Scheduled for Poster Session: Socio-, Cross-Cultural, and Motivational Concerns Impacting Sport and School Contexts, Thursday, April 10, 2008, 1:15 PM - 2:45 PM, Convention Center: Exhibit Hall, Reseach Consortium Poster Sessions


Assessing Clinical Treatment of Childhood Obesity Using Mixed Methods

Karen A. Werner1, Brian F. Geiger2, Mike A. Perko3, David Macrina4 and Marcia O'Neal4, (1)Centers for Disease Control and Prevention, Atlanta, GA, (2)UAB School of Education, Bessemer, AL, (3)University of Alabama, Tuscaloosa, AL, (4)University of Alabama-Birmingham, Birmingham, AL

Mixed methods research combines qualitative and quantitative approaches within a single study. A combined approach enables researchers to maximize the strengths while reducing the weaknesses of both methods. The researcher utilized a concurrent sequential design (Tashakkori & Teddlie, 2003) to inform the development of a quantitative cross-sectional survey. Survey content was based on the Systems Model of Clinical Preventive Care (Walsh & McPhee, 1992). The intent was to measure clinical practices of Alabama family physicians' and pediatricians' who treat pediatric obesity. The researcher completed a thorough process to estimate the reliability and validity of the new instrument following a modified protocol as suggested by Dillman (2002). The researcher conducted a literature review and three focus groups with Alabama family physicians and pediatricians. The next step was to solicit review and comments from eight expert panelists. The researcher modified survey items based on suggestions from these panelists. The third phase was to administer the pilot survey to 74 physicians across four settings. The reliability of the pilot survey was determined using test-retest and internal consistency measures. The alpha coefficients for hypothesized subscales ranged from 0.71 to 0.88. The researcher administered the identical survey to a small group of physicians one week apart, and then examined the correlation between the test-retest for each of the hypothesized subscale scores using Pearson's correlation. Correlations for each of the hypothesized subscales were moderate, ranging from 0.51 to 0.77 (p<0.01). The researcher solicited perceptions from 10 physicians who reviewed survey content and format during think-aloud interviews. The researcher edited the title of the survey, terminology, explanation of informed consent, item stems, and response scales based on these interviews. The last step before statewide dissemination was final review by three professionals who were otherwise uninvolved in the study. The researcher used a cross-sectional research design to distribute the survey to 1,662 Alabama pediatricians, family physicians, and general practitioners via a mass faxing service. Surveys were also distributed to 112 family physicians who attended a statewide professional meeting held in December of 2006. Eighty-six eligible physicians completed the survey, i.e., 62 from the mass faxing and 24 from the statewide meeting. Data from this study can be used to design training opportunities for family physicians and pediatricians and to plan interventions for children and families in the primary care setting.


Keyword(s): measurement/evaluation, obesity issues

Back to the 2008 AAHPERD National Convention and Exposition (April 8 - 12, 2008)