Health Education Research, Vol. 17, No. 1, 7-18,
February 2002
© 2002 Oxford University Press
Ethnic differences in social correlates of diet
1 Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, 1100 Bates Street, Houston, TX 77030,
2 Department of Behavioral Science, Box 243, University of Texas M. D. Anderson Cancer Center, Houston, TX 770204-6321,
3 Department of Health and Human Performance, University of Houston, Houston, TX 77030 and
4 Department of Behavioral Sciences and Health Education, The Rollins School of Public Health, Emory University, Atlanta, GA 30322, USA
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Little is known about whether culture influences social correlates of dietary behaviors. Questionnaires on parent- and child-reported family and peer influences on children's fruit, juice and vegetable consumption were analyzed for ethnic group differences in responses. Grade 4–6 students completed the questionnaires in the classroom and their parents completed telephone or in-home interviews. Analyses of variance across ethnic categories and
2 analysis of differences in ethnic group composition between clusters of scales were conducted. Few ethnic group differences were detected, suggesting substantial commonality among respondents. Ethnic differences might be accommodated by interventions tailored to particular behaviors among ethnic groups. |
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Culture influences one's beliefs, values, norms and attitudes (Marin et al., 1995
). Some have proposed that effective health promotion programs should be culturally sensitive and tailored to the specific target ethnic group (Kumanyika et al., 1992). Cultural sensitivity has been defined as `the extent to which ethnic/cultural characteristics, experiences, norms, values, behavioral patterns and beliefs of a target population as well as relevant historical, environmental, and social forces are incorporated in the design, delivery, and evaluation of targeted health promotion materials and programs' (Resnicow et al., 1999). A model for understanding and engineering cultural sensitivity for public health intervention delineated two dimensions: surface and deep structure sensitivity (Resnicow et al., 1999). Surface sensitivity requires incorporating the normal cues and symbols reflecting the target groups' culture. Deeper structure sensitivity requires understanding the dominant beliefs and values of a culture, and tailoring materials that incorporate and resonate with those beliefs and values. Which beliefs and values constitute a culture's deeper structure have not been clearly elucidated. Interventions addressing audiences with people from several cultures present interesting complications for `multicultural sensitivity'. The extent to which there is overlap in deeper structure behavior and beliefs among cultures could vary from no overlap to complete overlap. In a pluralistic society such as the US, all families are not necessarily committed to subcultural values and beliefs, and levels of acculturation vary. This adds other dimensions to designing culturally appropriate interventions.
Dietary behaviors have important implications for health (Potter, 1997) and a multiculturally sensitive approach to changing dietary behaviors should be based on a thorough understanding of deep structure differences in related beliefs. Along these lines, the `cultural consensus' model `has been used to study individual knowledge levels, or competencies from the pattern in interinformant agreement' (Weller, 1987). However, more commonalties than differences have been found across groups that varied substantially by culture (Weller et al., 1999). Whether there are cultural differences in the influences on dietary behaviors is not known.
There has been substantial interest in family and peer influences on what people eat (Baranowski, 1997). Fruit, juice and vegetable (FJV) consumption has been shown to be protective from most cancers (Potter et al., 1997), heart disease (Ness and Powles, 1997) and obesity (Lloyd et al., 1998; Raynor et al., 1999). Based on a social cognitive framework and extensive focus groups with children and parents from three ethnic groups (Cullen et al., 2000a), several measures of possible parent and peer (social–environmental) influences on children's consumption of FJV were developed and shown to have acceptable reliability (Cullen et al., 2000b, 2001). Ethnic differences in FJV consumption were noted: Euro-American students consumed more fruit (0.63 serving) and total vegetables (1.18 servings) compared with African-American (0.31 serving fruit and 0.93 serving vegetable) or Hispanic 0.43 serving fruit and 0.85 serving vegetable) students (P < 0.001 for both) (Cullen et al., 2000b). Several of the parent and child measures were significantly related to child FJV consumption, with most in the directions that were hypothesized (Cullen et al., 2000b, 2001). For example, child-reported parental FJV modeling was significantly positively correlated with child fruit and juice consumption (P < 0.01 for both) (Cullen et al., 2001). Parent-reported permissive parenting practices measures were significantly negatively related to child fruit (P < 0.05) and vegetable (P < 0.01) consumption; child fruit consumption was negatively correlated with parent-reported negative home FJV barriers (P < 0.05); and parent-reported FJV planning self-efficacy was positively correlated with child fruit consumption (P < 0.05) (Cullen et al., 2000b). Since ethnic group differences were obtained in FJV consumption and the family variables correlated with FJV consumption, it is reasonable to believe that differences in family variables would account for the identified ethnic group differences in consumption. The purpose of this study was to assess possible differences in family influences on child dietary behaviors in order to provide guidance to those designing nutrition education programs for children from differing ethnic groups.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Sample
The data were obtained as part of a study to identify social environmental correlates of children's FJV consumption. This study was approved by the Institutional Review Board of the University of Texas M. D. Anderson Cancer Center. Grade 4–6 classrooms from 19 parochial schools in the greater Houston, TX, area were recruited. Parental consent and child assent were obtained for all data collections.
Procedures
Social–environmental questionnaires were developed and tested with grade 4–6 children and parents in the spring of 1998 (Cullen et al., 2000a,b, 2001). Focus group discussions with representatives of each ethnic group were used in the design of these questionnaires in order to tap deeper structure beliefs in regard to diet (Cullen et al., 2000a). Following psychometric assessment (Cullen, et al., 2000b, 2001), these questionnaires were administered to a second group of grade 4–6 children and their parents in the fall of 1998. All questionnaires were in English. Approximately 65% of eligible students returned consent forms. No data were available for those students who did not participate in this study.
The same procedure was used for each child data collection period. Trained data collectors (undergraduate and graduate nutrition students) visited each classroom and read each question to the students, who marked their answer on scannable forms. A rover circulated around the classroom to provide assistance as needed. Participating children (n = 221 in spring and n = 287 in fall) received small gifts.
In-home interviews with the majority of the available parents (76 of 109) were conducted in the spring of 1998. The remainder of the parents (n = 33) were interviewed by phone due to time and budget limitations. Trained data collectors administered the questionnaires to the parent. Parents received a gift certificate to a local grocery store for participating. All parent interviews were conducted by phone in the fall of 1998 (n = 138), using similar procedures.
Parent scales
Parenting style refers to methods used by parents to maintain or modify children's behaviors. Two scales, supportive (e.g. I listen to what my child has to say) and permissive parenting (e.g. I forget the rules I make for my child) practices, were obtained from the Authoritative Parenting Index (Jackson et al., 1998). Internal consistencies for the 11-item supportive parenting practices factor (
= 0.72) and five-item permissive parenting practices factor ( = 0.73) were acceptable.
Two subscales for a parent food socialization–encouraging questionnaire (Olvera-Ezzell et al., 1990) were identified. Encouraging expectancies ( = 0.79) included seven items like `How often do you tell your child this food is good for him/her?'. The encouraging consequences scale ( = 0.70) included five items like `How often do you tell your child if s/he eats the food you will give him/her dessert?'. A 14-item parent food discouraging practices scale (Olvera-Ezzell et al., 1990) (e.g. How often do you tell your child the food is too sweet?) had good internal consistency ( = 0.84).
Five scales related to meal planning and food preparation were obtained from a new questionnaire (Cullen et al., 2000b). A six-item parent meal-related planning practices scale (e.g. I plan menus before doing shopping) had modest reliability ( = 0.68). A four-item child shopping influence scale (e.g. My child goes shopping with me) had modest reliability ( = 0.67). An 11-item parent FJV preparation practices scale (e.g. How often do you put fruit into your child's lunch?) had acceptable reliability ( = 0.73). The four-item child lunch/snack FJV preparation scale (e.g. How often does your child put fruit in his lunch?) had good reliability ( = 0.82). The three-item child dinner FJV preparation (How often does your child prepare his own dinner?) scale also had good reliability ( = 0.84).
A new barriers questionnaire yielded three FJV barriers scales (Cullen et al., 2000b). The internal consistencies varied from low to good: a nine-item negative family barriers scale (e.g. I don't have time to fix vegetables) ( = 0.68), a three-item cost/spoilage scale (e.g. Fresh fruit and vegetables cost too much) ( = 0.83) and a three-item canned/frozen foods scale (e.g. Canned fruit and vegetables are not as healthy as fresh ones) ( = 0.53).
Parental self-efficacy to promote healthy diets among their children was measured with three new scales (Cullen et al., 2000b). Parent self-efficacy for FJV modeling/socialization ( = 0.78) included six items like `How sure are you that you can regularly tell your child you like fruit for snack?'. Parent self-efficacy for FJV planning/encouraging ( = 0.75) included eight items like `How sure are you that you can regularly plan menus that have one serving of vegetables at dinner?'. Parent self-efficacy for FJV availability/ accessibility ( = 0.70) included six items like `How sure are you that you can regularly have cut up fruit for your child's snack?'. All had acceptable reliability.
Child questionnaires
Several peer and parent normative influences questionnaires were developed (Cullen et al., 2001). Parent and peer perceived norms for eating FJV (i.e. what FJV children think their family and friends are eating) were identified. Internal consistency for this 12-item scale was good ( = 0.83). FJV normative beliefs (i.e. what children believe their parents and friends think about eating FJV) for peer (n = 6 items; = 0.88) and family (n = 6 items; = 0.85) had good reliability. Parent (n = 7 items; = 0.88) and peer FJV (n = 7 items; = 0.85) normative expectations (i.e. children's beliefs about whether parents or friends encourage child to eat FJV) were also measured with good reliability.
Two scales, supportive and permissive parenting practices, were also obtained from the child version of the API (Jackson et al., 1998). These scales contained the same items as the parent version, with modified phrasing (e.g. My mother listens to what I have to say). Internal consistency was acceptable for the supportive parenting practices scale (n = 12 items; = 0.77), but modest for the permissive parenting practices scale (n = 6 items; = 0.62).
Scales to measure parenting style in regard to child-food control issues were obtained from a new questionnaire (Cullen et al., 2001). Parent control (11 items like: `My mother plans all my meals'), permissive eating (four items like `My mother lets me eat whatever I want for dinner') and food self-preparation (four items like `My mother lets me prepare my own lunch') had acceptable internal consistencies ( = 0.77, 0.76 and 0.76, respectively).
Peer (n = 13 items) and parent (n = 15 items) FJV modeling scales (e.g. My friends/parent eats fruit at dinner when I am with them) had high internal consistency ( = 0.82 and 0.89, respectively).
Data analyses
The spring and fall samples were combined to provide a larger sample. Only data from African-American, Euro-American and Hispanic students (n = 456) and parents (n = 218) were analyzed. Cronbach's was calculated for each scale for each ethnic group separately. The traditional way of estimating differences by ethnic group is to test for differences in mean values across ethnic groups; therefore, one way ANOVA by ethnic group analyses were conducted for each scale. Alternatively, the `cultural consensus model' (Romney and Weller, 1986) assesses consensus among individuals, which is comparable to cluster analysis (Glanz et al., 1998). Thus, cluster analysis procedures were used to categorize the individuals into homogenous subgroups in patterns of responses to each of the family scales. The Kmeans clustering algorithm was employed (Gruvaeus and Wainer, 1972; Hartigan, 1975; Hartigan and Wong, 1979). This method produces `Kmeans' or clusters of individuals who score similarly on a set of items. In a two cluster solution, for example, Kmeans procedure starts with one cluster or group and splits it in two by picking the individual farthest from the center in a multidimensional space as a seed for the second cluster. All individuals are then assigned to one or the other cluster depending on which center they are closest to in a multidimensional space. The objective of clustering is to identify the smallest number of groups that both maximize distance between groups and maximize frequency within each cluster. Several techniques were used to determine the number with clusters. First, multiple cluster solutions were compared for distances between clusters. In regard to the frequency within each cluster, if the number of individuals was zero, then too many groups had been specified. Graphing item responses by cluster provided a visual analysis of how clusters differed. The result of both analysis techniques in the items within scales suggested that in most cases two clusters were most appropriate for the individual item analysis. The results of both analysis techniques across scales suggested three clusters were most appropriate. Ethnic group distributions for each cluster were determined and 2 tests (cluster by ethnicity) were conducted. Data were analyzed using SAS.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The students whose parents completed interviews were similar to the total student sample (Table I
). Mean age of the parents was 40 ± 6.6 years. The majority of Hispanic parents were born in the US (75%). More Hispanic parents reported completing a high school education or less compared with African-American or Euro-American parents (P < 0.001). More African-American parents reported single parent families compared with Hispanic or Euro-American parents (P < 0.001).
|
Reliability was adequate to excellent for all but two parent scales (permissive parenting practices and canned/frozen food barriers) (Table II
) and one child-reported scale (permissive parenting practices) (Table III). There were some differences in values by ethnic groups.
|
|
Significant ethnic differences in mean scale scores were found for three parent-reported (Table II
) and three child-reported scales (Table III). Hispanic parents reported higher permissive parenting practices compared with Euro-American parents (P < 0.05). Euro-American parents reported the lowest encouraging expectancies (P < 0.05). Euro-American parents reported more meal planning practices compared with Hispanic parents (P < 0.01). African-American students reported higher peer normative beliefs for eating FJV compared with Euro-American students (P < 0.05). Euro-American students reported higher peer FJV modeling compared with Hispanic students (P < 0.01). Euro-American students reported less permissive eating compared with African-American and Hispanic students (P < 0.001).
There was only one significant difference in cluster membership for the parent scales by ethnicity (Table IV). Greater percentages of African-American and Euro-American parents belonged to the `more meal planning' group compared with Hispanic parents (P < 0.05).
|
There were only two significant differences in cluster membership for the child-reported scales by ethnicity (Table IV
). More African-American and Euro-American children belonged to the cluster reporting higher FJV modeling by peers compared with Hispanic children (P < 0.05). More Euro-American children belonged to the cluster reporting higher parent FJV normative expectations (P < 0.01).
Three parent and three child clusters were obtained from cluster analyses across scales (Table V). There were no statistically significant ethnic differences in cluster membership for either the parent or child clusters.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
There were no ethnic differences in cluster membership based on all psychosocial scales for parents or children. Few ethnic group differences in responses to questionnaires on social environmental influences on children's FJV consumption were identified and what differences were detected appear to be small in comparison to the ranges of possible values. Correcting for the large number of tests conducted would reduce even further the number of statistically significant differences. Only two scales (supportive and permissive parenting practices) were completed by both parents and students. Cluster membership for each of these scales did not differ by ethnic group for students or parents. Significant ethnic differences were found only for mean scale values for the parent-reported permissive parenting practices, with Hispanic parents reporting significantly higher mean permissive parenting practices. We know of no related literature on ethnic differences in parenting with which to compare these results.
No significant ethnic differences were found in cluster membership across psychosocial scales for either the parents or children. A similar Kmeans cluster analysis was conducted on responses to personal, behavioral and environmental questions across five domains of health-related behaviors (tobacco and alcohol use, exercise, nutrition, and weight control) by a national sample of adults (Glanz et al., 1998). However, our clusters are not comparable to theirs because we clustered on different scales and ethnic differences in their cluster membership were not reported. We know of no other reports using cluster analysis on correlates of diet.
Significantly more Euro-American students were members of the `more peer FJV modeling' cluster and reported significantly higher mean peer FJV modeling values than Hispanic students. Cluster membership for student-reported peer FJV normative beliefs did not differ by ethnicity, but African-American students reported significantly higher peer FJV normative beliefs than Euro-American students. The scale items represented all positive FJV behaviors; no negative items like teasing because the child was eating vegetables were employed. Among 12- to 16-year-old students, friends' consumption significantly influenced eating `unhealthy foods' (Woodward et al., 1996). We know of no other reports on ethnic differences in peer-related dietary behaviors. African-American students who participated in the Atlanta Gimme 5 intervention reported no differences in a social norms measure (i.e. most people in my family or most kids my age are eating FJV) compared with Euro-American students (Baranowski et al., 2000).
Although membership in the child-reported parent FJV normative expectations cluster varied significantly, with more Euro-Americans appearing to belong to the `higher parent normative expectations' cluster, mean values did not differ by ethnicity. Cluster membership for student reported permissive eating did not differ by ethnicity, but Euro-American students reported significantly lower permissive eating mean values compared with African-American or Hispanic students. As noted previously, Hispanic parents reported more permissive parenting practices which could be related to more permissive eating reported by Hispanic students. Similarly, Olvera-Ezzell et al. found Hispanic mothers to be permissive when encouraging their children to eat a new food (Olvera-Ezzell et al., 1990). No other dietary behavior literature in this area was found. In a study on parental perceptions and behaviors in regard to teen smoking, African-American parents reported more anti-tobacco use socialization compared with Euro-American parents (Clark et al., 1999). The 51 African-American parents were more likely to set ground rules regarding tobacco use by children and communicated these rules to their children compared with the 216 Euro-American parents (Clark et al., 1999). African-American parents were also more likely to believe they could influence their child's tobacco use (Clark et al., 1999). It appears likely that parental response to adolescent risk behaviors varies with the particular behavior.
The low frequency of substantial ethnic group differences in responses to questionnaires measuring social–environmental influences on children's FJV consumption, despite lower educational attainment of Hispanic parents and more African-American single-parent households, was surprising. Differences in health enhancing behaviors and chronic disease risk factors by educational level have been documented for adolescents (Lowry et al., 1996; Koivusilta et al., 1999) and adults (Winkleby et al., 1990; Erkkila et al., 1999). However, data on recent dietary trends in the US indicated that differences in the diets of whites and blacks of all socioeconomic levels were relatively similar (Popkin et al., 1996).
There are at least five potential explanations for such lack of differences: (1) unreliability of assessment minimized the ability to detect true differences by ethnic group, (2) these scales did not capture the true beliefs of the participants, (3) these beliefs did not reflect core or deeper structure cultural beliefs, (4) biased selection of sample and (5) true lack of ethnic group differences in those beliefs. With regard to reliability, the internal consistency values for most of the scales were at acceptable levels and there were only a few substantial differences in reliability across ethnic groups. Among children, one scale for which ethnic group differences appeared was among those at the lowest end of the distribution of reliability coefficients. This suggests that either this was a strong ethnic group difference or the finding resulted from type 1 error. Unreliability of scales does not appear to provide a reasonable explanation for the overall pattern of findings.
These items were generated using a theoretical foundation (Cullen et al., 1998) and substantial focus group discussions with children and parents from all the ethnic groups (Cullen et al., 2000a). Two existing prevalidated scales were incorporated in the research (Olveral-Ezzell et al., 1990; Jackson et al., 1998) and items for the other scales were generated to reflect statements in the focus groups within the theoretical framework. These procedures appear to minimize the likelihood that the scales did not capture true beliefs.
What constitutes a deeper structure belief has not been clearly delineated. For example, the African-American community has been characterized by the core values of communalism, religion/spiritualism, commitment to family, and connection to ancestors and history (Resnicow et al., 1999), while the Hispanic community has been characterized by the core values of family, respect for elders, fatalism and the importance of social interactions (Resnicow et al., 1999). We anticipated that because of the central role of family in all these cultures, that scales assessing family's relationships to food would have tapped deeper structure beliefs. Perhaps this is not the case. If so, investigators must more intensively investigate how these deeper structure beliefs relate to food, if at all.
The sample for these analyses was obtained from 11 parochial schools in the greater Houston area. Religious affiliation was not a requirement for enrollment in these schools and the schools enrolled children from all social strata. The high proportion of parents with a high school diploma or less and the almost equal distribution across ethnic groups suggests reasonable success in tapping a broad range of social strata. It remains possible, however, that there is a self selective bias in families enrolling in parochial schools which results in more homogenous beliefs and limits generalizability. Further research needs to be conducted with children more assuredly representative of all ethnic groups and social strata, and include measures of acculturation.
Finally, it is possible that there really are few differences across ethnic groups in beliefs about social environmental influences on children's FJV consumption. The standard deviations for some of these scales were substantial. This suggests that there is variability in many of these beliefs, but ethnicity did not account for that variability or, put another way, there is substantial within group variability for many of these beliefs. Most of these parents were born in the US. This pattern of findings among people already in this country could be explained by substantial acculturation among all ethnic groups by a dominant common media and prevalent similar food-related institutions (e.g. fast food outlets, supermarkets). The findings by Weller et al. of lack of differences in disease-related beliefs across groups that did not share common media or food related institutions (i.e. Connecticut, Texas, Mexico, Guatemala) raises the possibility that beliefs in regard to factors influencing food consumption are pancultural or otherwise not related to culture (Weller et al., 1999). Influences on other behaviors may show similar results. Lack of significant ethnic differences in scores for the AIDS Risk Self-Efficacy Scale for Sexual Activity have been reported (Faryna and Morales, 2000). Further research must be conducted across cultural groups within the US and across countries to address these issues.
Several limitations should be noted. The data were collected at two time points and two different methods were used for collecting the parent data which may have added error to the responses. Generalizability is limited because the students and parents were recruited from a small number of parochial schools.
The minimal number and relatively small size of the ethnic group differences obtained in the variables in this study suggests that dietary change interventions can have a substantial common core with a small amount of tailoring to each ethnic group. In regard to tailoring, one strategy would attempt to change the relevant behavior (behavior tailoring). For example, messages to Euro-American parents could incorporate and emphasize planning for meals, while messages for the other ethnic groups could promote greater meal planning behaviors. Several possible problems with this approach include: more meal planning may not result in more FJV consumption; perhaps all groups could benefit from higher meal planning; or these types of family behaviors could be highly resistant to change. A second tailoring strategy would be to find other ways to change dietary behavior that complemented or compensated for the particular planning behavior (compensatory tailoring). For example, to achieve the same ends without relying on a strong sense of meal planning, greater importance could be placed on purchasing more FJV to have more FJV available in the Hispanic home environment. Thus, when a meal was assembled, FJV would be an obvious selection. Much research needs to be conducted to determine when each of these strategies might be appropriate.
Similarly, the low level of permissiveness for eating among Euro-American families and the higher level of permissiveness for eating among Hispanic and African American families is inconsistent with the high levels of food control reported in all three groups. This requires further investigation. Perhaps intervention with the Hispanic community needs to reduce the level of permissive eating on the part of the parents (behavior tailoring) or work more intensively with the children to instill more self-control (compensatory tailoring). Interventions varying these components could be implemented and thoroughly evaluated.
The analyses reported herein should be considered exploratory or hypothesis generating, and interpreted with caution. The intriguing findings, however, merit further research.
|
Acknowledgments |
|---|
This work is a publication of the USDA/ARS Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine and Texas Children's Hospital, Houston, TX. This material is based upon work supported by the Cooperative State Research, Education, and Extension Service, US Department of Agriculture, under Agreement no. 9700578. This project has been also been funded in part by federal funds from the USDA/ARS under Cooperative Agreement no. 58-6250-6001. The contents of this publication do not necessarily reflect the views or policies of the USDA, nor does mention of trade names, commercial products or organizations imply endorsement by the US Government.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Baranowski, T. (1997) Families and health action. In Gochman, D. (ed.), Handbook of Behavioral Medicine: Vol. 1. Determinants of Health Behaviors: Personal and Social. Plenum, New York, pp. 179–205.
Baranowski, T., Davis Hearn, M., Resnicow, K., Baranowski, J., Doyle, C., Smith, M., Lin, L. and Wang, D. T. (2000) Gimme 5 fruit, juice and vegetables for fun and health: outcome evaluation. Health Education and Behavior, 27, 96–111.
Clark, P. I., Scarisbrick-Hauser, A., Gautam, S. P. and Wirk, S. J. (1999) Anti-tobacco socialization in homes of African-American and white parents, and smoking and non smoking parents. Journal of Adolescent Health, 24, 329–339.[Web of Science][Medline]
Cullen, K. W., Bartholomew, L. K., Parcel, G. S. and Kok, G. (1998) Intervention mapping: use of theory and data in the development of a nutrition program to increase fruit and vegetable intake in girls ages 9–12. Journal of Nutrition Education, 30, 188–195.
Cullen, K. W., Baranowski, T., Rittenberry L. and Olvera N. (2000a) Social–environmental influences on children's diets: results from focus groups with African-, Euro-, and Hispanic-American children and their parents. Health Education Research, 15, 581–590.
Cullen, K. W., Baranowski, T., Rittenberry, L., Cosart, C., Owens, E., Hebert, D. and deMoor, C. (2000b) Social–environmental influences on children's fruit, juice, and vegetable consumption reported by parents: reliability and validity of measures. Public Health Nutrition, 3, 345–356.[Medline]
Cullen, K. W., Baranowski, T., Rittenberry, L., Cosart, C., Owens, E., Hebert, D. and deMoor, C. (2001) Child-reported social–environmental influences on children's fruit, juice, and vegetable consumption: reliability and validity of measures. Health Education Research, 16, 187–200.
Erkkila, A. T., Sarkkinen, E. S., Lehto, S., Pyorala, K. and Uusitupa, M. I. (1999) Diet in relation to socioeconomic status in patients with coronary heart disease. European Journal of Clinical Nutrition, 53, 662–668.[Web of Science][Medline]
Faryna, E. L. and Morales, E. (2000) Self-efficacy and HIV-related risk behaviors among multiethnic adolescents. Cultural Diversity and Ethnic Minority Psychology, 6, 42–56.
Glanz, K., Basil, M., Maibach, E., Goldberg, J. and Snyder, D. (1998) Why Americans eat what they do: taste, nutrition, cost, convenience, and weight control concerns as influences on food consumption. Journal of the American Dietitians Association, 98, 1118–1126.
Gruvaeus, G. and Wainer, H. (1972) Two additions to hierarchical cluster analysis. British Journal of Mathematical and Statistical Psychology, 25, 200–206.
Hartigan, J. A. (1975) Clustering Algorithms. Wiley, New York.
Hartigan, J. A. and Wong, H. (1979) Clustering Algorithms. Wiley, New York.
Jackson, C., Henriksen, L. and Foshee, V. A. (1998) The Authoritative Parenting Index: predicting health risk behavior among children and adolescents. Health Education and Behavior, 25, 319–337.
Koivusilta, L. K., Rimpela, A. H. and Rimpela, M. K. (1999) Health-related lifestyle in adolescence—origin of social class differences in health? Health Education Research, 14, 339–355.
Kumanyika, S. K., Morssink, C. and Agurs, T. (1992) Models for dietary and weight change in African-American women: identifying cultural components. Ethnicity and Disease, 2, 166–175.[Medline]
Lloyd, T., Chinchilli, V. M., Rollings, N., Kieselhorst, K., Tregea, D. F., Henderson, N. A. and Sinoway, L. I. (1998) Fruit consumption, fitness, and cardiovascular health in female adolescents: the Penn State Young Women's Health Study. American Journal of Clinical Nutrition, 67, 624–630.[Abstract]
Lowry, R., Kann, L., Collins, J. L. and Lolbe, L. J. (1996) The effect of socioeconomic status on chronic disease risk behaviors among US adolescents. Journal of the American Medical Association, 276, 792–797.
Marin, G., Burhansstipanov, L., Connell, C. M., Gielen, A. C., Helitzer-Allen, D., Lorig, K., Morisky, D. E., Tenney, M. and Thomas, S. (1995) A research agenda for health education among underserved populations. Health Education Quarterly, 22, 346–363.[Web of Science][Medline]
Ness, A. R. and Powles, J. W. (1997) Fruit and vegetables, and cardiovascular disease: a review. International Journal of Epidemiology, 26, 1–13.
Olvera-Ezzell, N., Power, T. G. and Cousins, J. H. (1990) Maternal socialization of children's eating habits: strategies used by obese Mexican-American mothers. Child Development, 61, 395–400.[Web of Science][Medline]
Popkin, B. M., Siega-Riz, A. M. and Haines, P. S. (1996) A comparison of dietary trends among racial and socioeconomic groups in the United States. New England Journal of Medicine, 335, 716–720.
Potter, J. D. (ed.) (1997) Food, Nutrition and the Prevention of Cancer: A Global Perspective. World Cancer Research Fund and American Institute for Cancer Research, Washington, DC.
Raynor, H. A., Epstein, L. H. and Gordy, C. C. (1999) Effects of increased fruits and vegetables and decreasing high-fat and or high sugar during obesity treatment. Annals of Behavioral Medicine, 21 (Suppl.), S019.
Resnicow, K., Baranowski, T., Ahluwahlia, J. S. and Braithwaite, R. L. (1999) Cultural sensitivity in public health: defined and demystified. Ethnicity and Disease, 9, 10–21.[Medline]
Romney, A. K. and Weller, S. C. (1986) Culture as consensus: a theory of culture and informant accuracy. American Anthropologist, 88, 313–338.[Web of Science]
Weller, S. C. (1987) Shared knowledge, intracultural variation, and knowledge aggregation. American Behavioral Scientist, 31, 178–193.
Weller, S. C., Baer, R. D., Pachter, L. M., Trotter, R. T., Glazer, M., Garcia de Alba Garcia, J. E. and Klein, R. E. (1999) Latino beliefs about diabetes. Diabetes Care, 22, 722–728.
Winkleby, M., Fortmann, S. and Barrett, D. (1990) Social class disparities in risk factors for disease: eight-year prevalence patterns by level of education. Preventative Medicine, 19, 1–2.[Web of Science][Medline]
Woodward, D. R., Boon, J. A., Cumming, F. J., Ball, P. J., Williams, H. M. and Hornsby, H. (1996) Adolescents' reported usage of selected foods in relation to their perceptions and social norms for those foods. Appetite, 27, 109–117.[Web of Science][Medline]
Received on August 11, 2000; accepted on March 30, 2001
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Watson, T. Baranowski, and D. Thompson Item response modeling: an evaluation of the children's fruit and vegetable self-efficacy questionnaire Health Educ. Res., December 1, 2006; 21(suppl_1): i47 - i57. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. C. Whitaker and S. M. Orzol Obesity Among US Urban Preschool Children: Relationships to Race, Ethnicity, and Socioeconomic Status Arch Pediatr Adolesc Med, June 1, 2006; 160(6): 578 - 584. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||