Abstract
This study examines the influence of acculturation and language on the emotional response of three print advertisements (English, Spanish, Code Switched) through an experimental methodology founded on the Revised Hierarchical and the Conceptual Feature Models. The advertising treatments and surveys were administered to a convenience sample of 272 respondents in a large southwestern metropolitan area from five local Hispanic churches. The Bidimensional Acculturation Scale for Hispanics was used to measure acculturation (Low, High, Bicultural). Emotional responses to the print advertisements were measured using the Emotional Quotient Scale and the condensed Reaction Profile. Results revealed some significant differences in advertising preferences. Results also suggested that a code-switched advertisement could be both culturally relevant and appropriate for reaching the majority of the Hispanic market. Implications, limitations, and recommendations for future research are discussed.
Keywords
As the United States’s Hispanic population increases, advertisers are faced with the challenge of deciding the best approach for appealing to this demographic. Previous research has focused on the portrayal of Hispanics in advertising (Ferle & Lee, 2005; Fullerton & Kendrick, 2000) and whether the language of an ad should be in Spanish or in English (Bishop & Peterson, 2010; Noriega & Blair, 2008). Research has previously identified that Hispanics exhibit higher likeability when advertisements are in their dominant language (e.g., Bishop & Peterson, 2010; Carlo-Casellas, 2002), and that a translation of a campaign designed for the English-speaking market does not always work well (Carlo-Casellas, 2002). Still, some advertisers use re-treads—ads that have been edited for the Hispanic market (Fullerton & Kendrick, 2000). This research seeks to understand whether Hispanic-advertising campaigns should remain the same, be translated or culturally adapted, and whether code-switched advertising serves as appropriate cultural adaptation. It examines the roles culture, language, and cognition play in advertising to Hispanics.
U.S. Hispanic Market
Hispanics are the largest, fastest growing minority in the United States (Witt, 2008). According to the U.S. Census Bureau (2009), the Hispanic population will represent roughly one third of the entire population by 2050. As of 2006, Hispanic growth accounted for half of the nation’s growth (24.3%)—more than 3 times the growth rate of the entire population (U.S. Census Bureau, 2006). Notably, the United States has become the third-largest Spanish-speaking country in the world (Carlo-Casellas, 2002) with a market that can be broken down into many nationalities and cultures (Gurliacci, 2004). This has led marketers to pay attention to the buying power—between US$630 billion and US$1 trillion—of this unique segment. (DiMaria, 2007).
Advertising and Culture
Culture—as a way of life or specific practices—divides people into different types of consumers (Spencer, 2006). Too often, marketers and advertisers have segmented audiences by race and ethnicity, instead of by culture (Pettigrew, 2009). Patterson and Wilkins (2008) explained, “Advertising needs to take seriously the role of culture in our lives. That means that advertising must authentically reflect the diverse voices that comprise our culture” (p. 72). Advertising is a form of social communication that reflects the standards and values of a society (Graham, 2002). It relies on culture because it is constructed by culture, and in order for an advertisement to be successful, it must have a clear message to which consumers can relate. However, the ability to identify with an advertisement is affected by identification with a specific society, culture, or race (Ueltschy & Kramf, 1997).
Advertisers must realize that the term melting pot does not accurately describe present America anymore. Although there is evidence for this widely supported homogeneity theory, assimilation is the older paradigm that suggests people adopt the new culture’s traits and submerge their own (Korgaonkar, Karson, & Lund, 2000). In the United States, minorities move toward acculturation rather than assimilation (Noriega & Blair, 2008). According to marketer, Maria Perez, “There’s just this automatic assumption that all people from all different races react to the same product the same way” (Gurliacci, 2004). Although a more significant emphasis should be placed on culture, ethnicity and race cannot be disregarded (Pettigrew, 2009), mostly because the degree to which one identifies with an ethnic group determines commitment to cultural norms, standards, and ideas (Ueltschy & Kramf, 1997).
Research on marketing to the Hispanic consumer has focused on three distinct strategies: the “change the language” approach, the “completely different” approach, and the “nothing different” approach (Faber & O’Guinn, 1991). Hispanic marketers have noted that the same language does not necessarily reach all segments of the population (Gurliacci, 2004). In some cases, Hispanics have felt offended by certain advertisements because advertisers have placed language over culture. This was the case in 2005, when Mexican pop star Thalia became the spokesperson for Hershey’s cajeta. The advertisements ran on Spanish radio and television stations across America without regard to proper definitions or segmentation. The word cajeta is solely a Mexican Spanish term. For Argentineans, cajeta has very explicit sexual meaning, and in Ecuador, very few have heard the term (Wentz, 2005). In its defense, Hershey claimed to use the term because they were targeting Mexican population—which is 67% of Hispanics in the United States. However, if a parallel were drawn, any marketing agency targeting all White Americans would never select a word that could possibly offend any segment of the population.
In 1996, however, the Colgate-Palmolive Company took a different approach, by first entering the U.S. Hispanic market and then the general market. When the company introduced Mexican-produced Suavitel fabric softener and Fabuloso hard surface cleaner to the U.S. Hispanic market, it gained a 15% market share among Hispanic consumers. The Suavitel brand took the number two slot, overtaking Snuggle (Neff, 2000). Julio Gaviria, former director of Hispanic and Asian marketing for Colgate, stressed that both brands experienced success among Hispanics because of their strong fragrance. The fragrance became the center of their advertising strategy for the general market and also led to the development of a new line of botanical-scented dish soaps that positioned the company with a 40% market share in the category, an all-time high. Although Gaviria described their strategy as experimental, the multicultural approach worked (Neff, 2000).
Acculturation
As Hispanics assimilate to life in the United States, they begin to learn a culture that is different from their own, a process called acculturation (Valencia, 1985). This process often occurs as immigrants change their behavior and attitude toward the host society as they come into constant contact with different cultures (Berry, 1997).
Acculturation occurs at both a group level and on an individual level (Berry, 2005). At the group level, it requires change in social structure and cultural practices, while at the individual level, it requires change in personal behavior. Both undergo acculturation by the interaction of two components: attitudes (an individual’s preference about how to acculturate) and behaviors (a person’s actual activities). A person in the acculturation process must make the distinction between maintaining one’s heritage, culture, identity, and participating in the larger society with other ethno-cultural groups. Individuals then choose to (1) integrate, (2) assimilate, (3) separate, or (4) marginalize. This integration process allows immigrants to become a part of the new society while still holding on to their own heritage (Graham, 2002). These individuals are recognized as bicultural and may or may not be bilingual; they internalized two cultures, and speak the language associated with each culture (Luna, Ringberg, & Peracchio, 2008).
Although biculturals speak both languages, a particular language can activate distinctive sets of culture-specific mental frames, which draw upon with different aspects of their identities (Luna et al., 2008). Notably, bicultural bilinguals have often reported feeling as though they are two separate people as they switch between languages (LaFromboise, Coleman, & Gerton, 1993). Scholars have suggested that biculturals may have separate cognitive process associated with each culture and language and use frame switching to jump between the culture-specific mental frameworks (Briley, Morries, & Simonson, 2005; Luna et al., 2008). In order for frame switching to occur, a bilingual must also be bicultural; bilinguals who are not bicultural do not have distinct cognitive frameworks (Luna et al., 2008). Still, code switching, switching between two languages within a sentence, has been shown to have an effect on persuasion. Luna and Peracchio (2005) found that code switching from the minority language to the majority resulted in greater persuasion among bilinguals than code switching from the majority to the minority language. Furthermore, the mental frames bilinguals experience may consist of entirely separate identities, perspectives and behaviors (Briley et al., 2005). As new concepts are learned, each is associated with a specific language and culture. Consequently, words that may have exact translations in another language likely have a culture-specific definition or conceptual association (Kroll & de Groot, 1997).
The constructs of acculturation are at times difficult to define because they are concerned with culture—something that no one person knows everything about (Matsumoto, 2006). Still, research has been able to identify key constructs to the acculturation process. One of the most significant signs of acculturation is language. When it comes to acculturation among Hispanics, language preference falls in four categories: Spanish only (isolated), predominantly Spanish (low acculturated), predominantly English (high acculturated), and English only (assimilated). The acculturation process can be observed across attitudes, values, behaviors, and cultural identity—in both media and language use (Ryder, Alden, & Paulhus, 2000).
Bilingual Lexicosemantic Organization
Research has suggested that language plays a role in constructing, expressing, and sharing cultural standards and beliefs (Buttjes & Byram, 1991; Ringberg, Luna, Reihlen, & Peracchio, 2010). Scholars have also noted that individuals have the ability internalize two or more cultures and speak the language(s) associated with each culture (Ringberg et al., 2010). Language is vitally linked to culture. A person’s thought process is created by internalized mental models that are influenced by a cultural model. Beyond being bilingual, an individual may also be bicultural. In the same way, an individual may be bilingual but not bicultural. In an effort to explain this information processing, two theories that complement one another have become prominent: the Revised Hierarchical Model (RHM) and the Conceptual Feature Model (CFM). Together, they can be seen as a framework by which one can study a bilingual individual’s lexicosemantic organization (Kroll & de Groot, 1997).
RHM
The RHM is a widely accepted theory in psycholinguistics that posits two levels of representation: lexical (word) and conceptual (meaning) (Luna & Peracchio, 1999). The lexical level is separated into separate compartments by language, while the conceptual level is a single compartment in which words access a common representation or meaning (Caruana & Abdilla, 2004). Bilingual individuals hierarchically arrange words and concepts that separate at a lexical level, but use a semantic system that connects across languages (Dufour & Kroll, 1995).
RHM explains why second language messages may require more effort to process (Caruana & Abdilla, 2004; Luna & Peracchio, 1999). Connections at the lexical level across words in different languages are called lexical links or word associations. Connections across languages between words and their meanings are called conceptual links. Conceptual links have been shown to be stronger in a bilingual individual’s first language when compared with the second language, an imbalance that remains over time (Caruana & Abdilla, 2004). A message in a bilingual individual’s first language is more easily related to its relevant semantic storage and often easier for the individual to process (Kroll & Stewart, 1994). An advertisement in a secondary language in which the text lacks congruency with the image can often demand too much from a bilingual individual’s cognitive resources. To compensate, the individual will turn their attention to less demanding tasks and disregard the advertising message (Luna & Peracchio, 1999).
CFM
While building on RHM, the CFM has suggested that the same word in different languages may have different interpretations (Ringberg et al., 2010). Specifically, words in each language have characteristics that activate conceptual features that connect to corresponding concepts (Luna & Peracchio, 1999). Ringberg et al. (2010) noted that CFM provides a cognitive explanation on the role mental models play in frame switching, a rather recent concept in literature (Noriega & Blair, 2008). According to Noriega and Blair (2008), frame switching not only occurs in language but culturally as well, where certain symbols can cue different personalities and cultural cues. For example, in English, the word “friend” might be associated with McDonald’s and honesty. The same word in Spanish, “amigo,” may be associated with honesty and male (Luna & Peracchio, 1999). Similarly, the word “beach” may be associated with the words picnic and sun, while its Spanish translation, “playa,” may be associated with the Caribbean and palm trees (Ringberg et al., 2010). These mental models are obvious to the individual who possesses them, for the individual treats that mental model as if it were an obvious fact of the world.
Studies using both the RHM and the CFM have found that first language messages are more likely to be processed at the conceptual level than second language messages (Luna & Peracchio, 1999). Previous research has also found that bilinguals tend to rely on imagery more than monolinguals (Paivio & Lambert, 1981). The language in which the consumer-related concept is presented activates a language-specific schema (Luna & Peracchio, 1999). Mandler (1985) argued that this produced congruity-based affect, in which a comparison occurs when an individual receives new information. If this new information fits with prior knowledge schemas, then a mild positive affective reaction occurs. In short, CFM can be used to conceptualize the effects of language on affective reactions to advertising (Luna & Peracchio, 1999).
Some advertisers have picked up on the schema activated by certain words in Spanish and have incorporated them into their strategy, known as code-switched phrasing. This technique emphasizes alternating between two or more languages in order to appeal to the dual identities of bilinguals and biculturals (Bishop & Peterson, 2010). One of the first examples of this was an advertisement for AFLAC insurance that appeared in Hispanic magazine that read, “Twenty Million Hijas [Daughters] Are Covered by AFLAC. Is Yours?” (Luna & Peracchio, 1999). In 2008, an advertisement in Latina magazine used a tagline with both Spanish and English words (Bishop & Peterson, 2010). The advertisement stated, “In Arizona Spring Has a Way of Warming Los Corazones [Hearts].”
Hypotheses and Research Questions
“Unlike immigrants from earlier in the history of the United States, Hispanics today can participate in society while still retaining strong aspects of their Latino culture” (Anderson, 2009). Highly acculturated Hispanics are typically second, third, fourth, or fifth generation, and very rarely first generation. Population statistics support this notion. In fact, 39% of native-born Hispanics (second, third, fourth, and fifth generation) only speak English at home, and only 4.1% of foreign-born Hispanics only speak English at home (Pew Hispanic Center, 2010). According to the RHM, bilinguals likely process first language, or dominant language, messages at the conceptual level than second language messages (Luna & Peracchio, 1999). Although highly acculturated Hispanics speak Spanish, the aforementioned studies would predict this group dominates English better than Spanish. In light of this information, the first hypothesis is as follows:
Consistent with accommodation theory, research has also suggested that Hispanics with low-acculturation levels prefer advertisements completely in Spanish (Luna & Peracchio, 2001; Ueltschy & Kramf, 1997). Because lowly acculturated Hispanics prefer Spanish, and according to RHM and the CFM, certain schemas are activated in the first language that increase the ease of processing (Noriega & Blair, 2008), the following hypothesis is proposed:
More recent research has explained that bicultural bilinguals have the ability to access dual mental frameworks or identities, and frame switching occurs not only in language but also in culture (Luna et al., 2008). Research in cultural frame switching has suggested that, “language and symbols can cue different personality and cultural characteristics in bicultural-bilingual people” (Noriega & Blair, 2008, p. 71). CFM has suggested that depending on the language in which they are presented, words evoke different affective reactions (Kroll & de Groot, 1997; Luna & Peracchio, 1999). Because bicultural bilinguals experience frame switching, there is a possibility that by accessing both mental frameworks through code switching, the emotional response to an advertisement is affected. Therefore, the following hypothesis is proposed:
The research also sought to understand the following research questions:
Method
As reflected in the literature, language and culture are vitally linked as one is used to express the other (Buttjes & Byram, 1991; Ringberg et al., 2010). The goal of this research was to better understand the affect a bilingual consumer takes to an advertisement based on language and the implications language applies through culture and cognition. This research tested a Hispanic consumer’s reaction to an English print advertisement, a translated Spanish print advertisement, or a partially accommodating, culturally relevant code-switched advertisement.
Variables
The independent variables in this experiment were the language of the advertisement (English, Spanish, or Code Switched), and the demographic variables. The independent demographic variables included: gender, age, education level, generation, and length of stay in the United States. The moderating variable was the respondent’s acculturation level (High, Low, Bicultural). The dependent variable is the respondents’ emotional reaction to the advertisement.
Advertising Treatments
The print advertisement selection and design were guided by the literature. An all-purpose Mr. Clean product was chosen in light of the success of Colgate-Palmolive’s products and strategy. Advertisements focused on scent as the main sales pitch, since Gaviria, former director of Hispanic and Asian marketing for Colgate, stressed that both brands experienced success because of their strong fragrance (Neff, 2000). Three variations of the same print advertisement were professionally created. The English print advertisements stated, “Bring home a new tropical scent.” The print advertisement was a direct translation that read, “Llévate a casa un nuevo aroma tropical.” The third print advertisement was a code-switched Spanish and English language advertisement that stated, “Bring a new tropical scent a su casa.” All advertising treatments were identical in visual appeal, and only varied in the language of the text.
Instruments
Acculturation level was measured using the Bidimensional Acculturation Scale (BAS) for Hispanics. The BAS calculates acculturation through 12 Likert-type questions that measure three areas: language use, linguistic proficiency, and electronic media (Marin & Gamba, 1996). The BAS has evidence of reliability, validity, and internal consistency (Davis, Engel, & Gurin, 2010; Marin & Gamba, 1996). Scholars have noted that the BAS largely relies on language-based items (Cabassa, 2003), and this serves as a limitation of the instrument. However, as this experiment focuses mainly on the use of language and culture, this limitation is not a grave issue. The BAS allow for biculturation and this “will help researchers and practitioners in better understanding the processes Hispanics go through as they acculturate” (Marin & Gamba, 1996). Ultimately, the BAS was chosen as it works best with Mexican Americans and Central Americans (Cabassa, 2003; Marin & Gamba, 1996). The experiment was conducted in a southwestern metropolitan area, which is 17.2% Hispanic, and 80% of that Hispanic population is Mexican (U.S. Census Bureau, 2010).
Response to the advertisement was measured using Wells Emotional Quotient Scale (EQ) and a condensed version of the Reaction Profile. The EQ measures general emotional reactions to an advertisement through 12 Likert-type questions (Wells, 1964). The condensed Reaction Profile consists of three 5-point semantic differential questions that indicate whether the advertisement was good/bad, pleasant/unpleasant, or favorable/unfavorable. Previous research in the cultural aspects of advertising has often used the EQ and the Reaction Profile (Graham, 2002; Leach & Liu, 1998).
The experiment was conducted using print advertisements, and both the EQ and Reaction Profile were developed using print advertisements (Wells, 1964; Zinkhan & Fornell, 1985). Although different scales can predict attitude toward a brand, the Reaction Profile is superior in measuring purchase intent when compared with other scales (Zinkhan & Fornell, 1985). Research has shown that attitude toward an advertisement influences attitude toward brand and may predict purchase intentions (MacKenzie, Lutz, & Belch, 1986). The survey also included a section of demographic data to ensure a representative sample. These demographic questions included gender, age, origin, education level, generation, and length of residency in the United States.
Sample and Procedure
A convenience sample was used. Participants of 18 years of age and older were recruited from five local Hispanic churches. Subjects were given one of the randomly assigned print advertisements and the survey. Respondents were given the choice to complete the survey in Spanish or in English in order to maximize participation and improve responses. The generation of participants was determined through self-selection, where five generations and operational definitions were given as options.
The study sought to find whether significant relationships existed between the independent variables of type of advertisement (English, Spanish, Code Switched), and acculturation level (High, Low, Bicultural), and the dependent variable of the respondents’ emotional reaction to the advertisement. The data were collected from 283 Hispanic adults from congregations in a large southwestern metropolitan area during March 2012 (see Table 1). Eleven extensively incomplete surveys were removed from the sample, for a total sample of 272. Survey respondents were 39.7% male and 60.3% female. The average respondent’s level of education was relatively low, with 53.3% indicating high school or lower. First generation respondents, those born in a country other than the United States represented 58.3% of the sample. Second generation Hispanics accounted for 28.8% of the sample. In addition, the majority of respondents (55.9%) were from Mexico.
Demographics of Population Sample.
The “Other” category includes the Philippines, Spain, and several other countries of origin.
In order to examine the effects of acculturation among the three advertisements (English, Spanish, Code Switched), the sample was divided into three acculturation groups (High, Low, Bicultural) using the BAS for Hispanics. Of all survey respondents, 25.8% were lowly acculturated (n = 70), 33% were highly acculturated (n = 90), 25.8% were bicultural (n = 70), and 15.4% (n = 42) could not be classified. The 15.4% that could not be classified were respondents who scored less than 2.5 in both the Hispanic and non-Hispanic domains—this group was then classified as “non-culturally associated.”
Results
Upon conducting 3 × 3 × 2 repeated measured ANOVA, results did not reveal main effects for the advertising treatments by acculturation (F(1, 266) = .328, p < .721), gender (F(1, 266) = 1.266, p > .261), or as an interaction (F(2, 262) = .090, p > .914) between the variables for the EQ. Similarly, the Reaction Profile did not reveal any interaction effects for the advertising treatments by acculturation (F(6, 260) = .89, p > .501). Levene’s test did not indicate a violation of the equal variance assumption, (F(11, 260) = 76, p > .679). Thus, none of the hypotheses were supported. However, upon further exploration, other results were found to be statistically significant.
The data indicated significant differences in preference for the basic advertisement type on the EQ (F(6, 260) = 4.17, p < .017). Participants preferred the Spanish-only ad (M = 3.25), followed by code switch (M = 3.20), and then English (M = 3.02). Overall, advertisement type accounted for 3.1% of the variation, indicating a very weak positive relationship (r = .176). The advertisement types for the Reaction Profile, however, were not statistically significant. Upon looking at the specific results, the data showed that Hispanics in the not culturally classified category (M = 3.45) had the highest Emotional Quotient mean for Spanish advertisements. Overall, the Spanish print advertisements (M = 3.25) were significantly favored over English print advertisements (F(2, 260) = 4.17, p < .008). In addition, men (M = 2.41) and women (M = 2.27) did not have any significant differences (t(270) = .980, p > .245) in terms of preference for the Spanish-only ad.
As expected, the English print advertisements were favored most by the highly acculturated (M = 3.267) segment. Generally, the English print advertisements (M = 3.021) were the least favored of all three advertising treatments. This was statistically significant against both Spanish (F(2, 260) = 4.17, p < .008) and code-switched (F(2, 260) = 4.17, p < .032) advertisements. Finally, men (M = 3.02) were less receptive to the English print ad than women (M = 3.24; t(270) = −3.19, p < .001).
The highly acculturated segment (M = 3.27) favored the code-switched advertisements more than other groups, followed by the lowly acculturated segment (M = 3.197), than by the bicultural segment (M = 3.196) and lastly by the not culturally classified segment (M = 3.147). Overall, the code-switched advertisements (M = 3.203) were favored over English print advertisements (F(2, 260) = 4.17, p < .032), but not more than Spanish advertisements (F(2, 260) = 4.17, p < .032). See Table 2 for a complete breakdown. Men (M = 2.90) were significantly less receptive than women (M = 3.04) to the code-switched ads (t(270) = −2.61, p < .001).
Emotional Quotient Scores by Advertisement Treatment and Acculturation Level.
In addition, the data indicated that acculturation level played a significant factor (F(6, 260) = 3.08, p < .028) in how the ads were interpreted on the EQ (see Table 3), with highly acculturated individuals having the most positive reaction to the ads (M = 3.28) followed by non-classified (M = 3.19), bicultural (M = 3.15), and lowly acculturated (M = 3.02). Acculturation level accounted for 3.4% of the variance in the EQ and explained a very weak positive relationship (r = .184). Since acculturation level was statistically significant and the variable had three levels, a post hoc test was conducted. The Tukey–Kramer method found there was a statistically significant difference between the Emotional Quotient mean for lowly (M = 2.99) and highly (M = 3.28) acculturated Hispanics (F(3, 260) = 3.08, p < .003). Highly acculturated Hispanics (M = 3.28) favored the advertisements the most, followed by the bicultural segment (M = 3.16), then by the not culturally classified segment (M = 3.12), and lastly by the lowly acculturated segment (M = 2.99).
Emotional Quotient Scores by Acculturation for all Ad Treatments.
Discussion
In today’s diverse market, it is essential to understand not only the market itself but also the motivations behind it. A discussion on advertising mishaps showed that, in some cases, Hispanics have felt offended because advertisers have placed language over culture. The literature has shown that ethnic identity is core to consumer attitudes and behaviors in purchase decisions (Palacios, 2011). Because the Hispanic market is so heterogeneous, Hispanics may be more effectively segmented by acculturation (Graham, 2002; Kara & Kara, 1990; Ueltschy & Kramf, 1997).
In an effort to understand the link between language and culture, this study utilized an experimental methodology founded on the RHM and the CFM. The goal was to better understand the affect a bilingual consumer takes to an advertisement based on language and the implications language applies through culture and cognition. This research tested a Hispanic consumer’s reaction to an English print advertisement, a translated Spanish print advertisement, and a partially accommodating, culturally relevant code-switched advertisement. Although the study did not reveal any interaction effects—and consequently none of the hypotheses were supported—statistical analysis did indicate main effects. Advertisement type and acculturation level had statistically significant effects on the emotional response to the advertisement.
Hypothesis 1 stated that Hispanic Americans with a high-acculturation level would have a more positive emotional reaction to the print advertisement. Although the lack of interaction effects between acculturation and advertisement type (English, Spanish, Code Switched) rendered this hypothesis unsupported, the statistically significant main effect for advertisement type did show some support for this notion. The highly acculturated segment did have the highest emotional response for the English print advertisements. This suggests that perhaps with a larger sample, the difference would have been greater and thereby possibly supporting this hypothesis.
Hypothesis 2 stated that Hispanic Americans with a low-acculturation level would have a more positive emotional reaction to the translated Spanish print advertisement. As previously stated, the lack of statistically significant interaction effects between acculturation and advertisement type makes this an unsupported hypothesis. Unlike the circumstances in Hypothesis 1, the main effect for type of advertisement did not provide any support for this hypothesis. The not culturally classified segment exhibited the most likeability for the Spanish print advertisement, followed by the highly acculturated segment, then by the bicultural segment, and lastly by the lowly acculturated segment. These results would indicate that Hispanics hold on to their native language, but the Spanish language alone may not be enough to reach the lowly acculturated Hispanic market. Previous research found that bilinguals tend to rely on imagery more than monolinguals (Paivio & Lambert, 1981). Perhaps for the lowly acculturated who are in the process of becoming bilingual, imagery becomes significantly more important. The lack of a model using the product may have had a detrimental effect on the emotional response to the advertisement for this segment.
Hypothesis 3 stated that bicultural Hispanic Americans would have a more positive emotional reaction to the targeted/code-switched Spanish and English print advertisement. Again, the lack of statistically significant interaction effects makes this hypothesis unsupported. The statistically significant main effect for the code-switched advertisement was further analyzed. This showed that that the highly acculturated segment exhibited the highest emotional response, followed by the lowly acculturated segment, the bicultural segment, and lastly by the not culturally classified segment. Notably, however, the results between the emotional responses for the code-switched print advertisements had the lowest standard deviation of the three advertisement types. This indicates that perhaps a code-switched advertisement can effectively target the majority of Hispanics. This offers some support that code-switched messages are culturally relevant.
The demographic sample was relatively evenly distributed. Interestingly, 89% of the highly acculturated segment described themselves as first generation, and 51% of the lowly acculturated segment described themselves as second generation. Furthermore, 73% of the bicultural segment classified themselves as first generation. Research has noted that the process of acculturation occurs more slowly among Hispanics (Anderson, 2009). This would suggest that acculturation is a process that is unique for each person and occurs in a more complex way than the BAS for Hispanics can measure.
In fact, this study came across a set of circumstances that suggest change is necessary in the BAS. In the sample, 42 respondents could not be classified under the BAS standards. These respondents had a low affinity for both cultural domains. The majority of respondents in this category were first and second generation under 34 years of age. The demographic data for the not culturally classified segment closely resembled the lowly acculturated segment, but this group’s overall emotional response to the advertisements was between that for the lowly acculturated segment and the bicultural segment. As the BAS relies heavily on media use and linguistic proficiency, this uncategorized segment may have less of an affinity for media use and acculturate through other routes.
Implications
As the results of this study seem to suggest, highly acculturated Hispanics can be reached with a mainstream English language advertisement designed for the Anglo consumer. Some research has stressed that it is extremely important to advertise to low or bicultural Hispanics in Spanish (Koslow, Shamdasani, & Touchstone, 1994) while others have found that bicultural and highly acculturated Hispanics prefer English (Graham, 2002; Ueltschy & Kramf, 1997). This study has offered a middle ground. As bilingual lexicosemantic research has suggested, language is intrinsically linked to culture. As a person can be both bilingual and bicultural, it is easy to understand the mixed results of this study. The mental models that the bicultural consumer processes are influencing them in numerous ways that allow them to create associations with products and cultures. As RHM suggests, bicultural consumers create a hierarchy within their mental models of word preferences and cues. Often, to reduce cognitive processing and utilize the pre-established lexical links, the bicultural consumer prefers the ad that is most closely related to their native tongue—as the data suggest. This supports RHM in that it shows the level of acculturation is directly linked to the desire for word association in second language advertising messages. The mainstream, English language advertisement was adapted to contain a code-switched message that the Anglo consumer could understand, but the Hispanic consumer found culturally relevant. By providing the code-switched material, the lowly acculturated became more comfortable with a message and product. Using a code-switched advertisement can be both culturally relevant and appropriate for reaching the Hispanic market.
Limitations
Several limitations should be considered when analyzing the results of this study. One limitation of this study was the respondents’ education level. Although the level of education variable was relatively well distributed, there is more variance in the level of education than the statistics show. This occurs because the standards of education vary between countries. All together, 58.3% of respondents were first generation, which indicates that more than half of the sample is from countries in Central and South America. Upon conducting the surveys, a number of respondents required clarification for negatively phrased questions, and some survey responses seemed to contradict one another. Other measures should be taken to accommodate the more lowly educated segment. Perhaps the scales should be adapted to three or four points and disregard negatively phrased questions. In addition, many older Hispanics are not familiar with completing questionnaires and surveys of this type.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.