If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Examine acculturation and gender on intention to eat a healthful diet among Latino adolescents using the Theory of Planned Behavior.
Design
Secondary analysis of data set and condensed version of the Short Acculturation Scale for Hispanics (SASH).
Setting
Data collected from 34 randomly selected high schools in San Bernardino, CA.
Participants
265 Latino high school adolescents.
Main Outcome Measures
Effects of acculturation and gender on variables of the Theory of Planned Behavior on intention to eat a healthful diet.
Analysis
Multiple regression analysis examined acculturation/gender differences, and modifications on the prediction of intention. General linear modeling determined differences across gender and acculturation groups.
Results
Females had stronger intention, more positive attitude, and greater subjective normative influence. Females indicated feeling healthy and looking good and males indicated good athletic performance as contributors to eating healthfully. Mother was influential for both genders, and stronger for females. Siblings were influential for less acculturated males, and friends were influential for highly acculturated females. Less acculturated adolescents had stronger intention to eat healthfully, more tolerance to give up liked food items, and more support and encouragement.
Conclusions and Implications
Professionals need to take into account gender and acculturation differences when making dietary recommendations for Latino adolescents.
The potential impact of acculturation on dietary practices for Latino adolescents, especially those of high school age, is not well documented. Few dietary studies with Latinos examine effects in this age group, despite substantiated research indicating that as adolescents increase in grade level, their negative diet behaviors increase in frequency.
Unhealthful dietary behavior is a major contributor to the rising rates of obesity among this Latino adolescent population and requires careful examination to reduce the persistence of obesity into adulthood.
Obesity rates among Latinos are disproportionately high when compared to whites. National samples show clear evidence that Latino children have higher rates of obesity than white children.
According to data from the California Health Interview Survey, California Latino adolescents are more overweight than any other ethnic group in the state, at 37%, compared to a close 35% for African-American, 23% white, and 16% Asian adolescents.
Adolescent obesity has detrimental effects on many health and social factors, including: early onset high cholesterol, high blood pressure, type 2 diabetes, social stigmatization, cancer, persistence of obesity into adulthood, atherosclerosis, and premature mortality.
It is estimated that the development of preventive measures now (such as healthful dietary consumption) that help to normalize weight by as little as 5%-10% will save the state of California $6-$13 billion over a 5-year time span.
Though some positive dietary patterns are present among some Latino adolescents (greater than 5 servings of fruits and vegetables daily, and low-fat diet consumption), significant negative dietary patterns include higher soda and fast-food consumption compared to white and Asian adolescents.
More California teens consume soda and fast food each day than five servings of fruits and vegetables [Policy Brief]. UCLA Center Health Policy Research,
Los Angeles, CA2005
Differences in energy, nutrient, and food intake in a US sample of Mexican-American women and men: findings from the third National Health and Nutrition Examination Survey, 1988-1994.
Researchers state that Latino adolescent acculturation into American culture causes dietary health practices to deteriorate, shifting from plant-based food (corn tortillas, beans, fruit, fruit juices) to processed convenience food (chips, soda), with little evidence to support such conclusions.
Differences in energy, nutrient, and food intake in a US sample of Mexican-American women and men: findings from the third National Health and Nutrition Examination Survey, 1988-1994.
Differences in energy, nutrient, and food intake in a US sample of Mexican-American women and men: findings from the third National Health and Nutrition Examination Survey, 1988-1994.
Diet and food insufficiency among Hispanic youths: acculturation and socioeconomic factors in the third National Health and Nutrition Examination Survey.
Overall, higher acculturation status for Latinos is positively associated with increased undesirable behaviors like substance abuse, smoking, and alcohol use and a desirable behavior around usage of preventive health screening services; but it is negatively associated with birth weight in infants, healthful dietary behavior, and exercise frequency.
One study involving Latino adults examining gender and acculturation influences on dietary knowledge and beliefs reveals that acculturation for males is more positively associated with increased nutrition knowledge.
Females and those who are highly acculturated have higher dietary fat avoidance scores compared to males and those who are less acculturated. Highly acculturated males also have lower belief scores (that dietary change could lead to better health) compared to highly acculturated females. A study with middle school adolescents finds acculturation positively associated with reduced low-fat food and higher fast-food consumption.
More research with older adolescents (high school age) that examines the influence of acculturation on dietary behaviors is needed to prevent the impending persistence of obesity into adulthood. The purpose of this study was to perform a secondary analysis of data collected in 1999 to examine the potential influence of acculturation and gender on constructs of the Theory of Planned Behavior, as related to a healthful diet among Latino high school adolescents.
Theoretical Model
The Figure displays the Theory of Planned Behavior variables with the inclusion of the study variables proposed here in this research. The Theory of Planned Behavior is an established framework to help explain and predict various adolescent behaviors.
The theory postulates that psychosocial aspects affect a person's attitude toward performing behaviors, their subjective norm influence, and their personal sense of perceived behavioral control.
When a person has a more favorable attitude, subjective norm, and perceived control, he or she is more likely to have stronger intention to perform a behavior. Intention and perceived behavioral control are hypothesized to predict the behavior, as some behaviors may have limited volitional control. The Theory of Planned Behavior is a valuable theory in which to examine the relationship of acculturation, a highly complex process.
FigureTheory of planned behavior as modified and reprinted from Backman D. Haddad EH. Lee JW. Johnston PK. and Hodgkin GE. Psychosocial predictors of healthful dietary behavior in adolescents. J of Nutr Educ Behav, 2002; 34(4): 184-193.
In a 1999 study, Backman and colleagues used the framework of the Theory of Planned Behavior to identify factors related to dietary behavior in a group of multi-ethnic high school adolescents from San Bernardino, CA.
Influence of gender and ethnicity on psychosocial predictors of healthy dietary practices and exercise behaviors in adolescents [doctoral dissertation].
Influence of gender and ethnicity on psychosocial predictors of healthy dietary practices and exercise behaviors in adolescents [doctoral dissertation].
An initial 121-item questionnaire assessed the main theoretical constructs of intention, attitude, subjective norm, perceived behavioral control, behavioral beliefs and outcome evaluations, normative beliefs and motivation to comply, and control beliefs and perceived facilitation, each measured using 3 separate items. These items were measured using a 10-point bipolar Likert scale ranging from 0 to 9, arranged randomly to minimize response bias. Internal consistency Cronbach α of the items assessing intention, attitude, subjective norm, and perceived behavioral control were 0.87, 0.83, 0.74, and 0.73, respectively.
A follow-up study questionnaire, given 1 month after the initial questionnaire, measured actual dietary practices during the past month using a 67-item food frequency questionnaire. The outcome dietary behaviors (healthful diet) and food frequency items were determined during pilot-testing with 50 San Bernardino high school students using a 24-hour recall analysis. The recall data and food frequency questionnaire were provided to the researchers by a Loma Linda University nutrition faculty member conducting nutritional research in the same San Bernardino high school population. The instruments were also verified by a 4-person panel of public health nutrition faculty.
The 67-item food frequency questionnaire examined the outcome dietary behaviors of: “eating 2-3 pieces of fruit daily; consumption of 2-3 cups of raw or cooked vegetables daily; eating low-fat or fat-free dairy products daily; adding very little fat (salad dressings, butter, oils, or lard) to food; eating small servings of meat, poultry, and fish (about the size of a deck of cards); and limiting fried food, snack food, and fast food.” These behaviors were then organized into the study variables of daily total calories, percentage calories from fat, and servings of fruits and vegetables with the guidance of nutritional experts and faculty from Loma Linda University.
Study results showed distinct ethnic and gender differences, and ethnicity and gender interactions for many dietary behavior outcomes. Latina females showed stronger intention, more positive attitude, and greater subjective normative influences than Latino males for eating a healthful diet, whereas males reported significantly higher daily consumption of calories. Although the condensed version of the Short Acculturation Scale for Hispanics (SASH) was included in the questionnaire and the geographical area had a high proportion of Latino immigrants, researchers did not examine the effects of acculturation for the Latino ethnic subgroup.
The purpose of this study was to conduct a secondary analysis of data collected by Backman and colleagues to examine the potential effects of acculturation and gender on the original study variables.
Methods
Subjects
This study was a secondary analysis of an existing data set, collected prospectively in 1999, consisting of 265 Latino participants.
Influence of gender and ethnicity on psychosocial predictors of healthy dietary practices and exercise behaviors in adolescents [doctoral dissertation].
The original data set consisted of a total of 780 adolescents, ranging in age from 14 to 19, sampled from 34 randomly selected science classes at 4 high schools in San Bernardino, CA. The demographic characteristics of the initial study are described in detail elsewhere by Backman and colleagues.
Participants were included in the present analysis if they self-identified as “Hispanic” and answered the condensed SASH acculturation scale questions.
Range of acculturation scores from 1 to 5 (1 = Spanish only; 5 = English only) for the following questions: (1) In general, what language(s) do you read and speak? (2) What language(s) do you usually speak at home? (3) In which language(s) do you usually think? and (4) What language(s) do you usually speak with your friends? Numbers in parentheses represent median cutoffs of mean acculturation score of above questions.
Range of acculturation scores from 1 to 5 (1 = Spanish only; 5 = English only) for the following questions: (1) In general, what language(s) do you read and speak? (2) What language(s) do you usually speak at home? (3) In which language(s) do you usually think? and (4) What language(s) do you usually speak with your friends? Numbers in parentheses represent median cutoffs of mean acculturation score of above questions.
Males
65
24
Females
67
25
a Acculturation as measured using condensed version (language scale) of the Short Acculturation Scale for Hispanics (SASH).
b Range of acculturation scores from 1 to 5 (1 = Spanish only; 5 = English only) for the following questions: (1) In general, what language(s) do you read and speak? (2) What language(s) do you usually speak at home? (3) In which language(s) do you usually think? and (4) What language(s) do you usually speak with your friends? Numbers in parentheses represent median cutoffs of mean acculturation score of above questions.
As displayed in the Figure, this secondary analysis included the original study variables (intention, attitude, subjective norm, perceived behavioral control, corresponding beliefs, as well as the behavioral variables of daily total calories, percentage calories from fat, and servings of fruits and vegetables) from the Theory of Planned Behavior. External study variables added for this study consisted of gender, acculturation, and grade level (not featured in the Figure) as a control variable. These external variables were imposed on the theoretical model because current findings indicate factors related to dietary behaviors are highly influenced by gender, acculturation, their interaction, and grade level for Latinos.
Table 2 displays the behavioral, normative, and control beliefs from the original study also examined in this secondary analysis. Belief product scores consisted of a product of scores for beliefs (0 to 9) and each corresponding evaluation component ranging from -4.5 to 4.5, which were not modified in any way from the original study.
Table 2Behavioral, Normative, and Control Beliefs as Determined by Backman and Colleagues in Original Study
Behavioral
Normative
Control
Like the taste of healthful food items
Mother
Knowledge about how to eat a healthful diet
Feel good about self
Sister/brother
Availability of healthful food
Tolerance of giving up liked food items
Friend
Motivation
Lose weight/maintain healthful weight
Father
Access to money
Feel healthy
Coach
Time
Look good
Teacher
Support and encouragement
Good shape
Good energy level
Good athletic performance
Behavioral beliefs that form attitude, normative beliefs that form subjective norm, and control beliefs that form perceived behavioral control as determined in original study by Backman et al.
The condensed scale consisted of 4 questions having to do with the language factor: (1) In general, what language(s) do you read and speak? (2) What language(s) do you usually speak at home? (3) In which language(s) do you usually think? and (4) What language(s) do you usually speak with your friends? Answers were provided on a 5-point Likert scale ranging from 1 (“Spanish only”) to 5 (“English only”). This variable was averaged across items, and a median split of 3.50 was used to organize groups into 4 “high” and “low” acculturation and gender groups: highly acculturated males, highly acculturated females, less acculturated males, and less acculturated females. A median split has been used with this scale in previous research on acculturation and dietary practices with Latino adults,
although the average acculturation of the previous study was considerably lower (mean [M] = 2.15, standard deviation [SD] = 0.65) than the average reported here (M = 3.64, SD = 0.96). This finding is not surprising, as research indicates that young Latino populations are often more acculturated than adult Latino populations.
The median split of this variable provided adequate sample sizes for each gender by acculturation grouping to maintain sufficient power for statistical analyses.
Human Subjects Approval
The Loma Linda University Institutional Review Board determined this secondary analysis to be exempt from human subjects review, as the data are anonymous and contain no personal identifiers. The study protocol for the original 1999 study was approved by both the Loma Linda University Institutional Review Board and the San Bernardino Unified School District prior to commencement.
Descriptive statistics were used to summarize the sample demographics (Table 1). Multiple regression analysis of the main theoretical constructs with the added variables of acculturation (high/low), gender (male/female), and the interaction of these variables were entered simultaneously to determine modifications in the prediction of intention to eat a healthful diet from attitude, subjective norm, and perceived behavioral control. In a second step, grade level was then entered into the model to examine modifications in this analysis. Additionally, independent belief predictors of attitude, subjective norm, and perceived behavioral control constructs were examined for each gender by acculturation group (highly acculturated males, highly acculturated females, less acculturated males, and less acculturated females).
General linear modeling was used to identify main effects of acculturation and gender and their interaction for each of the 4 main model constructs (attitude, subjective norm, perceived behavioral control, and intention), 3 outcome dietary behaviors (total calories, percentage calories from fat, and fruit and vegetable servings), and each belief score.
Statistical analysis was carried out using the SPSS statistical package, version 13.0 (SPSS Inc., Chicago, IL, 2004). The criterion α to determine statistical significance was set at .05. Bonferroni adjustments were made to reduce the probability of a type 1 error for each portion of the analysis and consisted of the formula of .05 divided by the total number of independent variables in each analysis.
Results
Acculturation and Gender on Model Predictors
Multiple regression analysis of the main model variables of intention to eat a healthful diet regressed on attitude, subjective norm, and perceived behavioral control (while controlling for gender, acculturation, the interaction of gender and acculturation, and grade level) produced varying results. The variance explained in intention by the main model variables increased significantly when gender, acculturation, and their interaction were simultaneously added into the model (R2 = 0.433 vs 0.458; P = .008). However, gender, acculturation, and their interaction were not significant independent predictors of intention. The addition of grade level in the model did not significantly increase the explained variance of intention in the model (P = .766). Acculturation was negatively correlated with intention to eat healthfully (r = -0.18, P = .002). No significant correlations were found between grade level with the main model and outcome dietary behavior variables.
Table 3 displays means, 95% confidence intervals, and significant findings from general linear modeling analysis for the 4 gender (male/female) and acculturation (high/low) groups. Gender effects on the model variables showed that females have significantly stronger intention to eat a healthful diet compared to males, a more positive attitude, and stronger subjective normative influences. These results are similar to findings from the original study.
Influence of gender and ethnicity on psychosocial predictors of healthy dietary practices and exercise behaviors in adolescents [doctoral dissertation].
When the effects of acculturation were examined, those less acculturated had significantly greater intention to eat a healthful diet than those in the highly acculturated group. There were no significant gender or acculturation differences for total calories, percentage calories from fat, fruit and vegetable servings, or perceived behavioral control. There were no significant interactions of acculturation and gender found for any of the model variables.
Table 3Means and 95% Confidence Intervals for Key Model Variables by Gender and Acculturation
Range of intention, attitude, subjective norm, and perceived behavioral control is 0 to 9. Sample size is N=238 for dietary outcome behaviors because of loss to follow-up in original study. Bonferroni adjustment resulted in criterion for P ≤ .016
a Range of intention, attitude, subjective norm, and perceived behavioral control is 0 to 9. Sample size is N=238 for dietary outcome behaviors because of loss to follow-up in original study. Bonferroni adjustment resulted in criterion for P ≤ .016
b Significant difference between genders (P ≤ .001)
c Significant difference between high and low acculturation groups (P ≤ .005).
Table 4 displays the results among Latinos of general linear modeling on belief product scores examined by gender, acculturation, and the interaction of gender and acculturation. Analysis of variance revealed that females believed feeling healthy and looking good were stronger contributors to attitude toward eating a healthful diet compared to males. For males, good athletic performance was a stronger belief for attitude to eat a healthful diet compared to females. Significant differences for gender were found on the influence of the mother as a referent for subjective norm influence, with females reporting greater influence than males (M = 14.13, SD = 2.90 vs M = 6.70, SD = 3.29 for males). Though significant findings were not found for gender by acculturation, highly acculturated males reported the mother as the least influential, whereas less acculturated females reported the mother as the most influential referent (M = 4.98, SD = 4.44 vs M = 15.98, SD = 3.89). The referent of sibling (sister/brother) was substantially higher for females (M = 4.735, SD = 1.92 vs M = 1.24, SD = 2.17) compared to males (P ≤ .019), but just failed to meet the Bonferroni corrected criterion value of P ≤ .016. Significant findings for acculturation included less acculturated Latino adolescents compared to highly acculturated Latinos having more tolerance to give up liked food items as a healthful dietary belief (M = 6.25, SD = 3.19 vs M = -1.01, SD = 3.19). Having support and encouragement as a resource for perceived behavioral control was significantly different, with less acculturated Latino adolescents reporting greater support than highly acculturated Latinos (M = 9.95; SD = 3.42 vs M = 3.90; SD = 3.41). No significant interactions of gender and acculturation were found for any of the belief product scores.
Table 4Means and 95% Confidence Intervals for Q10 Belief Product Scores by Gender and Acculturation Status
Belief product scores consist of product score of belief (range 0 to 9) times the evaluation component range -4.5 to 4.5). Only significant findings between gender and acculturation reported here. No interaction of gender and acculturation found for any belief product scores. Bonferroni adjustment resulted in criterion for P ≤ .016
Significant difference between high and low acculturation groups (P ≤ .016)
4.08 (.79, 7.37)
3.71 (1.08, 6.34)
5.91 (.56, 11.26)
13.98 (9.72, 18.24)
a Belief product scores consist of product score of belief (range 0 to 9) times the evaluation component range -4.5 to 4.5). Only significant findings between gender and acculturation reported here. No interaction of gender and acculturation found for any belief product scores. Bonferroni adjustment resulted in criterion for P ≤ .016
b Significant difference between high and low acculturation groups (P ≤ .016)
c Significant difference between genders (P ≤ .016).
Multivariate regression of attitude, subjective norm, and perceived behavioral control on their respective belief product scores by gender and acculturation group revealed unique influences. Lose weight/maintain healthful weight was a strong independent positive predictor of attitude for highly acculturated females (β = .40; P ≤ .004). No other single belief product score was a significant independent predictor of attitude across each gender and acculturation group. Results for subjective norm showed that across all 4 gender and acculturation groups, the mother was a strong significant independent predictor of subjective norm. Sibling (sister/brother) was a strong significant independent predictor for less acculturated males (β = .46; P ≤ .008), and friend was a significant independent predictor for highly acculturated females (β = .32; P ≤ .008). For perceived behavioral control, no single belief product score was found to be a significant independent predictor in any of the 4 gender acculturation groups.
Discussion
This study reinforces the effectiveness of using the Theory of Planned Behavior to examine predictors of healthful diet practices specific to Latino adolescents. Gender, acculturation, and the interaction of acculturation and gender significantly modify the prediction of intention to eat a healthful diet for this Latino adolescent sample, warranting the inclusion of such variables in future studies with this population. Key differences exist for many of the key model constructs and belief product scores, and these differences are discussed below.
Female Latinas have higher positive attitudes, intention, and subjective normative influences compared to Latino males for eating a healthful diet. These findings are consistent with previous research on adolescent influences over healthful dietary choices and were described in Backman's originally published findings.
Influence of gender and ethnicity on psychosocial predictors of healthy dietary practices and exercise behaviors in adolescents [doctoral dissertation].
Gender differences on belief product scores indicate that Latina females emphasize feeling healthy and looking good as strong independent contributors to eating a healthful diet, compared to their male counterparts, who emphasize good athletic performance as a strong independent predictor. Professionals working to reduce adolescent obesity and improve healthful living with this population should take these gender differences into account when planning. When Latina adolescent females are given health information on how specific dietary behaviors affect health and overall physical appearance, adoption of positive dietary practices may follow as a result. Although no interaction of acculturation and gender were found on any belief product scores in this study, some research indicates that cultural ideals for Latina adolescents may shift as they acculturate to American ways, changing their preferences for body size and image.
More exploratory studies are needed to contextualize the meaning of feeling healthy and looking good for Latinas.
Existing gender differences for the influence of athletic performance is not surprising, as among Latinos, females have lower rates of physical activity engagement than males and many other gender ethnic groups, and most males emphasize sports over lifestyle physical activity.
When planning nutrition interventions, professionals should take into account these findings and structure dietary recommendations that support physical activity performance for males.
The influence of the mother in healthful dietary behaviors is clearly important for both female and male Latino high school age adolescents, regardless of acculturation status. Though the role of the mother is important for both groups, among Latinos the influence is more than twice as strong for female compared to male adolescents (M = 14.13 vs M = 6.70) as a social pressure to eat healthfully. Studies in adolescents support the role of the parent as a strong influence over adolescent dietary behaviors, especially for eating dinner, but few studies have examined the differences by gender within the Latino adolescent population.
This finding supports the need for inclusion of the mother in healthful nutrition planning, especially for females. Additional research is needed to examine the direct influence of the mother as a referent for healthful eating, relative to other familial and cultural parameters (including acculturation) for Latino adolescents.
Greater acculturation has negative effects on intention, tolerance for giving up liked food items, and support and encouragement for eating a healthful diet in Latino adolescents, with less acculturated groups showing more favorable outcomes. Given the effect of intention on actual behavior, this result supports other findings documenting a negative relationship between acculturation status and healthful dietary behavior for adolescents.
In addition, compared to highly acculturated adolescents, less acculturated Latino adolescents have more tolerance for giving up liked food items and report more support and encouragement to eat healthfully. This finding provides a unique insight into the maintenance of healthful practices for Latino adolescents as they navigate through the process of acculturation. It is plausible that as adolescent Latinos acculturate more to American ways of eating, their orientation toward maintaining healthful eating practices (or their intention to do so) is also mediated by the support through their reference groups, with greater emphasis on siblings for less acculturated males (compared to highly acculturated males) and friends for highly acculturated females (vs less acculturated females). Research indicates that as Latino adolescents acculturate, their perceptions of family members as referents change to external referents.
Additional exploratory studies, which include other measurements of acculturation, are needed to examine normative influences by gender and acculturation for the high school adolescent population.
It is important to note some of the study limitations. First, difficulties in the operationalization of healthful eating as 3 variables may limit direct comparison to findings of other studies and may not sufficiently represent the primary dietary concerns of Latino adolescents (consumption of soft drinks and fast food). Secondly, the data used for this study were collected in 1997. Though difficult to measure, it can be reasonably assumed that the adolescent population has not changed drastically in the city of San Bernardino, as the area continues to have a large population of Latino immigrants. In addition, to date no published research has examined the effect of acculturation and gender on diet and exercise using the Theory of Planned Behavior in the adolescent population. Third, the SASH is a unidirectional acculturation scale and has not been widely used in the evaluation of dietary practices in Latino adolescents. However, substantial use of the scale has included Latino adolescent research on other high risk behaviors, such as smoking, alcohol use, and sexual practices, justifying its usage here.
Implications for Research and Practice
The results of this study provide valuable insight about the role of gender and acculturation on healthful dietary influences for Latino high school adolescents. Unique relationships between Latino adolescent genders discussed here provide contextual information for health educators and clinicians working to promote healthful eating among Latino adolescents. As an illustration, the role of the mother as the referent group was clearly more influential for female versus male adolescents, siblings for less acculturated males, and friends for highly acculturated females. There are far too few studies examining the effects of acculturation on family and peer referent influences for dietary behaviors, warranting the need for more research. To ensure competency in clinical outcomes, consideration of such influences should be incorporated into the planning of best practice models of effective healthful diet interventions.
Acculturation status appears to have a significant influence on intention to eat healthfully for high school adolescents. Studies show that the maintenance of cultural practices, like language as measured here, has protective outcomes for many Latino adolescents. As the Latino immigrant populations grow and acculturate in many states, professionals who serve them must work to include culturally competent measures in promoting healthful dietary practices with this group.
Acknowledgments
Funding was provided by Loma Linda University, School of Public Health, Loma Linda, CA.
References
Centers for Disease Control and Prevention (CDC)
Youth Risk Behavior Surveillance—United States. In: CDC Surveillance Summaries, June 9, 2006.
More California teens consume soda and fast food each day than five servings of fruits and vegetables [Policy Brief]. UCLA Center Health Policy Research,
Los Angeles, CA2005
Differences in energy, nutrient, and food intake in a US sample of Mexican-American women and men: findings from the third National Health and Nutrition Examination Survey, 1988-1994.
Diet and food insufficiency among Hispanic youths: acculturation and socioeconomic factors in the third National Health and Nutrition Examination Survey.
Influence of gender and ethnicity on psychosocial predictors of healthy dietary practices and exercise behaviors in adolescents [doctoral dissertation].
00010-4&id=gr1.jpg){kind=link}