| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Supplements |
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONIDENTIFICATION OF CONCEPT AREAS...CREATION, ADMINISTRATION, AND...FUTURE DIRECTIONS AND LESSONS...ConclusionREFERENCES |
|---|
Key Words: American Indians • attitudes • behaviors • children • knowledge • Pathways • schools • questionnaire development • reliability
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONIDENTIFICATION OF CONCEPT AREAS...CREATION, ADMINISTRATION, AND...FUTURE DIRECTIONS AND LESSONS...ConclusionREFERENCES |
|---|
Previous school-based nutrition and physical activity intervention studies in grade-school children have relied on self-reported measures to assess changes in KAB targeted by the intervention. For example, in the Child and Adolescent Trial for Cardiovascular Health (CATCH) (1), intended food choices, knowledge, and self-efficacy regarding healthful eating among children in grades 3–5 were measured by using a classroom-administered questionnaire. Postintervention healthy-food-choice scores were significantly higher in intervention than in control schools (1). Subsequent work corroborated these findings in children in grades 6–9 and showed that self-reported intentions are consistent over time (2, 3). These findings indicate that this type of questionnaire is sensitive to the effects of a school-based intervention, and can be used successfully in an ethnically and socioeconomically diverse student population such as that targeted by CATCH (1). The results from CATCH suggested that self-report classroom-administered questionnaire measures of behaviors and attitudes would be appropriate for use in the Pathways study.
Cross-sectional studies have shown that food consumption and physical activity are influenced by a variety of psychologic, social, cultural, and environmental factors in addition to biological and developmental factors (4, 5). These factors undoubtedly interact, although the importance of each factor and the nature of the interactions are not known because studies of multiple categories of factors are rare. In addition, there has been no systematic research across different age and ethnic groups. Thus, the relative importance of different factors for different groups and at different developmental periods is not known. The Pathways study will collect information on several factors related to food consumption and physical activity that will permit examination of interactions among these factors.
This article describes the development of a questionnaire to assess several aspects of KAB related to healthy eating and physical activity. Several previously developed questionnaires were adapted for use, and some previously tested questions were used intact. However, for the most part, suitable questionnaires were not available and new questions had to be developed. The Pathways KAB questionnaire needed to be culturally and developmentally appropriate for third- to fifth-grade American Indian children and targeted to test the Pathways intervention. Cost, time, and staffing constraints also had to be considered.
|
IDENTIFICATION OF CONCEPT AREAS TO BE MEASURED |
|---|
|
TOPABSTRACTINTRODUCTIONIDENTIFICATION OF CONCEPT AREAS...CREATION, ADMINISTRATION, AND...FUTURE DIRECTIONS AND LESSONS...ConclusionREFERENCES |
|---|
Physical activity
The physical activity portion of the KAB survey was designed to assess factors that were likely to influence a child's present level of activity or the likelihood that they would become active and remain active if exposed to an appropriate intervention. A variety of psychologic factors have been studied in relation to physical activity (4). Individual differences such as achievement motivation, tolerance of stress, social adequacy, movement satisfaction, and independence are either weakly or not associated with physical activity in youth (7, 8). Thus, these factors are probably not strong determinants of children's activity. Similarly, knowledge about fitness and physical activity is an inconsistent correlate of activity (9, 10), though it may have delayed rather than immediate influences.
In contrast, self-efficacy (confidence in one's ability) for physical activity, exercise intentions, and specific perceptions of barriers to activity are potent predictors of activity in adolescents and may be important in children as well. It is also likely that peers and parents have important influences on children's physical activity. For example, both peers and parents have been shown to influence youth sport selection (11) and consistent positive associations have been reported between parental encouragement and activity in young children (12–16). The roles of modeling, reinforcement, environmental prompts, and other potential mechanisms of parental influences have not been well described. Several characteristics of a child's physical environment may help determine physical activity. Time and place factors, such as geographic location and season of the year, are probably important because percentage time spent outdoors is strongly related to physical activity (17). Television viewing, the most commonly studied environmental factor, is strongly correlated with obesity (18), possibly because it replaces more active leisure-time pursuits.
Previous studies have shown an association between physical self-perception and level of participation in physical activities and organized athletics (18, 19) as well as other fitness indicators (20, 21). Physical self-perception in specific domains has also been shown to differ between individuals engaged in various types of physical activities. For example, participation in ball sports is associated with high self-perception in sports skills for male and female young adults (18). Physical self-perception has been reported to change with participation in physical activities, such as aerobics classes and Outward Bound programs (21, 22). In contrast, physical activity and fitness indicators do not appear to be associated with nonphysical domains of self-perception (20).
The choice of specific factors to measure was based on the strength of the reported associations between physical activity and its correlates, the nature of the planned intervention, and the acceptance of the items in all of the American Indian communities involved in Pathways. Four broad categories were chosen for this section of the questionnaire: physical activity self-efficacy, social support for physical activity, perceived barriers to activity, and physical self-perception. Other potential predictors of activity, including body composition and physical environment factors, such as actual availability and proximity of television watching and of play areas, facilities, and programs, were evaluated elsewhere in the study.
Diet
Children's eating behavior can be examined by looking at individual, behavioral, and environmental factors (23). Individual factors are personality dispositions, cognitions, or affective domains that increase or decrease the likelihood of engaging in a given behavior and include self-efficacy, knowledge, functional meanings of food, and perceived body image. Behavioral factors are those that affect actions directly and include one's behavioral repertoire, behavioral intentions, and skills. Environmental factors are those aspects of the environment that support, permit, encourage, or discourage engagement in a particular behavior and include family and peer influences, social support, social norms, and specific opportunities (23). To change eating patterns successfully, interventions must aim to modify all 3 factors.
Over the past decade, several nutrition-related programs have focused their intervention efforts on these 3 factors (5, 24–26). Outcome evaluation in these programs followed the intervention model and measured specific behaviors, knowledge, attitudes, and intentions as well as physiologic outcomes. Outcomes were generally favorable in terms of knowledge, attitudes, and intentions, and were less so in terms of behavior and physiologic change on nutrition-related measures (25, 26). Recent findings from the CATCH study showed significant improvements in dietary intentions, usual food choices, knowledge, and self-efficacy for eating lower-fat and lower-sodium foods in third-grade children in intervention compared with control schools (24). These findings support theory-based and combined individual and environmental interventions for changing health-related risk behaviors in children. Most of the nutrition education programs that resulted in behavior change used teaching strategies based on social learning theory. In such programs, increasing student knowledge is only one of many objectives. The relation between nutrition knowledge and behavior is weak (5, 25). It is well documented that although nutrition knowledge is essential, knowledge alone does not enable young persons to adopt healthy eating behaviors. Thus, behaviorally centered approaches (ie, goal setting, behavioral skill building, and efficacy-enhancing experiences) are needed (5). Furthermore, behavioral changes may not be maintained unless the school environment supports the changes and the home environment reinforces the intentions.
The diet measures chosen for assessment in the Pathways KAB questionnaire were based on specific constructs derived from social learning theory that are considered to be determinants of behaviors underlying obesity-related risk factors. The diet measures were also based on the concepts emphasized in the Pathways intervention, specifically, reducing consumption of dietary fat and sugared beverages and increasing consumption of healthful lower-fat foods. Student outcomes targeted for assessment included knowledge of high-fat foods, self-efficacy to make healthy food choices, perceived social support for eating healthful foods, intentions to choose healthful foods, and self-reported behaviors.
Weight-related attitudes and behaviors
Frequent dieting and exaggerated concern about body weight are common in this country. National data collected in 1990 from 11467 high school students as part of the Youth Risk Behavior Survey (27, 28) estimated that 34% of female students and 15% of male students considered themselves overweight. Forty-four percent of female students and 15% of male students reported that they were trying to lose weight. Students trying to lose weight had used several methods in the 7 d preceding the survey, including skipping meals (49% of females, 18% of males), taking diet pills (4% of females, 2% of males), and vomiting (3% of females and 1% of males). Few studies have examined weight-related attitudes and behaviors in preadolescent children. Mellin et al (29) surveyed middle-class girls in California and found that controlled eating was reported by 45% of third-grade and 80% of fourth- and fifth-grade girls. Inappropriate methods of weight control (vomiting, taking diet pills or laxatives, or skipping meals to lose weight) were reported by 10% of the 9- and 10-y-old girls examined.
We know of no similar data from American Indian children of this age. However, studies in older children and adults indicate that attitudes toward dieting and body size are similar in whites and American Indians (30), and that the use of inappropriate weight-loss strategies may be as prevalent or even more prevalent among American Indians than among whites. Smith and Krejci (31) administered the Eating Disorder Inventory and the Bulimic Test to American Indian, Hispanic, and white high school students (n = 545; mean age: 15.2 y). They found that the 129 American Indian students consistently scored the highest on each of the 7 items testing for disturbed eating behaviors and attitudes. Story et al (32) reported using national survey data from American Indian and Alaska Native children aged 12–18 y. Most participants, particularly girls, were dissatisfied with their weight and were worried about being overweight. Unhealthy weight-control practices were not uncommon.
Some voluntary control of food intake or attention to the selection of foods is appropriate; however, preoccupation with weight and extreme practices such as fasting and vomiting to lose weight are inappropriate at any age and are especially undesirable in children. In the Pathways study we developed questions to assess children's perceived and ideal body size, concern about body size, and prevalence of dieting practices. Our hypothesis is that the Pathways intervention will not be associated with negative changes in children's body image, nor will it increase the prevalence of dieting.
Cultural identity
An ethnic identity scale was developed based on the recommendation of American Indian collaborators in the study. The Pathways intervention was carefully designed to be culturally sensitive, but some measure of cultural identity was required so that we could assess whether the intervention was equally effective in children who identified more or less strongly with their American Indian culture. Measurement of cultural identity in Pathways provided the opportunity to gain information about and insights into the impact of cultural issues on the effectiveness of an intervention in this group.
Ethnic identity is known to be related to the concept of self-esteem (33–35). Phinney (36) remarked that "a positive self-concept may be related to the process of identity formation—that is, to the extent to which people have come to an understanding and acceptance of their ethnicity." Ethnic identity was found in some studies to be related to the level of acculturation to mainstream values and behaviors (37–40). These interrelations were of central concern to this study. Greater ethnic identity and self-esteem may be related to a decreased adoption of dominant American lifestyle attributes in American Indian children.
The only existing literature on developing ethnic identity scales for US children focuses primarily on white ethnic groups and African Americans; there is no such literature on American Indian children. Phinney (36) wrote an excellent review of ethnic identity, but focused on adults and adolescents rather than on children. In fact, the measurement of ethnic identity in children is a difficult proposition considering current models. Many models today consider ethnic identity to be a developmental process, in which individuals go through various life stages. For instance, the Racial Identity Attitude Scale developed by Parham and Helms (41) describes changes in attitudes during the stages of preencounter, encounter, immersion, and internalization. Young children have not been measured with use of these types of models because they are not old enough to have moved through these stages.
|
CREATION, ADMINISTRATION, AND REVISION OF THE KAB QUESTIONNAIRE |
|---|
|
TOPABSTRACTINTRODUCTIONIDENTIFICATION OF CONCEPT AREAS...CREATION, ADMINISTRATION, AND...FUTURE DIRECTIONS AND LESSONS...ConclusionREFERENCES |
|---|
Initial tests with children
To help determine validity of the questionnaire, a subset of 64 questions was pretested with 32 American Indian children by using semistructured interviews. The children were in the third, fourth, or fifth grade. Trained American Indian staff members conducted most of these interviews at the field site. In individual interviews, children were asked each question on the questionnaire. Follow-up probes included, "What do you think this question is asking?," "Does this question make sense to you?," and "Does this question seem important?"
Input from the American Indian members of the Pathways team
Feedback from American Indian colleagues in Pathways was obtained during a 2-d meeting attended by 16 Pathways staff members, 8 of whom were American Indian. The purpose of each section of the questionnaire was reviewed and then items were examined one by one. As a result, 40 more questions were eliminated and the remaining 130 were revised extensively. At the suggestion of some of the American Indians in attendance, the cultural identity section was added to the questionnaire.
After the questionnaire was approved by the Pathways Study Steering Committee, approval was obtained from all 5 tribes without further revisions.
Selection of protocols for administrators
The working group decided to administer the questionnaire in school classrooms by using a protocol patterned after that of the CATCH study (42) that was age appropriate and accommodated differences in reading level. The children marked their own answers to questions, but a trained test administrator read the questions to the class. This dictated that the questionnaire format had to be simple and easy to understand, with no embedded items. Illustrative pictures were included wherever appropriate. A standardized instruction manual for administration of the questionnaire was developed that oriented the administrator to the purpose of the questionnaire and provided detailed, step-by-step instructions on how to administer the questionnaire. The protocol called for 2 staff members to administer the questionnaire in the classroom, one to read the questions aloud and the other to act as proctor, circulating throughout the room to answer questions and keep the children on task. To accommodate differences in reading level it was decided that questions would be read twice to third- and fourth-grade children and once to fifth-grade children. Brief activities (
5 min) that engaged the children were included in the protocol to break up lengthy testing sessions. The test administrator distributed letters explaining the questions and protocol to the teachers before the start of the test session.
Pretesting in the classroom
The 130-item draft of the questionnaire was pretested in 2 classrooms at one of the field centers. It was given by an administrator and a proctor with 1 other staff member present to observe. The questionnaire was administered in 3 sections, each 
30 min in length, over 2–3 d. The children's general reactions to the questionnaire and their observed comfort level upon completion of the questionnaire were positive. Several modifications were made as a result of this testing. This process yielded version 1.0 of the questionnaire.
Description of version 1.0 of the questionnaire
The reading level of the questionnaire was evaluated as second-grade level (grade 2.4) by the RIGHT WRITER (43) and fifth grade level (grade 5.6) by the FLESCH GRADE LEVEL (44) software programs. Each section of the questionnaire is described below.
Physical activity
Scales were developed or adapted to assess the following constructs thought to be mediators or correlates of physical activity: social support, self-efficacy, perceived barriers, and self-perception. The Physical Self-Perception Profile (PSPP) (18) was chosen as the instrument on which to base the physical self-perception portion of the KAB questionnaire. The PSPP is widely used (18), psychometrically sound (17, 45, 46), and has been adapted for use with younger adolescents (45) and children, including fourth and fifth graders (46). In version 1.0 of the KAB, 2 subscales from the Harter (47) Self-Perception Profile for Children (scholastic competence and social acceptance) were also included. This is the best known and most widely used of the general multidimensional self-perception instruments and has good-to-excellent psychometric properties with children as young as third-grade age (48). The domains included from the Harter profile were those most likely to be associated with obesity or to change as a result of participation in a regular physical activity program.
Diet
The specific diet-related psychosocial constructs measured by the KAB questionnaire included dietary self-efficacy, social support for lower-fat food choices, dietary intentions and expectations, dietary-fat knowledge, and reported food consumption. All measures, with the exception of food consumption, were based on the health behavior questionnaire used in the CATCH study (1). These measures have been used in previous work including Hearty Heart (49) and Go for Health (50) or were developed specifically for CATCH (1). Although the KAB diet-related scales were derived from these measures, the questions, food items, and response categories were changed. The specific food items included in the KAB questionnaire were chosen on the basis of the results of a formative assessment of foods commonly consumed by American Indian children.
Weight-related attitudes and behaviors
Questionnaires to measure weight-related attitudes and behaviors in a variety of populations were available from previous studies (28, 50–60). Some questions in this section were used intact from these existing questionnaires; however, most were modified to simplify the language to make the questionnaire more suitable for third- to fifth-grade children. The concepts assessed included body image, concerns about weight, and attempts at weight loss. As a measure of perceived actual, ideal, and healthy body size, children were asked to select from 8 line drawings of children that ranged from very thin to obese. These drawings were conceptually similar to those previously created for use in adults (52). In this section, as in other sections dealing with weight-related attitudes and behaviors, some questions were included to provide balance. For instance, to balance "Would you like to be skinnier than you are now?" we included "Would you like to be chubbier than you are now?"
Cultural identity
The literature indicates that there are 4 potential components of ethnic identity: ethnic self-identification (the label you give yourself), sense of belonging to a particular ethnic group, positive and negative attitudes toward one's own ethnic group, and ethnic involvement (participation in social and cultural practices) (36). We found that the wording used to measure ethnic identity by "sense of belonging" and "positive or negative attitudes" was too complex for third-grade children. Therefore, our scale of ethnic identity included only those components that relate to self-identification and ethnic involvement. In particular, the ethnic involvement component focused on concrete behaviors, such as language spoken and activities practiced.
Test and retest of version 1.0
Test-retest reliability was assessed for version 1.0 of the KAB questionnaire by administering the questionnaire twice to the same fourth-grade children with an interval of 3–6 wk between tests. Fourth-grade children were chosen because they would give an intermediate view of what could be expected in terms of comprehension and performance from children in the third, fourth, and fifth grades. Questionnaire administrators and proctors were trained in a 3-h meeting during which instructions were reviewed and administrators were required to rehearse in front of the group.
The questionnaire was administered in 2 schools at each of 4 sites. Three separate sessions (separated by 
1 h) were conducted over a 2- or 3-d interval. An average of 70 min was required to administer the entire questionnaire. The paper forms were mailed to the coordinating center and all data were entered centrally. A total of 371 fourth-grade children from the 4 study sites took part in the reliability testing; 18% of the questionnaires were incomplete (a questionnaire was considered incomplete if 3 questions were not answered). This relatively high level of incomplete data was due to student absences and student participation in special programs during one or more of the 3 scheduled test times.
Qualitative feedback was requested from each center regarding problems that arose during the test-retest administration. Overall, these reports indicated that the test was well accepted. However, the body-image questions with the graded body shapes were felt to be insensitive, and teasing of overweight children was a problem at some sites. It was also noted that some children in special education programs were unable to keep up with the rest of the class and may not have provided valid data.
Statistics were calculated for each of the study questions as an estimate of repeatability (61). Cronbach's (62) 
statistics were calculated as measures of internal consistency of the scales and correlation coefficients were used to estimate the repeatability of scales (63). The following 8 scales had a Cronbach's and test-retest r 0.5: barriers to physical activity, knowledge of high-fat foods, diet self-efficacy, diet behaviors, food frequency, body image, attempts at weight loss, and cultural identity.
Revisions based on reliability data
The questionnaire was revised on the basis of the results of the first reliability testing. Some questions were dropped and others were added. In some cases, new response options were created and tested in an effort to improve repeatability. Selected examples of changes are presented below.
For the physical self-perception section, each subscale contained only 2 items, so internal consistency could not be assessed. To address this problem, the working group decided to include more items to assess each dimension. The scholastic competence and social acceptance scales were judged to be of little relevance to the central aims of the Pathways study and were eliminated. The response format was revised from that of the original PSPP to a Likert (64) format, consistent with the format used in the first KAB testing. Question wording was also revised to improve clarity. The revised version of the self-perception section of the KAB contained six 4-item subscales consisting of questions adapted from the original PSPP.
Neither the physical activity nor the diet social support scales performed adequately. A new 4-point response format (most of the time, sometimes, once in a while, and almost never) was developed and pilot tested that replaced the original 3-choice response set (most of the time, sometimes, and never or almost never). The food-frequency scale was modified because it appeared that children had difficulty conceptualizing the time frame associated with usual intake. In the revised version of the KAB questionnaire, specific foods were presented and children were asked to circle the picture if they had eaten the food the day before; thus, the time frame was more limited and concrete.
To identify the sources of incomplete questionnaires, procedures were put into place to identify children in special education programs, a protocol was written for obtaining data from children who were absent for a portion of the test, and a form was developed to collect information on the amount of effort required to produce higher rates of completion. On the basis of time logs from the first test and retest, it was decided to administer the test in 2 sessions rather than 3 to reduce the staff-associated costs of the administration.
Review of revised questionnaire
The revised questionnaire was sent out to the American Indian members of the Pathways team for review. After approval by that group and the Pathways Steering Committee, the revised questionnaire, version 2.0, was reviewed and approved by the participating tribes.
Test and retest of version 2.0
Sample questions from each section of version 2.0 of the KAB questionnaire are shown in Table 1
. This version was tested for repeatability between October and December of 1995. Training of the administrators and proctors was conducted in a 2-h conference call. Data were entered at the site rather than centrally for version 2.0, and an additional 2-h conference call was held to train field staff to enter the KAB data at the site. The completion rate of the questionnaires was >95% because Pathways staff members revisited schools to obtain missing data. The test was completed in 80 min in 2 sessions.
|
statistic by section are shown in Table 2, as well as the number of questions, Cronbach's , and r for repeatability for each scale. The correlation is the appropriate statistic to use for the assessment of the repeatability of scales; however, the average of the questions as used in the scales is also shown for descriptive purposes (61, 63). Because one of the goals of the testing was to produce a more concise questionnaire, efforts were made to reduce the number of items in some scales. The mean and Cronbach's statistics were recalculated after deleting poorly performing or unnecessary questions or after collapsing response options. Six of the 8 scales with both a Cronbach's and test-retest r >0.5 in the first version performed equally well in the second version (barriers to physical activity, knowledge of high-fat foods, diet self-efficacy, diet behaviors, body image, and attempts at weight loss). The repeatability and internal consistency of the physical activity self-efficacy scale improved such that the internal consistency and repeatability tests met the >0.5 criterion value. However, the performance of the food-frequency and cultural-identity scales declined and no longer met this criterion. Generally, the modifications in the questionnaire resulted in only small changes in the repeatability and internal consistency of the scales.
|
|
FUTURE DIRECTIONS AND LESSONS LEARNED |
|---|
|
TOPABSTRACTINTRODUCTIONIDENTIFICATION OF CONCEPT AREAS...CREATION, ADMINISTRATION, AND...FUTURE DIRECTIONS AND LESSONS...ConclusionREFERENCES |
|---|
was 0.84 and the test-retest r was 0.63 (60). In Pathways, the respective results were 0.76 and 0.59 for the 12-item scale tested. More research is needed on methods of questionnaire administration in children of this age. As noted above, an interesting problem emerged for items that required reverse scoring, apparently because they were negatively worded. A negatively worded item is one in which a negative response is keyed positively (65). For example, "Some kids do not like the way their body looks" is a negatively worded item from the KAB questionnaire. Such items are included in psychosocial questionnaires to guard against a tendency to answer affirmatively to all items. Children completing version 2.0 of the KAB questionnaire responded to such items as though they were positively worded (eg, "Some kids like the way their body looks"). Previous literature suggests similar problems when administering some psychologic questionnaires to children in this age group (66). Taken together with the present results, these findings suggest that negatively worded items may not be appropriate for use in the Pathways study and that other strategies for reducing response-set bias should be investigated.
Cultural issues
The poor performance of the negatively worded questions was likely related to the young age of our sample, but cultural issues may also have contributed. Other questions we found to be problematic were those designed to ascertain information about attitudes or behaviors that could be termed undesirable. Because it is intrinsic to many American Indian cultures to think and speak in a positive way and to avoid thinking or speaking in a negative way (67), questions that elicit negative thinking are inappropriate. In traditional Navajo culture, it is held that thoughts and spoken words have the power to shape reality and to control events. Therefore, asking a question about negative feelings or behaviors is believed to potentially cause those negative feelings and behaviors. An additional culture-bound belief that made questionnaire development challenging was the tradition of avoiding comparisons between individuals. Rating or ranking individuals is considered unacceptable in several tribes. Thus, questions that required a comparison of oneself to others were not acceptable.
Over the course of the development of the questionnaire we gained an appreciation of the differences between individual opinion and group consensus in the area of cultural input. In the early phases of the study, the KAB working group responded to the suggestions of the individual American Indian members of the team. As time passed and the number of American Indian members on the team grew and the study experienced turnover in staff, it became obvious that there were differences in the culture-bound beliefs between different tribes and between individuals in tribes. We have found it very useful to request formal recommendations from the American Indians on the project as a group rather than individually. Also, we found that occasional face-to-face meetings were essential for discussions of cultural appropriateness.
Our experience underscores the importance of involvement of American Indian members of the Pathways team in every phase of development of the KAB questionnaire. This was important not only in ensuring that the questionnaire was properly worded and culturally appropriate, but in obtaining tribal approval. Too often, instruments are developed for majority children and used with children from different ethnic communities without adequate data on reliability and validity for use within that population, and without input from the community. The American Indian members of the KAB team not only helped to produce a more culturally appropriate instrument, but also facilitated acceptance of the instrument by the larger community.
|
Conclusion |
|---|
|
TOPABSTRACTINTRODUCTIONIDENTIFICATION OF CONCEPT AREAS...CREATION, ADMINISTRATION, AND...FUTURE DIRECTIONS AND LESSONS...ConclusionREFERENCES |
|---|
and r values 0.6. The final questionnaire will be administered annually during the Pathways full-scale intervention trial. We look forward to learning more about the effect of the intervention in areas pertinent to the health of American Indian children.
|
FOOTNOTES |
|---|
2 Supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health (U01-HL-50869, U02-HL-50867, U01-HL-50905, U01-HL-50885, and U01-HL-50907).
3 Address reprint requests to J Stevens, University of North Carolina at Chapel Hill, Department of Biostatistics, CB#8030, Chapel Hill, NC 27514. E-mail: June_Stevens{at}UNC.EDU.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONIDENTIFICATION OF CONCEPT AREAS...CREATION, ADMINISTRATION, AND...FUTURE DIRECTIONS AND LESSONS...ConclusionREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Stock, C. Miranda, S. Evans, S. Plessis, J. Ridley, S. Yeh, and J.-P. Chanoine Healthy Buddies: A Novel, Peer-Led Health Promotion Program for the Prevention of Obesity and Eating Disorders in Children in Elementary School Pediatrics, October 1, 2007; 120(4): e1059 - e1068. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. I. Saksvig, J. Gittelsohn, S. B. Harris, A. J. G. Hanley, T. W. Valente, and B. Zinman A Pilot School-Based Healthy Eating and Physical Activity Intervention Improves Diet, Food Knowledge, and Self-Efficacy for Native Canadian Children J. Nutr., October 1, 2005; 135(10): 2392 - 2398. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Caballero, T. Clay, S. M Davis, B. Ethelbah, B. H. Rock, T. Lohman, J. Norman, M. Story, E. J Stone, L. Stephenson, et al. Pathways: a school-based, randomized controlled trial for the prevention of obesity in American Indian schoolchildren Am. J. Clinical Nutrition, November 1, 2003; 78(5): 1030 - 1038. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Davis, S. Hunsberger, D. M Murray, R. R Fabsitz, J. H Himes, L. K Stephenson, B. Caballero, and B. Skipper Design and statistical analysis for the Pathways study Am. J. Clinical Nutrition, April 1, 1999; 69 (4): 760S - 763S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L Helitzer, S. M Davis, J. Gittelsohn, S. B Going, D. M Murray, P. Snyder, and A. B Steckler Process evaluation in a multisite, primary obesity-prevention trial in American Indian schoolchildren Am. J. Clinical Nutrition, April 1, 1999; 69 (4): 816S - 824S. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
