This paper was presented at the Association for Education in Journalism and
Mass Communication in San Antonio, Texas August 2005.
If you have questions about this paper, please contact the author
directly. If you have questions about the archives, email
rakyat [ at ] eparker.org. For an explanation of the subject line,
send email to
[log in to unmask] with just the four words, "get help info aejmc," in the
body (drop the "").
(Feb 2006)
Thank you.
Elliott Parker
====================================================================
General Interest Magazine Language Preference among Hispanics
Anthony Galvez
Instructor
College of Mass Communications
Texas Tech University
Box 43082
Lubbock, TX 79409-3082
806-742-6500 ext. 245
[log in to unmask]
In 2003, the Census Bureau reported that one in eight people (13.3%
of the total population) in the United States were Hispanic (Ramirez
& de la Cruz, 2003). The growing number of Hispanics in the United
States has caught the attention of corporate America. During the
1990s marketers started building media franchises targeted at
Hispanic audiences. NBC's purchase of the Telemundo television
network, for example, showed the growing attention given to this
emerging demographic. The merger involving Univision, the nation's
largest television network targeted to Hispanics, and the Hispanic
Broadcasting Company (HBC), the nation's largest radio company
targeted to Hispanics, also showed the value being placed on reaching
out to Hispanics (Ahrens & Williams, 2003). These mergers, however,
do little to prove that Hispanic targeted media has garnered the
power to attract major advertising revenue (Napoli, 2002). One of the
reasons for questioning the power of one media entity to attract a
majority of the Hispanic populations is the issue of language. Due
to the major differences in the Hispanic populations and due to level
of acculturation, socio-economic status, education and country of
origin, it has become apparent that understanding media language
preference and the forces that drive it is critical in creating
ethnic-specific messages. This study seeks to answer questions
related to these types of language choices made by media channels and
their effect on Hispanics of different acculturation levels. To
answer these questions, this study identified the literature related
to acculturation and media preferences of Hispanics before testing a
variety of propositions related to different language types and
different acculturation levels.
Though past research (Delener & Neelankavil, 1990; Gibson, Hudson &
Melanson, 1999; Greenberg, Burgoon, Burgoon & Korzenny, 1983;
Hernandez & Newman, 1992; Melanson & Hudson, 1996; Rios & Gaines,
1998; Roslow & Roslow, 1980) has looked at language preference and
media use, little has been done to identify how to create print media
that will grab this unique audience. Valdes (2000) argues that
language choice, cultural nuances, habits, needs, and wants must all
be understood before the Hispanic market can effectively be targeted.
Simply translating English language articles into Spanish will only
attract those readers fluent in Spanish and can often lead to
confusion if not culturally correct (Valdes, 2000). Furthermore,
researchers (Gibson, Hudson & Melanson, 1999; Greenberg et al., 1983;
Hernandez & Newman 1992; Melanson & Hudson, 1996; Rios & Gaines,
1998) have discovered Hispanics do not always prefer Spanish language
media. Variables including, age, number of years in the United
States and acculturation all play a part in language preference
(Greenberg et al., 1983; Hernandez & Newman, 1992; Melanson & Hudson,
1996; Rios & Gaines, 1998). To date, however, there has been no
empirical test to determine if there is a relationship between the
way language is presented in print and Hispanics' perceptions of
magazine articles.
Purpose of the Study
The purpose of this study is to determine the relationships between
language presentation style and Hispanics' perceptions of news value,
affinity toward presenting media and understanding content. More
specifically, magazine articles written in English, Spanish, and a
mix of English with Spanish will be tested. Also, because previous
research has indicated acculturation as a predictor of media language
preference, this study will test level of acculturation as a
predictor of magazine language preference for bilingual Hispanics.
HISPANIC MEDIA USE AND LANGUAGE PREFERENCE
Researchers have looked into Hispanic media usage and preferences
(Delener & Neelankavil, 1990; Deshpande, Hoyer & Donthu, 1986;
Gibson, Hudson & Melanson, 1999; Greenberg et al., 1983; Melanson &
Hudson, 1996; Roslow & Roslow, 1980). Media language choice has also
been examined (Gibson, Hudson & Melanson, 1999; Greenberg et al.,
1983; Hernandez & Newman, 1992; Melanson & Hudson, 1996; Rios &
Gaines, 1998). Hispanic media usage and motivation has been a
subject of research as well (Albarran & Umphrey, 1993,
1994). Finally, there has been research into Hispanic media usage in
regard to its role in cultural identity maintenance (Deshpande, Hoyer
& Donthu, 1986; Jeffres, 2000; Rios & Gaines, 1998).
Overall, previous research has indicated a complex link between
culture, identity and media use. Language preference has also proved
to be determinate on many different factors. This chapter will
outline some of the variables associated with media use and language
preferences.
Media Use
The most comprehensive study examining Hispanic media habits, to
date, was conducted by Greenberg et al. (1983). They conducted
telephone surveys in seven southwestern cities. The research focused
on eight variables:
1. Ethnic identity and language usage.
2. Media use and access.
3. Media evaluations.
4. Media functions.
5. Sources of information.
6. Media content preferences.
7. Newspaper and television image.
8. Demographics.
The surveys revealed a statistically significant difference between
Hispanic and Anglo newspaper readership. Across the board Anglo
respondents reported reading the newspaper more often than Hispanics.
The research indicated no significant difference in time spent
watching local and national television news between Hispanics and
Anglos. Overall, Hispanics reported watching more TV than Anglos.
There was a significant difference in the percentage of Hispanics and
Anglos reporting to have recently read a magazine. Only 60% of the
Hispanics surveyed said they had read a magazine in the last week as
opposed to 73% of Anglos. Time spent with Spanish language media was
also gauged by the researchers. They found Hispanics spent 29.2
minutes a week reading Spanish language newspapers and Hispanics
spent, on average, 43.4 minutes reading Spanish language
magazines. The number of minutes spent, on average, with Spanish
language radio (557.3) and television (391.1) were much higher than
the amount of time Anglos reported. Hispanics also reported listening
to Spanish language tapes/records 108.1 minutes a week.
The findings indicated Hispanics who predominately spoke Spanish
were more likely to spend time with Spanish language
media. Hispanics who predominately spoke English in general
preferred English language media. Overall, there was a trend where
Hispanics used print media much less than Anglos and were inclined to
watch more television. The findings suggest that media heavily laden
with news and information (newspapers and magazines) is utilized less
than media heavy with entertainment (television and radio).
The findings of the Greenberg et al. (1983) study were further
validated by Delener and Neeleankavil (1990). Their findings
indicated Hispanics preferred television over other types of
media. The research also indicated that Hispanics listen to the
radio more than the rest of the population, especially when there is
access to Spanish stations. Not surprisingly, magazine and newspaper
usage was lower among Hispanic respondents when gauged against
African Americans and Anglos.
The overall findings of the past research into Hispanic media usage
indicate low usage of print media and high usage of electronic
media. In general the current body of research strongly suggests
that Hispanics use radio and television more often than newspapers or
magazines. It is still important, however, to understand which
variables effect print media language preference.
Language Preference
Advertisers have shown great interest in minority language
preference. Research into Hispanic language preference in
advertising, however, has been inconclusive at best. Faber and
O'Guinn (1991) found there was little difference in attitude toward a
commercial or brand when Hispanics were exposed to both English and
Spanish language advertisements. Hernandez and Newman (1992) examined
the body of research available at that time. They collected research
dealing with Hispanics and broadcast media and print media. They
concluded that the best way to reach the Hispanic market is by
speaking to them in their dominant language noting, level of
acculturation plays a role in the transition from Spanish to English
as the dominate language. Conversely, a study conducted by Koslow,
Shamdasai, and Touchstone (1994) found advertiser's choice to use
Spanish language copy resulted in a positive affect toward the ad.
Researchers have also looked beyond advertising and have examined
the general media language preferences of Hispanics. Melanson and
Hudson (1996) surveyed nine hundred and seventy three Hispanics in
Texas. Questions were asked to gauge Hispanic media use and language
preference. The results of the survey indicated Hispanics preferred
broadcast media. Interestingly enough, only 20% of the respondents
reported reading a newspaper in the previous month. Age was
identified as an indicator of media language preference with
Hispanics under the age of 35 reporting to speak English in the
home. Older respondents preferred to speak Spanish or Spanish and
English mixed. Ethnic identity and its effect on media language
preference was also examined. Rios and Gaines (1998) predicted that
low Latino heritage would mean low Spanish language media use as well
as unfavorable attitudes towards the Spanish language. Bilingual
Latino heritage will produce positive attitudes towards Spanish
language and high exposure to general media. High Latino heritage
will use Spanish language media more often than the other two groups.
Results of the research indicated low Latino Heritage individuals had
a significantly less favorable attitude toward Spanish and were
exposed less to Spanish language media. Bicultural Latino heritage
respondents had a more favorable attitude toward Spanish and were
exposed to both English and Spanish language media. High Latino
heritage respondents had high levels of exposure to Spanish language
television and radio. They also reported having low access to
English language media and low use of newspapers in either language.
One of the reasons low acculturated Hispanics prefer Spanish
language media can be explained by Jeffres' (2000) findings that
indicate ethnic media is a tool for cultural maintenance. A panel
study was conducted from 1976 to 1992. Participants from the Midwest
were given surveys every 4 years. Thirteen ethnic groups were
identified and utilized for the study. The initial survey had 768
respondents. By 1992 the number of respondents was down to 157 due
to issues of mortality and mobility. Ethnic media use positively
predicted strength of ethnic identity over time. The same was not
true, however, for ethnic identity predicting ethnic media
use. Still the importance of media use as a tool for cultural
maintenance was apparent.
While the majority of the studies have provided a wealth of
information, most of the methodologies have been ineffective in
validating links between changes in language and media preferences.
METHODOLOGY
This study examined if language presentation would cause Hispanics
to have differing opinions of the overall news value of general
interest articles. The relationship between language presentation
and understandability was also tested to see if the In-Culture
marketing approach would lead to a greater understanding of the
articles. This study was also constructed to test if language
presentation would cause Hispanics to experience a certain level of
affinity with the articles presented.
Recent trends in magazines targeting Hispanics have been to: 1) use
only Spanish text 2) use a mix of Spanish and English and 3) to print
entire articles in both Spanish and English. According to Johnson
(2000) publishers have for some time made available Spanish language
versions of popular magazines (Cosmopolitan, Good Housekeeping and
Harper's Bazaar). Magazines such as Latina and Latina Style, more
directly marketed to Hispanics, print articles in English while
mixing in Spanish. At times Latina magazine will print the same
article completely in English and then completely in Spanish. The
argument can be made that Spanish only publications serve Hispanics
new to the United States and magazines utilizing a mix may be the
mortar necessary to keep more highly acculturated Hispanics close to
their culture (Jeffres, 2000). Nonetheless, each one of these
language presentation styles is a different way of presenting the
same information. The choice to use a certain language presentation
style is made in order to reach a certain audience and to create a
certain affect in the reader. This study looked into the overall
effect differing language presentation styles had upon Hispanics and
whether an affinity was felt for the article. Language presentation
and subject's indication of news value and understandability were
also tested.
Acculturation refers to the process where by people from a minority
culture blend into a dominant culture. The process of acculturation
can be detected in changes of language, preferences and
behaviors. The tendency is for the minority population to take on
some of the traits of the dominant culture. As contact with the
dominant culture increases so should the amount of
acculturation. Those individuals with a greater reliance on the
language, preferences and behaviors of the dominant culture are said
to be more highly acculturated than those who still show a greater
reliance on the tenements of the country from which they immigrated.
For the purpose of this study, affinity is defined as a feeling of
kinship or positive social relationship with an object. It is argued
that Spanish language and mix articles would create a bond between
the reader and the article. Past studies (Rios & Gaines, 1998;
Jeffres, 2000) have identified media as a tool of cultural
maintenance by which relationships are created between media and the
reader. Such a bond may be similar to the bond a chef may feel with a
trade magazine dedicated to new standards in the food industry or a
person interested in technology and a television program dedicated to
the diffusion of new technology.
News value was defined as the feeling of overall importance of the
articles content. The measures newsworthy and informative were used
to determine the articles news value. Newsworthiness was defined as
the articles ranking among other issues requiring attention by the
news media. Informative was defined as the ability of the article to
relate information to the reader.
Understandability was defined as the ability of the text to easily
relate the intended message. Its overall function in this research
was to gauge if bilingual subjects would have a greater understanding
of English or Spanish language text. It was felt that a greater
understanding of the text would coincide with a greater new value and
affinity.
Another variable examined by this research was language preference.
As previous research has shown, (Ball-Rokeach, 1985; Faber et al,
1985; Ball-Rokeach & De Fleur, 1976) individuals approach the media
for specific gratifications. Relationships can be developed between
the individual and certain media. To determine if such relationships
can be built through type of language, preference was examined by
this research. By testing for language preference, this study will
add to the previous literature, which indicates a greater reliance on
English as acculturation increases. Language preference was
identified by testing the subjects' level of affinity for the article
through measures identifying level of enjoyment and if the respondent
found the article interesting. Language preference was also
identified by the respondents' feeling the article was both
understandable and newsworthy. For the purpose of this study,
language preference is gauged by and identified as a degree of
newsworthiness, affinity and understandability.
Hypotheses and Research Questions
When True et al. (2000) tested acculturation as a predictor of
Mexican-American's perceptions of advertising they found degrees of
acculturation can predict ad perception. Their conclusions were that
highly acculturated Hispanics would respond to stimuli in a similar
fashion to Anglos. When the advertisements were crafted so as to
mirror the respondent's level of acculturation, he/she was more
likely to respond positively. Ueltschy and Krampf (1997) tested
acculturation as a predictor of Hispanic's language and model
preference in advertisements. Their findings revealed low
acculturated Mexican-Americans preferred ads in Spanish. The
opposite was true for highly acculturated Hispanics. They preferred
advertisements in English.
Johnson (2000) found one of the many functions of ethnic media is to
preserve the native culture through language and pride. For that
reason, the assumption can be made, those members of an ethnic
minority still seeking a certain level of assimilation, while holding
on to their culture, will prefer media reflecting the duality of
their existence. Because of this duality, as well as the findings of
Ueltschy and Krampf (1997) and True et al. (2000) the following
research question and hypotheses were constructed:
RQ1: Is there a relationship between language presentation and a)
news value b) affinity c) understanding of general interest magazine articles?
RQ2: Is there a relationship between level of acculturation and a)
news value b) understanding of general interest magazine articles?
H1: Respondents with a low level of acculturation will prefer
general interest magazine articles written in Spanish.
H2: Respondents with a high level of acculturation will prefer
general interest magazine articles written in English.
H3: Respondents with a middle level of acculturation will prefer
general interest magazine articles written in both English and
Spanish as is dictated by the concept of In-Culture marketing.
Methodology
The experiment consisted of testing levels of acculturation and
language preference. The overall design of the experiment was a 3 X
3 factorial design with three levels of acculturation and three
different language presentation styles being tested. The stimulus
material was two sets of articles printed in Spanish, English, and
English with Spanish words mixed in and level of acculturation. Two
articles were used so that any effect created by the content of the
articles would be controlled for and also because the researcher
expected a small sample size. The dependent variables were news
value, affinity and understandability. The stimulus material and
questionnaire was examined by six bilingual Hispanics in order to
determine if there were any problems with the material not previously
identified by the researcher. It was determined by the researcher
to seek a higher level of external validity, than offered by a lab
environment, by conducting an intercept survey. It should be noted,
however, that this type of experiment will result in less external
validity than a true field experiment in which the stimulus material
would be tested at the location where respondents would normally be
exposed to it.
Hispanic subjects were recruited using both, a mall intercept type
data collection method as well as a snowball (Rios & Gaines, 1998)
type of data collection technique in two small and two medium-sized
west Texas cities. Respondents were given a packet, asked to read
two articles and fill out the attached questionnaire.
Stimuli
The stimuli for this experiment were two articles written in a style
consistent with general interest magazines. The first was a travel
article touting the merits of taking a cruise as a standard vacation
alternative. The second article was about the demise of the payphone
and the new era of cellular phones.
The article about cell phones was four paragraphs long with three
hundered words. The article about vacations was three paragraphs
long with three hundred and seven words. Each article was crafted so
as to indicate that the selection was only the beginning of a larger
article. The articles were laid out to look as if they were taken
directly from a magazine using Quark Xpress.
A Spanish teacher at an east Texas junior high school translated each
article into Spanish. The articles were then reviewed and edited by
two Mexican-Americans fluent in both English and Spanish during a
pretest of the research instrument. No problems were identified with
the translation.
The mix version of the two articles was constructed based on several
articles in the February 2003 issue of Latina. The articles were
identified as using In-Culture type articles because they were
primarily written in English and had Spanish words mixed in. The
articles were compared to other articles in the magazine and were
judged to be typical of the general writing style presented
throughout. The frequency of Spanish words in the articles was
recorded. It was determined that a Spanish word appeared on average
once out of every 100 words. The mix versions of the two articles
were then constructed using the formula of Spanish to English as
previously identified (1 per 100 words).
In all there were six different packets randomly distributed to the
respondents. The first packet contained both articles completely in
English with the vacation article first and the cell phone article
was second. The second packet contained both articles in English but
the order the articles appeared was reversed. The third and forth
packets were exact duplicates of the first two with the exception of
the language. Both articles in this group were written using
primarily English with a mix of Spanish words throughout. The fifth
and sixth packets were identical to the first two except for the
language used in the articles. In this group the articles were
written in Spanish.
Participants and Data Collection
The population of the study was Hispanics in west Texas over the age
of 18. Both males and females were encouraged to participate. Of the
two hundred and forty completed questionnaires, two hundred and
seventeen were deemed usable for the study.
Using a mall intercept type method, subjects for the experiment were
first recruited at a concert celebrating Cinco de Mayo in a medium
sized west Texas community. A table was set up and signs were hung
indicating participants in the research could register for a
drawing. The winner of the drawing would win a $100 gift card
provided by a local grocery store. A volunteer who was not
affiliated with the research collected the information for the
drawing separately from the research data. All efforts were made to
ensure there was no connection between the research and the drawing
information.
After looking at the initial sample size, it was determined more
data was needed. The event was a discount night at a local theme
park. Once again a table was set up and respondents were given the
opportunity to enter a drawing for a $50 gift card provided by a
local grocery store.
Due to a very small turn out at the second event, a third intercept
was set up. This time using the same incentive as the second data
collection, participation was solicited at a local grocery store in a
predominately Hispanic neighborhood. Once again the level of
participation in the experiment was low despite the incentive.
After another look at the sample size, it became apparent more was
needed. A switch in strategy led to the use of the technique
utilized by Rios and Gaines (1998). The technique is known as
snowball sampling. Adult volunteers were recruited through local
contacts of the researcher. The packets were delivered to the
volunteers at their places of work or at home, and the packets were
picked up at a later date.
Research Instrument
The questionnaires for all packets were identical and were written in
English. Because, the focus of the survey was on bilingual
Hispanics, the choice was made not to use both English and Spanish
versions of the ARSMA-II scale. Also, because English-only speaking
researchers did recruitment, testing with a Spanish version of the
questionnaire would have created translation problems when
administering the experiment.
At the end of each article were three questions about the article. A
five-point likert scale was used to gauge responses (1= I totally
agree, 5 = I totally disagree). Question #1 asked if the article was
newsworthy. Question #2 asked if the article was interesting and
question 3 asked if the article was informative. Four additional
questions were asked after the second article. These questions were
used to analyze the respondent's ability to understand the articles
and enjoyment level. The questions were as follows:
1. I felt this article was newsworthy
2. This article was interesting
3. I felt this article was informative
4. The article about cell phones was easy to understand.
5. The article about cruise ships was easy to understand.
6. I enjoyed the article about cell phones.
7. I enjoyed the article about cruise ships.
The questions asking if the articles were newsworthy and informative
were meant to gauge overall newsworthiness. The questions about
enjoyment and interest were constructed to gauge affinity and the
question asking if the article was easy to understand gauged
understandability.
Following the questions about the articles was the 30 question
ARSMA-II (Acculturation Rating Scale for Mexican Americans) scale
gauging level of acculturation. Answers were gauged using a 5-point
likert scale with 1 being not at all and 5 being almost always. The
ARSMA II scale, as created by Cuellar, Arnold and Maldonado (1995),
is commonly used by researcher to rank Hispanics level of
acculturation. Questions are asked in regard to language
preferences, cultural associations and self-identity. ARSMA-II tests
language preferences with questions such as, "I speak Spanish" and "I
enjoy watching English language television." Cultural associations
are gauged with questions such as, "My contact with the USA has
been…" and "My family cooks Mexican food." An example of
self-identity would be "I like to identify myself as Mexican
American" and "I like to identify myself as an American."
Questions are divided into two sub-categories; the Anglo orientation
scale (aos) and the Mexican orientation scale (mos). There are
thirteen aos questions and 17 mos questions. A score is given for
each category and the mos is subtracted from the aos to give a
composite score. The score are then lumped into one of the following
five groups:
Level I. Very Mexican oriented
Level II. Mexican oriented to approximately balanced bicultural.
Level III. Slightly Anglo oriented bicultural
Level IV. Strongly Anglo oriented
Level V. Very assimilated; Anglicized.
For the purpose of this study it was decided to regroup the
respondents into thirds based upon their ARSMA-II score. The groups
were then labeled as:
Group I: Low acculturation
Group II: Medium acculturation
Group II: High acculturation
The reason for condensing the groups was to better facilitate a 3x3
factorial design. Also, the reduction of groups helped to account
for problems with low participation rates. It was also decided that
breaking the groups into absolute thirds would help to even out the
distribution of the groups.
At the end of the questionnaire demographic data was collected.
Respondents were asked to identify their age, gender, religious
preference, amount of education, where they received their education
and generation based on how long their family has been in the United States.
RESULTS
An analysis of the population of this study revealed a majority of
the respondents were female (n = 129, 59.4%). Respondents ranged in
age from 18 to 68 with a Median age of 30. Education level of those
responding was widely dispersed with a majority of the respondents
reporting completion of at least grade 9 (see table 1). Of the 197
respondents reporting where they attended school, 193 attended school
in the United States and 4 attended school in Mexico. Generation in
the United States was gauged revealing, a fairly even distribution
with exception to 1st generation which was only 5.8% of those
answering this question (see table 2). Generation was determined by
the birthplace of the respondent's, parents and grandparents.
Utilizing SPSS, descriptive statistics were run to identify the mean
score of the five measures used test the level of affinity, news
value, and understanding for both articles. They were further broken
down by type of language presented and level of acculturation (see table 3).
For the cell phone article, all acculturation levels reported that
the Spanish version was less newsworthy. The same was also true for
the cell phone article and the interesting measure. As for the
informative measure, the low acculturated group reported a lower mean
score for the Spanish language presentation than the other two
language presentation styles. The remaining two groups found the
Spanish language presentation style less informative. The mean
scores for the understand measure and the cell phone article
indicated that the low acculturation group found the In-Culture
presentation style more understandable and the remaining two groups
once again ranked the Spanish language style as being less
understandable. Overall mean scores for the enjoy measure indicated
that all three groups found the Spanish language version of the
articles less enjoyable.
For the Cruise article similar trends were revealed. Respondents in
the low and middle acculturation groups reported a lower mean score
for the English and In-Culture articles than for the Spanish. Those
respondents in the high acculturation group found the In-Culture
version to be less newsworthy. The low acculturated group found the
In-Culture presentation style to be less interesting and the medium
and high acculturation group found the Spanish version to be less
interesting. All three levels of acculturation reported that the
Spanish language presentation was less informative. The same trend
was also seen with all three acculturation groups reporting that the
Spanish language presentation was less understandable. And for the
enjoyable measure all three groups reported a higher mean score for
the Spanish language presentation.
Within the ANOVA model, repeated measures within subjects were run to
identify any interaction or main effect between the two story
scenarios. There was no effect found between the two story scenarios
for all measures except for the Informative measure. A
Student-Newman-Kuels (SNK) post hoc test revealed that for the cell
phone article there was a difference between the Spanish and English
versions of the articles but no difference between the mix version of
the article and the Spanish and the English versions of the articles.
For the cruise article there was a statistically significant
difference between the Spanish version of the articles and the mix
and English versions but no difference between the mix and English
versions of the articles. Because there was no effect found for the
remaining measures, subsequent analysis was performed by running
ANOVAs on the collapsed data of the two articles with the exception
of the informative measure. The stimulus material was not collapsed
into a single group for analyzing the informative measure.
RQ1: Is there a relationship between language presentation and a)
news value b) affinity c) understanding of general interest magazine articles?
A relationship was found between language presentation and news
value. Analysis revealed a statistically significant difference
between type of language presented and respondents reporting that the
are articles newsworthy F(2, 207) = 7.49, p = .001 (see table 4). An
SNK post hoc test revealed the significant difference was between the
Spanish language version of the article and English articles as well
as the mix version of the articles. Mean scores indicated that
respondents found both the mix version and the English version more
newsworthy than the Spanish version. There was no difference,
however, between the English version and the mix version of the articles.
For the cell phone article there was no significant difference F(2,
319) = 2.31 p = .102 found between the English, Spanish and mixed
language versions of the stimuli and respondents finding the article
informative. The cruise article, however, did show a statistically
significant difference between the different versions of the articles
and informative (see table 3). Student-Newman-Keuls (SNK) post hoc
test revealed the difference for informative was between the Spanish
language articles and the English language article with no difference
between the mix article and Spanish or English. A look at the mean
scores for the cruise article indicated that respondents found the
English language version of the article more informative than the
Spanish version but not significantly more informative than the mix version.
As for affinity, there was a statistically significant difference
between language presentation and respondents finding the article
interesting F(2, 215) = 6.16, p = .003 (see table 5) and respondents
finding the article enjoyable F(2, 217) = 12.81, p = .001 (see table
6). Once again a SNK post hoc test showed the difference was between
the Spanish language articles and both the English language articles
and mix articles. For both measures, mean scores indicated that low
acculturated Hispanics preferred the mix version to the English
version. Mean scores also indicated that the respondents had a
greater affinity towards the English and mix than towards the Spanish
version of the articles.
There was also a statistically significant difference between
language presentation and understandability F(2, 307) = 10.13 p =
.001 (see table 7). An SNK post hoc test indicated the difference
for understandability was between the Spanish language articles and
both the English language and In-Culture articles.
RQ2: Is there a relationship between level of acculturation and a)
news value b) understanding of general interest magazine articles?
Overall, for RQ2 there was no statistical difference found between
level of acculturation and rating of the articles news value. The
measure for newsworthy was not statistically significant F(2, 207) =
.00, p = .993. Looking at the articles individually, for the
informative measure and the cell phone article there was no
significant difference F(2, 319) = .26, p = .772. For the cruise
article there was also no difference F(2, 306) = 1.47, p = .993.
There was also no statistically significant difference for level of
understanding (F(2, 307) = .97 p = .380) in general interest magazine
articles. It was not expected that a difference in level of
acculturation would result in a lower news value rating because of
the overall general nature of the articles presented.
H1: Respondents with a low level of acculturation will prefer
general interest magazine articles written in Spanish.
H1 was not supported by the data analysis. Analysis[1] revealed no
statistically significant difference between level of acculturation
and affinity for the articles presented. Affinity was measured by
testing the interest measure (F(2, 215) = 1.57, p = .210) or the
enjoyment measure F(2, 217) = 1.49, p = .228. It was predicted that
Hispanics ranked in the lower level of acculturation would prefer
Spanish language articles. Overall, low acculturated Hispanics ranked
their enjoyment (see table 8) and interest (see table 9) of all three
versions of the articles in the middle.
H2: Respondents with a high level of acculturation will prefer
general interest magazine articles written in English.
H2 was also not supported by the data analysis. It was predicted that
a higher level of acculturation would result in higher affinity
towards English language articles. Analysis revealed highly
acculturated Hispanics did not have a greater affinity for the
English language articles based on the interest measure (F(2, 215) =
1.57, p = .210) or the enjoyment measure (F(2, 217) = 1.49, p = .228)
used to gauge affinity.
H3: Respondents with a middle level of acculturation will prefer
general interest magazine articles written in both English and
Spanish as is dictated by the concept of In-Culture marketing.
There was no evidence to support H3. Respondents in the middle
acculturation group showed no greater preference for the mixed
language articles based on the interest measure (F(2, 215) = 1.57, p
= .210) or the enjoyment measure (F(2, 217) = 1.49, p = .228) used to
gauge affinity.
DISCUSSION AND CONCLUSION
The data analysis revealed that there was a relationship between
language presentation and how Hispanics rank the news value, affinity
and understandability of general interest magazine
articles. Subjects were more favorable to the articles written in
English and the mix of English and Spanish (In-Culture). These
findings were consistent with previous studies (Melanson, Hudson &
Watts, 1995; Melanson & Hudson, 1996) that have identified English as
the language of choice for many Hispanics. For all three measures,
the preference was for the articles in English and in English and
Spanish mixed (mix). Respondents consistently had a negative opinion
toward the Spanish only articles.
When breaking down the measures used for testing news value (news
worthy and Informative), one inconsistency did appear. For the
Informative measure, there was a statistically significant difference
between how respondents ranked the English and mix articles, and how
they ranked the Spanish only version for the cruise article, which
was consistent with the all other measures. For the cell phone
article, however, the difference appeared between the Spanish and the
English version of the article and not between the mix and the other
two articles. In other words respondents felt that the English
version of the cell phone article was more informative than the
Spanish version. The mix version, however, was not ranked either more
or less informative than the English and Spanish versions of the
article. This difference for the cell phone article may be
explained by taking a look at the time period in which the study took
place. During the time of the study, legislation had been passed
allowing for cell phone users to switch service providers with out
loosing their current cell phone number. The change in the
regulations had received a large amount of press thus leading to a
lower perception of the stories informative value. Nonetheless,
overall mean scores indicated that medium and highly acculturated
Hispanics felt the cell phone article in Spanish was less
informative. Low acculturated Hispanics, ranked the mix cell phone
article lowest. These findings indicate that when magazine
publishers attempt to reach the Hispanic audience they may want to
reevaluate their own perceived notions of the news value of Spanish
and mix article. If seeking to gain the attention of those Hispanics
who only speak or read Spanish, however, there is still no
alternative to Spanish language text.
For the purpose of this study, affinity was defined as a liking
and/or an inclination towards an object. Utilizing the enjoy and
interesting measures as an indication of affinity, the data revealed
that there was a statistically significant difference between the
Spanish language version of the articles and the English and mixed
language articles. People of all levels of acculturation reported
that they found the Spanish version less interesting than the mixed
language and English version of the articles. The same trend emerged
for the enjoy measure. The implications of these findings suggest
that print media written in English and mixed language will receive a
greater affinity from the Hispanic community. These findings,
however, can only be generalized to the bilingual community.
For the understand measure, respondents reported that the Spanish
language version was not as understandable as the English and mixed
language versions of the articles. Mean score showed that low and
high acculturated Hispanics reported a higher level of understanding
for the mixed language than the English language version. Yet there
was no statistically significant difference between the two language types.
As for the use of the In-Culture marketing approach to creating media
intended for the Hispanic market, though the findings indicated no
statistically significant difference between the mixed language and
English versions of the articles, overall the mean scores reported
for medium and high acculturation subjects indicated that Hispanics
found the mixed language articles less newsworthy, understandable and
had less of an affinity towards them. The differences of mean
scores, though not statistically significant, do little to support
the idea that using Spanish words mixed into a predominantly English
text will create an article more attractive to Hispanics.
Overall, little evidence was found to support previous research
(Rios & Gaines, 1998; Ueltschy & Krampf, 1997) indicating level of
acculturation could predict language preference. The assumption was
that those Hispanics reporting a lower level of acculturation would
prefer Spanish language media and those with a higher level of
acculturation would prefer English language media. The lack of
support for the previous acculturation studies may be due to the fact
that this study did not include Hispanics who only spoke Spanish. The
vary nature of this study was to find out the preference of bilingual
Hispanics. These findings did, however, support the previous
research (Hernandez & Newman, 1992) indicating that the best language
to use when marketing to Hispanics is the language they are most
comfortable with.
Acculturation's Role in Preference
For the editors and publishers, the findings of this study suggests
that when trying to reach the bilingual Hispanics, level of
acculturation may only play a small role in language
preference. Once a certain level of English proficiency is achieved
by Hispanics, print media in English and/or mixed language can be
relied upon to get the message across. Print media predominantly in
Spanish, however, will have a negative effect on the bilingual
Hispanic audience. And as the data showed there may be little
incentive for publishers to seek out the In-Culture marketing method
for print media.
Limitations of the Study
One of the greatest limitations of this study was the inability of
the researcher to sample the population of Hispanics who only speak
Spanish or who have a very limited grasp of the English language. By
using only the English version of the ARSMA- II questionnaire it was
impossible to get the responses of the Spanish language only
population. The purpose of this study was to get a sense of what the
bilingual Hispanic population would prefer. After all we know that
those who can only understand Spanish text would prefer it to any
other language.
Overall, Hispanics were reluctant to participate in the study. The
snowball method of recruitment was used when it became apparent that
the intercept method was not working. Where as the snowball method
proved to be more effective, it was also much more time
consuming. Because of this limitation the data was collected over an
eight month period of time. The snowball method of data collection
also limits the overall randomness of the sample.
Future Research
There has been a great deal of research looking into media portrayal
of Hispanics, media preference and media language preference. To
date, however, there have been few studies examining the
effectiveness of In-Culture marketing towards Hispanics. Future
efforts should be made to empirically test the overall effectiveness
of In-Culture marketing. Research should also be conducted to test
how In-Culture marketing is received by different subsets of the
Hispanic community, such as the Puerto Ricans, Cubans and South
Americans. Qualitative research should also be conducted to better
define the best method for creating the In-Culture marketing message.
As the Hispanic population in the United States continues to grow,
so should research into reaching this critical audience. Continued
research into the Hispanic segment of our society can only help in
the diffusion of important political, health and social messages.
REFERENCES
Ahrens, F. and Williams, K. (2003, August 11). Spanish-language
media expand. Washington Post, p. A 01.
Albarran, A. B., & Umphrey, D. (1993). An examination of television
motivations and program preferences by Hispanics, Blacks, and
Whites. Journal of Broadcasting & Electronic Media, 37, 95-103
Albarran, A. B., & Umphrey, D. (1994). Marketing cable and pay cable
services: Impact of ethnicity, viewing motivations, and program
types. Journal of Media Economics, 7, 47-57.
Ball- Rokeach , S. (1985). The origins of individual media-system
dependency: A sociological framework. Communication Research, 12, 485-510.
Ball- Rokeach , S. & DeFleur , M. (1976). A dependency model of
mass-media effects. Communication Research, 3, 3-21.
Cuellar, I., Arnold, B. & Maldonado, R. (1995). Acculturation rating
scale for Mexican Americans-II: A revision of the original ARSMA
scale. Journal of Behavioral Sciences, 17, 275-304
Delener, N., & Neelankavil, J. P. (1990). Informational sources and
media usage: A comparison between Asian and Hispanic subcultures.
Journal of Advertising Research, 30, 45-52.
Deshpande, R., Hoyer, W. D., Donthu, N. (1986). The intensity of
ethnic affiliation: A study of the sociology of Hispanic consumption.
Journal of Consumer Research, 13, 214-220.
Faber, R., Reese, S., & Steeves , H. (1985). Spending time with the
news media: the relationship between reliance and use. Journal of
Broadcasting & Electronic Media, 29, 445-450.
Faber, R. J. & O'Guinn, T. C. (1991). Does translating language
translate to more effective advertising?: Hispanic viewers
"perceptions of english versus spanish language ads," in Proceedings
of 1991 conference of American Academy of Advertising, Rebecca
Holoman ed., Reno, NV: the Academy.
Gibson, R., Hudson, J. C. & Melanson, W. (1999). Broadcast media
usage and preferences of Hispanics in three large southwestern
markets. Southwestern Mass Communication Journal, 15, 1-13.
Greenberg, B. S., Burgoon, M., Burgoon J. K. & Korzenny, F. (1983).
Mexican Americans & the Media. Norwood, N.J.: Ablex Publishing Corp.
Hernandez, S. A. & Newman, L. M. (1992). Choice of English vs.
Spanish language in advertising to Hispanics. Journal of Current
Issues and Research in Advertising, 14, 1-11.
Jeffres, L., (2000). Ethnicity and ethnic media use. Communication
Research, 27, 496-535.
Johnson, M. A. (2000). How ethnic are U.S. ethnic media: The case of
Latina magazines. Mass Communication & Society, 3:2, 229-248.
Koslow, S., Shamdasani, P. N. & Touchstone, E. E. (1994). Exploring
language effects in ethnic advertising: A sociolinguistic
perspective. Journal of Consumer Research, 20, 575-585.
Melanson, W. W., Hudson, J. C. & Watts, L. (1995), Hispanics prefer
English language media. In Communicating Cultural Chaos: Proceedings
of the Ninth Annual Red River Communication Conference, Fargo, North
Dakota, 65 - 70.
Melanson, W. W. & Hudson, J. C. (1996). Survey results of Hispanic
media usage, language preferences and perceptions of advertising. In
Gary B. Wilcox (Ed.) Proceedings of the 1996 Conference of the
American Academy of Advertising, Vancouver, BC, 200-201.
Muyskin, P. (2000). Bilingual speech: A typology of code-mixing.
Cambridge, Ma.: University Press.
Napoli, P. (2002). Audience valuation and minority media: An
analysis of the determinants of the value of radio
audiences. Journal of Broadcasting & Electronic Media, 46, 169-184.
Ramirez, R. R., de la Cruz, G. P., (2003). The Hispanic population in
the United States: March 2002, Retrieved March 16, 2004, from
http://www.census.gov/ prod/2003pubs/p20-545.pdf
Rios, D. I., & Gaines, S. O. Jr. (1998). Latino media use for culture
maintenance. Journalism and Mass Communication Quarterly, 75, 746-761.
Roslow, P. & Roslow, S. (1980). How U.S. Latins use radio: they
listen more than Non-Latins. Journal of Advertising Research. 20.
Skiba, R. (1997). Code switching as a countenance of language
interference. The Internet TESL Journal, 3 (10). Retrieved Feb. 16,
2004 from http://iteslj.org/Articles/Skiba-CodeSwitching.html
True, S. L., Winsor, R. D., Capozzoli, E. A., & Tudor, R. K., (2000).
Degree of consumer acculturation and perceptions of advertising: A
study of the Mexican-American market. TSU Business & Economic Review.
24, 6-13.
Ueltschy, L. & Krampf, R. F. (1997). The influence of acculturation
on advertising effectiveness to the Hispanic market. Journal of
Applied Business Research, 13, 87-101.
Valdes, I. M. (2000). Marketing to American Latinos: A guide to the
in-culture approach. New York: Paramount.
Table 1. Distribution of Education Level of Respondents.
Grade level Frequency Percent Valid
Percent Cumulative
Elem-6 4 1.8
1.9 1.9
7-8 10 4.6
4.6 6.5
9-12 93 42.9 43.1 49.5
1-2 yrs college 60 27.6 27.8 77.3
3-4 yrs college 22 10.1 10.2 87.5
Graduate/higher 27 12.4 12.5 100.0
Total 216 99.5 100.0
Table 2. Generation in the United States.
Generation Frequency Percent Valid Percent Cumulative
1st 12 5.5 5.8 5.8
2nd 43 19.8 20.8 26.6
3rd 46 21.2 22.2 48.8
4th 48 22.1 23.2 72.0
5th 58 26.7 28.0 100.0
Total 217
Table 3. Means of Measures for Language Presentation and Acculturation.
Article
Language Level of
Acculturation
Spanish Mix English Low
Mix High
Cell article
Newsworthy 3.17(1.23) 2.68(1.24) 2.40(1.15) 2.80(1.28) 2.82(1.32) 2.63(1.12)
Interesting 3.14(1.23) 2.61(1.12) 2.60(1.12) 2.69(1.18) 2.96(1.29) 2.72(1.09)
Informative 2.97(1.33) 2.72(1.21) 2.58(1.26) 2.76(1.36) 2.85(1.28) 2.66(1.18)
Understand 2.83(1.43) 2.04(1.30) 2.33(1.39) 2.26(1.33) 2.49(1.39) 2.25(1.44)
Enjoy 3.29(1.25) 2.36(1.09) 2.56(1.16) 2.71(1.18) 2.80(1.35) 2.67(1.17)
Cruise article
Newsworthy 3.40(1.22) 3.07(1.15) 2.85(1.32) 3.13(1.29) 3.00(1.32) 3.18(1.25)
Interesting 3.26(1.32) 2.80(1.14) 2.71(1.33) 2.91(1.29) 3.06(1.33) 2.79(1.23)
Informative 3.36(1.29) 2.93(1.10) 2.74(1.30) 2.97(1.27) 2.87(1.27) 3.17(1.23)
Understand 2.96(1.31) 2.19(1.29) 2.33(1.39) 2.43(1.39) 2.63(1.27) 2.41(1.44)
Enjoy 3.41(1.17) 2.86(1.13) 2.68(1.29) 2.85(1.32) 3.09(1.19) 3.01(1.18)
Note. Numbers in parenthesis are standard deviations. (1 = I totally
agree, 5 = I totally disagree)
Table 4. Mean Distributions of Acculturation and Language for
Newsworthy Collapsed.
Mean SD N
Spanish Mix English Spanish Mix English Spanish Mix English
Acculturation
Low 3.17 2.74 2.88 1.07 .95 1.07 32 17 21
Medium 3.50 2.85 2.50 1.04 1.10 1.10 20 23 24
High 3.26 3.04 2.54 1.15 .92 .78 17 27 27
Note. (1 = I totally agree, 5 = I totally disagree)
Table 5. Mean Distributions of Acculturation and Language for
Interesting Collapsed.
Mean SD
N
Spanish Mix English Spanish Mix English Spanish Mix English
Acculturation
Low 2.95 2.62 2.76 1.05 .91 1.19 32 17 21
Medium 3.50 3.02 2.69 1.22 .95 1.18 20 23 24
High 3.29 2.61 2.54 1.15 .77 .92 17 28 27
Note. (1 = I totally agree, 5 = I totally disagree)
Table 6. Mean Distributions of Acculturation and Language for Enjoy
Collapsed.
Mean SD N
Spanish Mix English Spanish Mix English Spanish Mix English
Acculturation
Low 3.19 2.34 2.55 1.13 .93 1.06 32 19 21
Medium 3.48 2.76 2.72 .97 .95 1.30 20 27 23
High 3.47 2.66 2.59 1.13 .78 .93 18 28 27
Note. (1 = I totally agree, 5 = I totally disagree)
Table 7. Mean Distributions of Acculturation and Language for
Understand Collapsed.
Mean SD
N
Spanish Mix English Spanish Mix English Spanish Mix English
Acculturation
Low 2.45 2.11 2.40 1.25 1.26 1.22 32 19 21
Medium 3.27 2.46 2.08 1.11 1.17 1.18 20 27 24
High 3.25 1.79 2.26 1.30 .99 1.46 18 28 27
Note. (1 = I totally agree, 5 = I totally disagree)
[1] The results of H1, H2 and H3 were based on the analysis of the
interest and enjoyment measures. The F values for all three are,
therefore, the same.
|
