the first claim. First, it allows minorities to acculturate into the dominant group, and second, it maintains pluralism. However, this study cannot be used to suggest how these portrayals affect the Hispanic culture. Further research would need to be conducted to detemrine how these portrayals effect Hispanic audience members. 27 Appendix A Tables One - Ten 28 Table One List of Variables for This Study Variable Range M SD Gender 1, 2, 9 Male, Female, Undistinguishable Ethnicity 1 - 5 Hispanic, African American, Asian American, Caucasian, Can’t Tell Occupation l — 8 Professional, Blue Collar/ Trade, Medical/Technical, Homemaker, Rural, Service, Safety, Other Context 1 — 5 Family, Social, Business, Nobody Else, Other Attractiveness 1-3, 99 Very, Moderate, Not at all, Not applicable Female 1 — 3 1.85 .786 Male 1 — 3 2.17 .816 Sexiness 1-3, 99 Very, Moderate, Not at all, Not applicable Female 1 — 3 1.85 .783 Male 1 — 3 2.17 .821 Body Size 1 - 6, 9 Very Underweight, Somewhat Underweight, Normal Weight, Somewhat overweight, Very Overweight, Extremely Overweight I — 6 2.63 .84 Cultural Cues l, 2, 9 Yes, No, Not Applicable - Hispanic Flags, Maps, Festivals, Textiles, Plants, Art, Food Setting 1 - 7 Professional, Blue collar, Medicalfl‘echnical, Home Indoor, Social, Natural, Other Theme 1 - 13 Food, Children/Family, Celebrity, Travel/Escape, Technical/Auto, Other, Drug/Wellness, Materialism, Cigarettes/Alcohol, Public Service 29 Table Two Calculated Scott’s Pi Intercoder Reliability of Variables Variable Pi Gender .96 Ethnicity .87 Occupation .91 Context .84 Attractiveness .84 Sexiness .82 Body Size .78 Cultural Cues .96 Setting .92 Message .96 Theme .86 Overall Pi = .9 30 Table Three Occupational Distribution of Hispanics Across Magazines People en Espanol People Occupation n % n % Professional 48 39 4 33 Homemaker 12 9 1 8 Service 2 2 0 0 Other 60 50 7 59 Total 123 100% 12 100% X2(3, N = 134) = .507 *, p < .92 *Invalid chi-square analysis due to em ty cells. P 31 Table Four Occupational Setting Distribution People en Espanol People Setting n % n % Home indoor 40 43 52 44 Social 23 24 19 16 Natural 3 l 33 46 40 Total 94 100% 29 100% X2(2, N: 211) = 2.39, p = .303 32 Table Five Attractiveness Ratings of Female Models 2 - 1.5 a 1 .. -0-People -l—People en Espanol 0.5 - o U I Hispanic All Other Ethnicities Main effect for attractiveness and magazine F (1, 357) = .38, p = .54 Results not significant Main ejfect for attractiveness and ethnicity F (1, 357) = .029, p = .86 Results not significant Interaction efi’ect of attractiveness, magazine and ethnicity F (1, 357) = 2.69, p = .10 Results not significant Sexiness Ratings of Female Models I. S - 1 .l —-o—Pe0ple —l—People en Espanol 0.5 . O t . Hispanic All Other Ethnicities Main efi’ect for sexiness and magazine F (l, 357) = .4, p = .53 Results not significant Main eject for sexiness and ethnicity F (l, 357) = .002, p = .96 Results not significant Interaction effect of sexiness. magazine and ethnicity F(1, 357) = 2.29, p = .13 Results not significant. 33 Table Six Attractiveness Ratings of Male Models 0.8 - 0.6 4 —o—People 0'4 . —l—People en Espanol 0.2 I o I I Hispanic All Other Ethnicities Main efi’ect for attractiveness and magazine F (l, 357) = .07, p = .8 Results not significant Main efl’ect for attractiveness and ethnicity F (1, 357) = .25, p = .6 Results not significant Interaction effect of attractiveness, magazine and ethnicity F (1, 357) = 4.83, p , .05 Results not significant. Sexiness Ratings of Male Models 0.8 'I 0.6 - +People 0'4 ' —I—People en Espanol 0.2 - 0 r 1 Hispanic All Other Ethnicities Main efi’ect for sexiness and magazine F (1, 357) = .005, p = .95 Results not significant Main eflect for sexiness and ethnicity F ( I, 357) = .19, p = .67 Results not significant Interaction eflect of sexiness, magazine and ethnicity F(1, 357) = 4.850, p , .05 Table Seven 34 Hispanic Cultural Cue Distribution in People en Espanol People en Espanol Queues n % Flag 1 3 Map 1 3 Festival 3 8 Textiles 3 8 Plant/landmarks 7 19 Art 3 8 Food 7 19 *Pastimes 4 11 *Colors 3 8 *Sayings 3 8 *Post hoc analysis Total 36 100% 35 Table Eight Context Distribution of Advertisements in People en Espanol People en Espanol People Context n % n % Family 44 24 29 15 Social 23 12 11 6 Business 10 5 23 12 Nobody else 96 52 116 60 Other 13 7 14 7 Total 186 100% 193 100% x2 (4, N = 379) = 14.238, p < .05 36 Table Nine Primary Theme Distribution People en Espanol People Primary Theme n % n % Beauty 65 23 41 1 1 Immigration Story 3 1 0 0 Romance 1 0 2 0 Domestic/Food 53 19 1 13 31 Children/Family 22 8 1 1 3 Celebrity 44 16 67 19 Travel/Escape 5 2 24 7 Technical/Auto 37 13 52 14 Can’t Tell 7 3 l 1 3 Drugs/Wellness 12 4 l9 5 Materialism 15 5 9 3 Ci garettes/Alcohol 10 3 8 2 Public Service 10 3 6 2 Total 284 100% 363 100% 37 Table Ten Theme Distribution Comparison People en Espanol People Primary Theme n % n % Beauty 65 24 41 12 Domestic/Food 53 19 1 13 32 Children/Family 22 8 1 1 3 Celebrity 44 16 67 19 Travel/Escape 5 2 24 7 Technical/Auto 37 14 52 15 Drugs/Wellness 12 4 l9 5 Materialism 15 6 9 3 Ci garettes/Alcohol 10 4 8 2 Public Service 10 4 6 2 Total 284 100% 363 100% x2(9, N = 623) = 46.018, p < .05 38 Appendix B Codebook & Operational Definitions 39 Coding Training & Codebook First write the advertisements number onto the code sheet. Then answer these questions. 1. Are any people present in the advertisement? Yes_(1)_ OR No (2) If yes continue through the following questions. If no proceed to question 2 What is the gender of the primary person in the advertisement? Male—(1)_, Female_(2)_, Undistinguishable (9) , 99 (not applicable) Primary person in the advertisement: A character that is very important to the advertising theme or layout, shown in the foreground usually holding or demonstrating the product. Parent is always the Primary when shown with a child under 16 years of age. What is the ethnicity of the primary person? , Asian American _(3) , Hispanic _(1)_, African American _(2) Caucasian _(4)_, OR can’t tell and not applicable _(5)_, Hispanic defined as round face, flatter nose, darker complexion What is the gender of the secondary person in the advertisement? Male_(1) , Female (2) , Undistinguishable/Not applicable (9) Secondary Person: A character that is unimportant to the advertising theme or layout, shown in the background assisting the primary person or used as decoration or context. What is the ethnicity of the secondary person? Hispanic _( 1)_, African American _(2)_, Asian American _(3)_, Caucasian _(4)_, OR can’t tell and other _(5) Not applicable _(99)_ Is anyone else in the advertisement Hispanic? Yes_( 1) (if yes please explain) No ( 2) Not applicable _(99)_ What is the occupation of the primary person in the advertisement? _( 1)_ Ptofessional _(6)_ Service _(2)_ Blue Collar/Trade _(7)_Safety _(3)_ Medical/ Technical _‘(8)_ Other _(4)_ Homemaker 99(not applicable) _(5)_ Rural Occupations Professional: Jobs such as executive, lawyer, CEO, marketing, sales, architect, secretary, assistant, or other such general jobs usually requiring dress clothes. Blue Collar/Trade: Factory worker, carpenter, builder, plumber, mechanic or other jobs such jobs Medical/ Technical: Doctor, nurse, clinician, lab personnel, computer operator, engineer, robotic engineer and other such jobs usually requiring high math and science aptitudes and often dressed in white coats or scrubs. Does not include a factory worker who operates the robotic machinery. Homemaker: A person caring for children, clcaning, cooking or otherwise acting as a caregiver. 41 Rural: A field worker, migrant worker, fishery personal or any other job that focuses on the outdoors. Excluding a nature, hunting, fishing guides or other jobs concerning service profession revolving around the outdoors. Service: Forest ranger, hunting, fishing or nature guide, restaurant worker, movie theater personnel, or other service oriented jobs. Safety: Police officer, fireman, DN R, Street cleaner or other jobs that protect the common good of man. Other: Any make believe job or one not categorized above. What is the primary person’s context in the advertisements relation to others in the advertisement? _(1)_ Family Context _(4)_ Nobody else in the _(2)_ Social Context advertisement _(3)_ Business Context _(5)_ Other Relationship 99(not applicable) Context Family Context: Includes husband and wife and any relationship between relatives, including children as well as extended family, such as aunts and uncles, grandparents, grandchildren, adopted children and foster children. (Taylor et al., 1995) Social Context; Includes friends or any other person depicted in a social setting, with the exception of family members depicted in a social context. (Taylor et al., 1995) Businem Context: A depiction of members of or workers in the same company, whose who are employed by the same company. Also, colleagues in the same profession or occupation, although they may be employed by different companies. Any relationship between employees or professionals who work together. (Taylor et al., 1995) 42 Nobody else in the advertisement: Choose this option only when one model appears by themselves in the advertisement. (Taylor et al., 1995) Other Relationship: Any other option not listed above or the relationship is unclear. (Taylor et al., 1995) If the primary person is female, how attractive would you rate her? _(1)_ Very attractive _(2)_ Moderately attractive _(3)_ Not at all attractive _ (99)_ Not applicable Attractiveness and Sex Appeal Attractiveness: How physically attractive the model is (Mastro & Atkins, 2002) Children under 16 always 99 for both. Sex Appeal: The sexiness of the model (Mastro & Atkins, 2002) If the primary person is female, what is her overall sex appeal? _(1)_ Very sexy _(2)_ Moderately sexy _(3)_ Not at all sexy _ (99)_ Not applicable If the primary person is male, how attractive would you rate him? _(1)_ Very attractive _(2)_ Moderately attractive _(3)_ Not at all attractive 43 _ (99)_ Not applicable If the primary person is male, what is his overall sex appeal? _(1)_ Very sexy _(2)_ Moderately sexy _(3)_ Not at all sexy __ (99)_ Not applicable Please rate the primary person’s body size. _(1)_ Very underweight _(6)_ Extremely overweight _(2)_ Somewhat underweight 99(not applicable)- so close only the _(3)_ Normal Weight face shows and no other information is _(4)_ Somewhat overweight given. _(5)_ Very overweight 2. Are there Cultural queues in the Advertisement? Yes (1) No (2)_ If yes answer the next series of questions. If no please proceed to question 3 Are any Hispanic flags used in the advertisement? _(1)_ yes OR _ (2)_ No, 99(not applicable) Flags: A11 non-American flags. Are any Hispanic maps used in the advertisement? _(1)_ yes OR _ (2)_ No 99(not applicable) 44 Maps: Non-American maps (i.e. not state maps, American cities, or country maps). Are any references to Hispanic festival used in the advertisement? _(1)_ yes OR _ (2)_ No 99(not applicable) Festivals: This would include depictions of references to Cino de Mayo, Spanish carnival, Spanish dancers or Day of the Dead. Are Hispanic textiles used or referenced in the advertisement? _(1)_ yes OR _ (2)_ No 99(not applicable) Are any Hispanic plants or landmarks used or referenced in the advertisement? _(1)_ yes OR _ (2)_ No 99(not applicable) ls Hispanic art used in the advertisement? _(1)_ yes OR _ (2)_ No 99(not applicable) ls Hispanic food used or referenced in the advertisement? _(1)_ yes OR _ (2)_ No 99(not applicable) If any other cultural cues are present please list Is the message (writing) in the advertisement Consistent with the cultural cues_(1)_ Offers no cultural cues _(2) 45 Offers additional cultural cues _(3)_ Consistent with the cultural cues: Offers the same cultural cues as the background of the advertisement. Offers no cultural cues: There is no reference to cultural cues in the message. Offers additional cultural cues: Offers cultural cues beyond the ones offered in the background of the advertisement. If the message offers additional cultural cues please list 3. What is the setting of the advertisement? _(1)_ Professional Setting (Taylor et al., 1995) _(2)_ Blue Collar setting (Taylor et al., 1995) _ (3)_ Medical/ Technical Setting _(4)_ Home indoor/outdoor (Taylor et al., 1995) _ (5)_ Social Setting Outside Home (Taylor et al., 1995) _ (6)_ Natural Settings (Taylor et al., 1995) _(7)_ Other Settings Professional Setting: Sales or office rooms (Taylor et al., 1995), boardrooms, or conference centers. Blue Collar setting: Factories (Taylor et al., 1995) and trade jobs such as construction, plumbing or other general labor. Medical] Technical Setting: Hospitals, clinics, medical labs, computer labs, robotic labs or engineering settings. Home indoor/outdoor: Recognizable as a residence, room, garage, yard, home or apartment, or driveway in which the primary is depicted. (Taylor et al., 1995) Social Setting Outside Home: Includes Public places, auditoriums, restaurants, movie thcaters, places were people meet for social purposes. (Taylor et al., 1995) Outdoors and Natural Settings: Forests, fields, rivers or oceans. (Taylor et al., 1995) Other: Artificial setting and any other settings not listed above (Taylor et al, 1995) 4. What is primary background color of the advertisement? _(1)_ Green _(4)_ Black _(2)_ Red _ (5)_ Blue _(3)_ Yellow _(6)_ Other color-no distinct Primary background is the main color surrounding the forefront. May be a person’s clothing. 5. What is the primary color of the text? _(1)_ Green _(4)_ Black _(2)_ Red _ (5)_ Blue _(3)_ Yellow _(6)_ Other color Primary text is the bold text. Excludes logos and taglines. 6 What is the primary theme in the advertisement? Primary theme is the person paying for the ad. _(1)_ Beauty _ (5)_ Children] family _(2)_ Immigration Story _(6)_ Celebrity _(3)_ Romance _(7)_ Travel/ Escape _(4)_ Domestic/ food _(8)__ Technical/auto 47 _(9)_ other (please list) * (l2) Cigarettes/Alcohol * (10) Drug/Wellness * (13) Public Service * (11) Materialism * Categories added Post Hoc Beauty: An advertisement focusing on outward appearances, selling beauty products (make—up, shampoo) or products which alter physical appearance. Immigration Story: A story about an immigrant who rose up to discover and grasp the “American Dream” Romance: An advertisement, which focuses on the sexual attractiveness, sex, or romance between partners. Domestic or food: An advertisement centered on cleaning products, cooking, cleaning and ones selling food. Children/Family: An advertisement focused on raising, nurturing or carrying for children or one that focuses on familial ties. Celebrity: All advertisements with celebrity endorsements (or celebrities in the ad) should be placed here. This should always be the primary theme. Travel] Escape: An Advertisement focused on escaping the riggers of this world or traveling to a destination. Technical/Auto: An advertisement focused on a technical product or automobile. Other: Non-codable or to little information to determine a theme Drug or Wellness: Prescription drugs, over the counter medicine, Supplements, Prescription glasses, and contacts Materialism: jewelry, clothes, shoes, special furniture and other decorative items Cigarettes and Alcohol: Depictions of cigarettes and alcohol — hard liquor and beer Public Service: Not-for-profit companies advertising common good 7. Please list any additional primary themes? 48 References 49 References Alaniz, M. & Wilkes C. (1995) Reinterpreting Latino culture in the commodity form: The case of alcohol advertising in the Mexican American community. Hispanic Journal of Behavioral Sciences, 1 7, 439-451. Albarran, A. B., & Umphrey, D. (1993). An examination of television motivations and programming preferences by Hispanics, Blacks and Whites. Journal of Broadcasting & Electronic Media, 37, 95-103. AdAge, Advertising Age. (2002). Top 300 Magazines, alphabetical Index. Crain Communications at AdAge.com. AHAA, Association of Hispanic Advertising Agencies. (2002). A brochure entitled: Missed opportunities: vast corporate under spending in the U .S . Hispanic market. McLean, Virginia. Budd, R. W., Thorp, R K., & Donohew, L. (1967). Content analysis of communication. New York: The Macmillan Company. Banerjee, M., Capozzoli, M., McSweeny, L., & Sinha, D. (1999). Beyond kappa: A review of interrater agreement measures. Canadian Journal of Statistics, 27, 3- 23. Czepiec, H., & Kelly, J. S. (1983). Analyzing Hispanic roles in advertising: a portrait of an emerging subculture. Current Issues & Research in Advertising, 5, 219-240. Dearing, J. W., & Rogers, E. M. (1996). Agenda-Setting. Thousand Oaks, California: Sage Publications. Faber, R. 1., O’Guinn, T. C., & Meyer, T. P. (1987). Televised portrayals of Hispanics: A comparison of ethnic perceptions. International Journal of Intercultural Relations, 1 I, 155-169. Fullerton, J.A., & Kendrick, A. (2000). Portrayal of men and women in US. Spanish- language television commercials. Journalism and Mass Communication Quarterly, 77, 128-142. Greenberg, B.S., Eastin, M.S., Hofshire, L., Lachlan, K.A., & Brownell, K. (2003). The portrayal of overweight and obese persons in commercial television. American Journal of Public Health, 93. In Press 50 Gomez, M., & Fassinger, R. E. (1994). An initial model of Latina achievement: acculturation, biculturalism and achieving styles. Journal of Counseling Psychology, 44, 205-215. Grier, S. A., & Deshpande, R. (2001). Social dimensions of consumer distinctiveness: The influence of social status on group identity and advertising persuasion. Journal of Marketing Research, 38, 216-224. Jhally, S. (1995). Image-based culture: Advertising and popular culture. In G.Dines, & J. M. Humex, (Eds.), Gender, race and class in Media: a text reader (pp. 77- 87). Thousand Oaks, California: Sage Publications. Kiousis, S., Bantimaroudis, P., & Ban, H. (1999). Candidate image attributes: Experiments on the substantive dimension of second level agenda setting. Communication Research, 26, 414-428. Krippendorf, K. (1980). Content analysis: an introduction to its methodology. Beverly Hills, CA: Sage Publications. Lafky, 8., Duffy, M., Steinmaus, M., & Berkowits, D. (1996). Looking through tendered lenses: Female stereotyping in advertisements and gender role expectations. Journalism & Mass Communication Quarterly, 73, 379-88. Mastro, D., & Atkin, C. (2002) Exposure to alcohol billboards and beliefs and attitudes toward drinking among Mexican American high school students. The Howard Journal of Communication, 13, 129-151. McCombs, M. E., & Shaw, D. L. (1993). The evolution of agenda-setting research: Twenty-five years in the marketplace of ideas. Journal of Communication, 43, 58-67. McCombs, M.E., Shaw, D. L., Weaver, D. H., & Hamm, B. F. (1999). Individuals, groups and agenda melding: A theory of social dissonance. International Journal of Public Opinion Research, I I , 2-24. Miller, R. E., & Wanta, W. (1996). Race as a variable in agenda setting. Journalism & Mass Communication Quarterly, 73 , 913- 92. Neuendorf, K. A. (2002). The content analysis guidebook. Thousand Oaks, CA: Sage Publications. Orbe, M. P. (1998). Constructing co-cultural theory. Thousand Oaks, CA: Sage Publications. Porter, W. J ., & Levine-Donnerstein, D (1999). Rethinking validity and reliability in content analysis. Journal of Applied Communication Research, 27, 258-284. 51 Riggins, S. H. (1992). The Media Imperative: Ethnic Minority Survival in the Age of Mass Communication. In R. A. White & M. Traber (Eds.), Ethnic minority media: An international perspective (pp. 1-20). Newbury Park, CA: Sage Publications. Roessler, P. (1999). The Individual agenda-designing process: How interpersonal communication, egocentric networks, and mass media shape the perception of political issues by individuals. Communication Research, 26, 666-700. 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. Taylor, C. R., Lee, J. Y., & Stern, B. B. (1995). Portrayals of African, Hispanic and Asian Americans in magazine advertising. The American Behavioral Scientist, 38, 608-621. Tharp, M. C. (2001). The marketing environment in a multicultural society. Marketing and consumer identity in multicultural America, (p. 1-29). Thousand Oaks, CA: Sage Publications. 52 i(flIj‘ygll‘llljllljliI