Search tips
Search criteria 


Logo of herLink to Publisher's site
Health Educ Res. 2009 August; 24(4): 586–595.
Published online 2008 December 15. doi:  10.1093/her/cyn059
PMCID: PMC2706493

Measurement characteristics of dietary psychosocial scales in a Weight Gain Prevention Study with 8- to 10-year-old African-American girls


Few measurement instruments for children's eating behaviors and beliefs have been specifically validated for African-American children. Validation within this population is important because of potential cultural and ethnic influences. Objectives were to evaluate established and newly developed or adapted dietary psychosocial measures in a sample of 303 preadolescent African-American girls and their caregivers. Acceptable internal consistency (Cronbach's α ≥ 0.70) was found for measures of girls’ self-efficacy for healthy eating, outcome expectancies for healthy eating, positive family support for healthy eating and household availability of low-fat food and fruit, juice and vegetables (FJV). Evidence for concurrent validity was found with significant associations between self-efficacy for healthy eating and lower intake of energy (r = −0.17) and fat grams (r = −0.16). Greater FJV availability was associated with greater FJV intake (r = 0.14) and lower body mass index (BMI) in girls (r = −0.12). Positive family support for healthy eating was associated with higher BMI in girls (r = 0.41). These results contribute to the development of scales to evaluate prevention interventions related to dietary intake in African-American children.


The prevalence of childhood obesity is increasing in the United States [1, 2], with a disproportionate rise among African-American girls [3, 4]. The long-term health consequences of obesity in African-American children are compounded by the fact that African-American adults are at elevated risk for obesity-related morbidity from diabetes, hypertension and cardiovascular disease [5]. To address the obesity epidemic and its health consequences, obesity prevention and intervention strategies have become a public health priority [6]. Development of programs for preventing accelerated weight gain in African-American children is a current research focus for this high-risk group [7, 8].

Understanding psychosocial influences that contribute to eating behaviors is essential to developing effective weight gain prevention interventions [911]. In particular, greater insight is needed into the psychosocial and environmental factors that influence eating behaviors in African-American children [1214]. Social cognitive theory (SCT) [15] emphasizes self-efficacy and outcome expectancy as important behavioral mediators. Hence, these constructs are useful variables to consider in behavioral intervention efforts.

While assessing the efficacy and effectiveness of intervention programs, the quality of the measurement instruments themselves is crucial to drawing accurate conclusions [16]. Measures of children's eating behaviors and beliefs are available [17, 18], but few have been specifically validated for African-American children [8, 16]. Validation within this population is important because of potential ethnic and cultural differences in nutritional beliefs and behaviors [8].

This paper examines the characteristics of self-reported dietary psychosocial measures from both girls and their caregivers in Phase 2 of the Girls’ health Enrichment Multi-site Studies (GEMS), an intervention designed to prevent excessive weight gain in 8- to 10-year-old African-American girls. Choice of psychosocial measures was guided primarily by the SCT framework. The objectives of this paper are (i) to evaluate pre-existing self-report instruments within this population and (ii) to evaluate instruments that were newly developed or adapted to this population during the GEMS Phase 1 pilot study.


Study design

The GEMS project is aimed at preventing obesity in preadolescent African-American girls. In Phase 2, field centers at the University of Memphis and Stanford University conducted separate interventions over two years. Detailed description of GEMS Phase 2 is available elsewhere [19, 20]. The current report is based on baseline data collected at the Memphis site from July 2002 through February 2004 after protocol approval by the University of Memphis Institutional Review Board, an external Data Safety Monitoring Board, and the National Heart, Lung and Blood Institute.


The target population was 8- to 10-year-old African-American girls whose body mass index (BMI) was more than or equal to the 25th percentile based on 2000 Centers for Disease Control and Prevention (CDC) growth charts [21]. Any girl who did not meet this criterion was eligible if their caregiver's BMI ≥25 kg/m2. Further exclusion criteria were medical conditions or medications affecting growth, conditions that would restrict participation in physical activity interventions, conditions affecting participation (e.g. two grade levels behind in reading/writing) and failure to give informed consent.

Of 463 participants screened, 303 (65%) girls and their caregivers were found eligible. Excluded were girls who did not meet eligibility standards or opted not to participate (n = 107). Girls related to a randomized participant (n = 53) were invited to participate as ‘guests’, but their data were not analyzed.



Anthropometric measures included girl's height, weight and BMI. Height was measured with a Shorr Height Measuring Board (Olney, MD, USA) and weight with a Scaletronix 5602 Model calibrated scale (White Plains, NY, USA). BMI was calculated as weight/height squared (kg/m2). BMI percentiles were generated with a SAS program for CDC 2000 growth charts available from


Potential covariates were girl's age, caregiver's education (1 = high school or less, 2 = some college and 3 = college graduate) and annual household income (categorized into five $20 000 increments from <$20 000 to ≥$80 000). Other information collected but not included was caregiver's age, other children and adults in the household, home ownership, household possessions and caregiver's cultural beliefs.

Child measures

Dietary recall

Dietary intake was assessed with three 24-hour dietary recall interviews and analyzed with Nutrition Data System for Research (NDS-R) 2005 software, developed by the Nutrition Coordinating Center, University of Minnesota, Minneapolis, MN, USA. Participants completed an initial face-to-face dietary recall interview. Subsequent recall interviews were conducted by telephone. Interviews occurred on non-consecutive days with attempts to include at least one weekend day. Typically, the amount of time from the first to last recall was 2–3 weeks. Dietary intake was averaged across recalls to produce daily averages.

Variables of interest were total energy intake (kcal/day); total fat (g); percent energy from fat and average daily servings of sweetened beverages, fruit, juice and vegetables (FJV) (including legumes). These variables were based on previously established behavioral food coding methods [22].

Self-efficacy for healthy eating

A 19-item questionnaire, adapted from an instrument developed by Reynolds et al. [23], measured self-efficacy for healthy eating with statements such as ‘How sure are you that you can eat some fruits and vegetables each day?’ Response options were 1 = ‘not sure’, 2 = ‘sort of sure’ and 3 = ‘very sure’. Higher scores corresponded to greater self-efficacy. Cronbach's α of 0.86 was reported in a sample of fourth grade schoolchildren [23].

Outcome expectancies for healthy eating

This 12-item instrument was developed and used in the ‘High 5’ study to examine the mediating influence of expectancy on dietary intake [23]. It was adapted for use in GEMS Phase 2. Girls indicated their level of agreement (1 = ‘disagree’, 2 = ‘sort of agree’ or 3 = ‘agree’) with statements that measured expected consequences of eating a healthy diet (e.g. ‘Eating healthy will make you smarter’). Higher scores indicated more positive expectancies. Internal consistency reported in the High 5 study was 0.67.

Beverage preferences

A 16-item questionnaire, modified from an 11-item measure developed in GEMS Phase 1, was used to assess preferences for sweetened (seven items) and non-sweetened beverages (nine items). Girls rated their preference for soft drinks (regular and diet), sports drinks (i.e. Powerade or Gatorade), punch, Kool-Aid (regular and diet), sweetened fruit drinks (i.e. Fruitopia, Sunny Delight and Capri Sun), Snapple (regular and diet), sweetened and unsweetened tea, 100% fruit juice, vegetable juice and water. In GEMS Phase 1, Cronbach's α for the entire scale was 0.71 with a 12-week test–retest intraclass correlation coefficient (ICC) of 0.79 [12]. Response options were 1 = ‘I do not like this’, 2 = ‘I like this a little’ and 3 = ‘I like this a lot’, so that higher scores reflected greater preference. Scores were calculated by dividing the sum of the responses by the total number of responses (e.g. rating all seven sweetened beverage items as ‘I like this a lot’ would yield a score of 21/7 = 3).

Social desirability

Potential social desirability response bias was measured with the nine-item Revised Children's Manifest Anxiety Scale (RCMAS) [24]. Girls answered 1 = ‘yes’ or 0 = ‘no’ to questions such as ‘I never get angry’. Higher scores reflected greater social desirability. In the first RCMAS validation report, internal consistency estimates for African-American girls were 0.59, 0.64 and 0.66 for 8-, 9- and 10-year-olds, respectively [24]. In GEMS Phase 1, internal consistency (excluding the first item of the original measure) was 0.78, with a test–retest ICC of 0.62 [25].

Parent measures

Availability and accessibility of fruit, 100% juice and vegetables

Caregivers completed a 44-item questionnaire on the availability (38 items) and accessibility (six items) of fruit, 100% juice and vegetables in their home. This instrument was adapted from a validated parent-report measure of FJV availability [26, 27], developed for school-aged children in the ‘5-A-Day’ project [28]. In GEMS Phase 1, internal consistency was 0.77, with a test–retest ICC of 0.50. The six items related to accessibility (e.g. cut-up fresh vegetables in refrigerator) were introduced in GEMS Phase 2.

Availability of low-fat/fat-free alternatives

Availability of high-fat and reduced fat foods was assessed with a questionnaire developed during GEMS Phase 1 [22]. Items in the scale derived from information acquired in focus groups with elementary schoolchildren [29]. Caregivers indicated whether particular food items (e.g. milk) had been available at home in the past two weeks. For food items that had both high-fat and reduced-fat options, caregivers specified which items were available (e.g. whole milk and skim milk). Subscales corresponding to high-fat foods (15 items) and low-fat foods (17 items) were generated. Ratio scores (e.g. available high-fat items/total high-fat items) indicated relative availability. In GEMS Phase 1, internal consistency estimates for high-fat and low-fat food availability were 0.59 and 0.64, respectively.

Food preparation

Food preparation practices during the past month were measured with a 25-item instrument that was developed in GEMS Phase 1 and derived from existing measures for adults [30] and children [31]. Two factors were reported in GEMS Phase 1 that corresponded to high-fat and low-fat preparation practices but explained only 22% of the variance, with Cronbach's α values of 0.58 and 0.66, respectively. Response options were 1 = ‘almost never’, 2 = ‘sometimes’ and 3 = ‘almost always’ to questions such as ‘How often did you serve bacon or sausage for breakfast?’

Family support for healthy eating

Caregiver support for healthy eating habits was assessed with an 11-item adapted measure originally developed by Sallis et al. [32], who reported internal consistency estimates >0.80 in an ethnically diverse sample of young adults. Caregivers indicated how often they engaged in behaviors such as ‘Complimented her on changing her eating habits’ and ‘Ate high-fat or high-salt foods in front of her’. Response options were 1 = ‘none’, 2 = ‘rarely’, 3 = ‘a few times’, 4 = ‘often’ and 5 = ‘very often’.

Statistical analyses

Analyses were performed with SPSS 15.01 [33]. Scales were scored by averaging across response values. A scale score was set to missing if responses were missing for 25% or more of the items.

Descriptive statistics were examined for distributional normality. Significant outliers discovered for dietary intake were excluded (e.g. consumption of 22 heads of cabbage in one day was unlikely). For three measures (food preparation, family support for healthy eating and beverage preferences), factor structure was explored using principal components analysis.

Internal consistency analyses (Cronbach's α) were performed. For measures with acceptable internal consistency estimates (≥0.70), concurrent validity was assessed by comparing scale scores to dietary intake and BMI using two-tailed Pearson correlations. Significance level was set at P <0.05.


Descriptive statistics are presented in Table I. Average age for girls was 9.3 years and average BMI was 21.8 kg/m2. Average BMI percentile rank was the 77th percentile, with 16.5% of girls between the 85th and 95th percentile and 40.6% at or above the 95th percentile at baseline. About 31% of caregivers graduated college, 53% had some college or technical school training and 16% had a high school education or less. Fifteen percent of caregivers reported an annual income over $60 000, 21% between $40 000 and $60 000, 40% between $20 000 and $40 000 and 24% <$20 000.

Table I.
Baseline characteristics of participants (n = 303) in Phase 2 of GEMS

Of 303 participants, 94% (n = 284) successfully completed all three dietary recalls. Among the remaining 19 girls, 17 completed two recalls. Two girls with only one recall were excluded. Most girls (82%) had at least one weekend recall (n = 250). Analysis of variance testing for differences in dietary intake between weekend and weekday recall indicated no significant differences (P > 0.05).

Measurement characteristics

Factor structure

Principal components analysis with varimax rotation was used to explore the factor structure of two caregiver questionnaires (food preparation and family support for healthy eating) and one questionnaire completed by girls (beverage preferences). Preliminary tests were performed to assess whether data were appropriate for factor analysis, namely Bartlett's test of sphericity and the Kaiser–Meyer–Olkin measure of sampling adequacy. Table II shows a two-factor solution for food preparation (low-fat and high-fat preparation) that accounted for 29% of the total variance. Eight items failed to load significantly on either factor and were removed. This solution was similar to that reported in GEMS Phase 1 [12] although there were some individual item differences. Table III shows a two-factor solution for family support for healthy eating that closely corresponds to concepts of ‘supportive behavior’ and ‘non-supportive behavior’ described by Sallis et al. [32]. Total explained variance was 53%. One item (Ate the same foods as she ate) did not load on either factor and was removed. For the beverage preferences questionnaire, no clear factor structure emerged.

Table II.
Factor structure for ‘food preparation’ questionnaire (n = 301)
Table III.
Factor structure for ‘family support for healthy eating’ questionnaire (n = 303)

Internal consistency

Internal consistency coefficients are presented in Table IV. Among girls’ measures, the self-efficacy for healthy eating and outcome expectancies for healthy eating scales had the highest internal consistency estimates, Cronbach's α = 0.82 and 0.72, respectively. In addition to no clear factor structure, the beverage preferences instrument showed poor internal consistency, Cronbach's α = 0.53.

Table IV.
Descriptives and internal consistency estimates (Cronbach's α) for girl and caregiver measures

Among caregiver measures, FJV availability was the most internally consistent (Cronbach's α = 0.85). Most frequently available items were corn (85%), green beans (85%), lettuce (79%) and orange juice (74%). One item (French fries) was excluded from calculations because it was not counted in the NDS-R vegetable servings calculation. Internal consistency of the six-item FJV accessibility subscale was low at 0.60.

For the high-fat and low-fat food availability subscales, internal consistency was 0.67 and 0.72, respectively. High-fat foods were generally more available than low-fat items. The most frequently available high-fat foods, which ranged from 80 to 85%, were cheese; snack chips; cookies, pies and cakes; and margarine or butter. Reduced fat (2%) milk was the most available low-fat alternative (48%). For the food preparation scale, internal consistency was 0.59 for high-fat preparation and 0.69 for low-fat preparation. Within the family support for healthy eating questionnaire, the six-item supportive behavior subscale had good internal consistency, with a Cronbach's α of 0.83. The non-supportive behavior subscale was less reliable, with a Cronbach's α of 0.60.


Potential confounding effects of sociodemographic characteristics and response bias were addressed before calculating scale intercorrelations for measures that showed acceptable internal consistency (≥0.70). Bivariate Pearson correlations and partial correlations controlling for girl's age, caregiver education and household income were examined. For girls’ measures, associations were also examined for potential social desirability response bias. Examination of partial correlations showed that adjusting for demographics and social desirability did not substantially alter direction or strength of relationships, so unadjusted correlations are reported.

Among girls’ measures, self-efficacy for healthy eating was positively associated with outcome expectancies for healthy eating (r = 0.26, P ≤ 0.01). Among caregiver measures, low-fat food availability correlated both with FJV availability (r = 0.38, P ≤ 0.01) and supportive behavior for healthy eating (r = 0.17, P ≤ 0.01). There were no significant correlations between girls’ and caregivers’ measures.

Concurrent validity

Only scales and subscales with acceptable internal consistency (Cronbach's α ≥ 0.70) were examined for their association with outcome variables. Table V shows that higher self-efficacy for healthy eating scores were significantly associated with lower total energy intake (r = −0.17, P ≤ 0.01) and grams of fat intake (r = −0.16, P ≤ 0.01). Caregiver-reported FJV availability was positively related to girls’ FJV intake (r = 0.14, P ≤ 0.01). A significant negative correlation between FJV availability and BMI (r = −0.12, P ≤ 0.05) indicates that caregivers of girls with lower BMI reported greater availability of FJV. The significant positive correlation between supportive behavior for healthy eating and BMI (r = 0.41, P ≤ 0.01) indicates that caregivers of heavier girls reported greater support.

Table V.
Pearson correlations among dietary psychosocial scales, dietary intake and BMI


Findings from this study add to a small literature on dietary psychosocial scales validated for use among African-American girls and their families. The scales reported here were specifically adapted or designed to measure dietary attitudes and practices in African-American girls and caregivers participating in a weight gain prevention intervention.

Girls’ measure of self-efficacy for healthy eating demonstrated strong internal consistency in this population, comparable to that found for the normative group evaluated by Reynolds et al. (>0.80) [23], and the measure of outcome expectancies for healthy eating showed slightly higher internal consistency than in the original High 5 study (0.72 versus 0.67) [10]. In addition, self-efficacy for healthy eating was associated with lower total energy intake and fewer grams of fat intake. This finding for preadolescent African-American girls is similar to results that link self-efficacy with children's eating behaviors [3436]. However, although significant, the proportion of variance explained for these two criterion variables was quite low (3%). Moreover, self-efficacy for healthy eating was not significantly correlated with FJV consumption, a relationship that has been previously reported [37], nor was it related to sweetened beverage intake.

Among caregiver measures, high internal consistencies were shown for two adapted measures: FJV availability and the family support for healthy eating subscale corresponding to supportive behavior (Cronbach's α = 0.85 and 0.83, respectively). FJV availability had higher internal consistency than that reported in GEMS Phase 1 (0.85 versus 0.77) [22]. Moreover, the supportive behavior subscale mimicked Sallis and colleagues’ findings of Cronbach's α > 0.80 in ethnically diverse samples [32]. As supported by recent literature, children's dietary intake is influenced by what foods are available and accessible in the environment [28, 3840]. FJV availability showed a positive association with self-reported FJV intake although the amount of variance explained was low (2%). No relationship was found between the supportive behavior subscale and girls’ dietary intake. However, supportive behavior was positively associated with girls’ BMI, so the question of construct validity arises with this measure. Some have speculated that parental concern over a child's weight status can lead to more restrictive child feeding practices [41]. With this in mind, it is conceivable that the caregiver-reported supportive behavior subscale is tapping other constructs, too, e.g. ‘control’ or ‘restrictiveness’ [42]. Research has shown that attempts to restrict children's food intake, however well intentioned, may promote the development of undesirable eating behaviors [39, 43] and impede development of dietary self-control [44, 45]. Overall, caregivers of heavier girls may verbally encourage healthier eating, yet perhaps inadvertently derail these efforts by attempting to exert control. Both the positive association of BMI with the supportive behavior subscale and the negative association of BMI with FJV availability suggest that this is a plausible explanation. In fact, none of the caregiver measures showed any significant relationship with girls’ self-efficacy for healthy eating. Perhaps the child's perception of caregiver behavior, rather than the caregiver's perception, would have been a better predictor [37]. A family support measure from the girls’ perspective may have proven informative, but was not used in this study.

One subscale from an instrument developed during GEMS Phase 1 showed some improvement in internal consistency. Low-fat food availability improved from 0.64 in GEMS Phase 1 to 0.72 in GEMS Phase 2. Internal consistency for the high-fat food availability subscale remained below an acceptable threshold of 0.70.

Sweetened beverage consumption has been associated with increased energy intake and higher BMI in children [4648], and preference has been reported as one of the strongest predictors of children's consumption [49]. However, no clear factor structure emerged for the beverage preferences measure. One potential limitation for this measure is item selection in that the items may not have been adequately representative of separate sweetened and non-sweetened beverage constructs. A high percentage of girls had never tried some of the listed beverages and thus could not indicate preference. Further work is needed to develop a useful measure for beverage preferences in this population.

While these results provide unique information about dietary psychosocial constructs in a robust sample of preadolescent African-American girls, the study has limitations. First, although neither social desirability nor age was found to be influential, psychosocial measures may have been susceptible to unmeasured sources of variability such as lack of knowledge and developmental differences in cognitive ability, experience or cultural influence. Finally, results may have limited generalizability for other African-American populations (e.g. males and other age groups) or for Caucasian and other ethnic groups.

In summary, quality instruments are essential to identifying intervention influences on children's dietary beliefs and behaviors. Results of this study contribute to the development of scales to assess SCT-based prevention interventions related to dietary intake in African-American children. One of the girls’ psychosocial measures and one caregiver measure show promise in assessing dietary psychosocial constructs in African-American girls. The measure of self-efficacy for healthy eating showed encouraging associations with total energy and grams of fat intake and the caregiver-reported measure of FJV availability appears to be a useful measure owing to the positive association with FJV intake in the girls. Further refinement of existing scales and development of new instruments will help improve the assessment of mediators in interventions targeted at obesity prevention in this understudied population.


National Heart, Lung and Blood Institute, National Institutes of Health (U01 HL62662, U01 HL62663).

Conflict of interest statement

None declared.


1. Ogden CL, Flegal KM, Carroll MD, et al. Prevalence and trends in overweight among US children and adolescents, 1999–2000. JAMA. 2002;288:1728–32. [PubMed]
2. World Health Organization (WHO) Obesity: Preventing and Managing the Global Epidemic. Report of a WHO Consultation on Obesity, 3–5 June 1997, Geneva. Geneva, Switzerland: World Health Organization; 2001. WHO/NUT/NCD 98.1.
3. Ogden CL, Carroll MD, Curtin LR, et al. Prevalence of overweight and obesity in the United States, 1999–2004. JAMA. 2006;295:1549–55. [PubMed]
4. Crawford PB, Story M, Wang MC, et al. Ethnic issues in the epidemiology of childhood obesity. Pediatr Clin North Am. 2001;48:855–78. [PubMed]
5. Cossrow N, Falkner B. Race/ethnic issues in obesity and obesity-related comorbidities. J Clin Endocrinol Metab. 2004;89:2590–4. [PubMed]
6. US Department of Health and Human Services (USDHHS) The Surgeon General's Call to Action to Prevent and Decrease Overweight and Obesity. Rockville, MD: US Department of Health and Human Services, Public Health Service, Office of the Surgeon General; 2001.
7. NHLBI Study Group. Obesity and cardiovascular disease risk factors in black and white girls: the NHLBI Growth and Health Study. Am J Public Health. 1992;82:1613–20. [PubMed]
8. Obarzanek E, Pratt CA. Girls health Enrichment Multi-site Studies (GEMS): new approaches to obesity prevention among young African-American girls. Ethn Dis. 2003;13:1–5. [PubMed]
9. Tercyak KP, Tyc VL. Opportunities and challenges in the prevention and control of cancer and other chronic diseases: children's diet and nutrition and weight and physical activity. J Pediatr Psychol. 2006;31:750–63. [PubMed]
10. Reynolds KD, Franklin FA, Binkley D, et al. Increasing the fruit and vegetable consumption of fourth-graders: results from the high 5 project. Prev Med. 2000;30:309–19. [PubMed]
11. Cole K, Waldrop J, D'Auria J, et al. An integrative research review: effective school-based childhood overweight interventions. J Spec Pediatr Nurs. 2006;11:166–77. [PubMed]
12. Cullen KW, Baranowski T, Klesges LM, et al. Anthropometric, parental, and psychosocial correlates of dietary intake of African-American girls. Obes Res. 2004;12:20S–31S. [PubMed]
13. Spruijt-Metz D, Li C, Cohen E, et al. Longitudinal influence of mother's child-feeding practices on adiposity in children. J Pediatr. 2006;148:314–20. [PubMed]
14. Molaison EF, Connell CL, Stuff JE, et al. Influences on fruit and vegetable consumption by low-income Black American adolescents. J Nutr Educ Behav. 2005;37:246–51. [PubMed]
15. Bandura A. Social cognitive theory: an agentic perspective. Annu Rev Psychol. 2001;52:1–26. [PubMed]
16. Baranowski T, Klesges LM, Cullen KW, et al. Measurement of outcomes, mediators, and moderators in behavioral obesity prevention research. Prev Med. 2004;38:1–13. [PubMed]
17. Contento IR, Randell JS, Basch CE. Review and analysis of evaluation measures used in nutrition education intervention research. J Nutr Educ Behav. 2002;34:2–25. [PubMed]
18. De Bourdeaudhuij I, Klepp KI, Due P, et al. Reliability and validity of a questionnaire to measure personal, social and environmental correlates of fruit and vegetable intake in 10-11-year-old children in five European countries. Public Health Nutr. 2005;8:189–200. [PubMed]
19. Glasgow RE, Klesges RC, Godding PR, et al. Evaluation of a worksite-controlled smoking program. J Consult Clin Psychol. 1984;52:137–8. [PubMed]
20. Robinson TN, Kraemer HC, Matheson DM, et al. Stanford GEMS phase 2 obesity prevention trial for low-income African-American girls: design and sample baseline characteristics. Contemp Clin Trials. 2008;29:56–69. [PMC free article] [PubMed]
21. U.S. Department of Health and Human Services, Centers for Disease Control and Prevention. CDC Growth Charts. Hyattsville, MD: National Center for Health Statistics; 2000.
22. Cullen KW, Klesges LM, Sherwood NE, et al. Measurement characteristics of diet-related psychosocial questionnaires among African-American parents and their 8- to 10-year-old daughters: results from the Girls’ health Enrichment Multi-site Studies. Prev Med. 2004;38:34–42. [PubMed]
23. Reynolds KD, Yaroch AL, Franklin FA, et al. Testing mediating variables in a school-based nutrition intervention program. Health Psychol. 2002;21:51–60. [PubMed]
24. Reynolds CR, Paget KD. National normative and reliability data for the revised Children's Manifest Anxiety Scale. School Psychol Rev. 1983;12:324–36.
25. Klesges LM, Baranowski T, Beech B, et al. Social desirability bias in self-reported dietary, physical activity and weight concerns measures in 8- to 10-year-old African-American girls: results from the Girls Health Enrichment Multisite Studies (GEMS) Prev Med. 2004;38:78–87. [PubMed]
26. Cullen KW, Baranowski T, Rittenberry L, et al. Child-reported family and peer influences on fruit, juice and vegetable consumption: reliability and validity of measures. Health Educ Res. 2001;16:187–200. [PubMed]
27. Marsh T, Cullen KW, Baranowski T. Validation of a fruit, juice, and vegetable availability questionnaire. J Nutr Educ Behav. 2003;35:93–7. [PubMed]
28. Hearn M, Baranowski T, Baranowski J, et al. Environmental influences on dietary behavior among children: availability and accessibility of fruits and vegetables enable consumption. J Health Educ. 1998;29:26–32.
29. Cullen KW, Baranowski T, Rittenberry L, et al. Social-environmental influences on children's diets: results from focus groups with African-, Euro- and Mexican-American children and their parents. Health Educ Res. 2000;15:581–90. [PubMed]
30. Kristal AR, Shattuck AL, Henry HJ. Patterns of dietary behavior associated with selecting diets low in fat: reliability and validity of a behavioral approach to dietary assessment. J Am Diet Assoc. 1990;90:214–20. [PubMed]
31. Dennison BA, Jenkins PL, Rockwell HL. Development and validation of an instrument to assess child dietary fat intake. Prev Med. 2000;31:214–24. [PubMed]
32. Sallis JF, Grossman RM, Pinski RB, et al. The development of scales to measure social support for diet and exercise behaviors. Prev Med. 1987;16:825–36. [PubMed]
33. SPSS for Windows, Rel. 15.0.1. Chicago: SPSS Inc.; 2006.
34. Ievers-Landis CE, Burant C, Drotar D, et al. Social support, knowledge, and self-efficacy as correlates of osteoporosis preventive behaviors among preadolescent females. J Pediatr Psychol. 2003;28:335–45. [PubMed]
35. O'Dea JA, Wilson R. Socio-cognitive and nutritional factors associated with body mass index in children and adolescents: possibilities for childhood obesity prevention. Health Educ Res. 2006;21:796–805. [PubMed]
36. Robinson CH, Thomas SP. The interaction model of client health behavior as a conceptual guide in the explanation of children's health behaviors. Public Health Nurs. 2004;21:73–84. [PubMed]
37. Young EM, Fors SW, Hayes DM. Associations between perceived parent behaviors and middle school student fruit and vegetable consumption. J Nutr Educ Behav. 2004;36:2–8. [PubMed]
38. Bandini LG, Cyr H, Must A, et al. Validity of reported energy intake in preadolescent girls. Am J Clin Nutr. 1997;65:1138–41. [PubMed]
39. Birch LL, Fisher JO. Development of eating behaviors among children and adolescents. Pediatrics. 1998;101:539–49. [PubMed]
40. Corwin SJ, Sargent RG, Rheaume CE, et al. Dietary behaviors among fourth graders: a social cognitive theory study approach. Am J Health Behav. 1999;23:182–97.
41. Costanzo PR, Woody EZ. Domain-specific parenting styles and their impact on the child's development of particular deviance: the example of obesity proneness. J Soc Clin Psychol. 1985;3:425–45.
42. Moens E, Braet C, Soetens B. Observation of family functioning at mealtime: a comparison between families of children with and without overweight. J Pediatr Psychol. 2007;32:52–63. [PubMed]
43. Birch LL, Davison KK. Family environmental factors influencing the developing behavioral controls of food intake and childhood overweight. Pediatr Clin North Am. 2001;48:893–907. [PubMed]
44. Birch LL, Fisher JO, Grimm-Thomas K, et al. Confirmatory factor analysis of the Child Feeding Questionnaire: a measure of parental attitudes, beliefs and practices about child feeding and obesity proneness. Appetite. 2001;36:201–10. [PubMed]
45. Johnson SL, Birch LL. Parents’ and children's adiposity and eating style. Pediatrics. 1994;94:653–61. [PubMed]
46. Blum JW, Jacobsen DJ, Donnelly JE. Beverage consumption patterns in elementary school aged children across a two-year period. J Am Coll Nutr. 2005;24:93–8. [PubMed]
47. Cullen KW, Ash DM, Warneke C, et al. Intake of soft drinks, fruit-flavored beverages, and fruits and vegetables by children in grades 4 through 6. Am J Public Health. 2002;92:1475–8. [PubMed]
48. Ludwig DS, Peterson KE, Gortmaker SL. Relation between consumption of sugar-sweetened drinks and childhood obesity: a prospective, observational analysis. Lancet. 2001;357:505–8. [PubMed]
49. Grimm GC, Harnack L, Story M. Factors associated with soft drink consumption in school-aged children. J Am Diet Assoc. 2004;104:1244–9. [PubMed]

Articles from Health Education Research are provided here courtesy of Oxford University Press