Preliminary Validation of the Eating Behaviors and Attitudes Inventory (EBAI)

The purpose of the present research study is to provide preliminary evidence of validity for the Eating Behaviors and Attitudes Inventory (EBAI) as an educational device and screening tool. The ultimate goal for the development of the EBAI is to provide an instrument that could be administered by 8-12th grade health education teachers or guidance counselors to facilitate learning about healthy and unhealthy self-regulatory eating behaviors and attitudes among adolescent women. A secondary goal for the EBAI is for the instrument to be utilized by counselors to help college women in a counseling setting to help them fine-tune and shape healthy eating patterns. The EBAI is an instrument which was initially developed through sampling college women in order to identify behavioral mechanisms and attitudes women employ to self-regulate eating behaviors when faced with three levels of eating stress, (a) baseline, (b) 3-5 day disturbance, and c) 10 lb. weight change. The item development of the EBAI yielded a total of 84 items: 60 behavior and 24 attitude items. The EBAI is comprised of two sections, the Assessment of Eating Behaviors (Part I) and the Assessment of Eating Attitudes (Part 2). The EBAI's format calls for a yes/no response to each of the behavior and attitude items. Participants are directed to recall three episodes of eating stress and to complete the instrument accordingly. The sample for the preliminary validation study consisted of a combined, assessable, and heterogeneous sample of 202 undergraduate women enrolled at three academic institutions. The participants were 18 to 44 in age and the mean age was 20.6 years. The sample was also comprised of: 40% freshmen, 17% sophomores, 26% juniors, and 17% seniors. In regards to race, 45% of the individuals self-identified themselves as majority participants and 55% of the individuals self-identified themselves as minority participants. One subsample was composed of 100 undergraduate female African American students who were enrolled in an introductory psychology course at a large historically African American university in the Southeast. A second subsample consisted of 42 undergraduate female cadets who were enlisted in an introductory psychology course and attended a military academy in the Northeast. The third subsample was comprised of 60 undergraduate women who were enrolled in either a science of nutrition, communication and human relations, or introduction to educational psychology course at a large state university in the Southeast. During the last 30 minutes of a class period within their respective course, each of the participants received and completed a packet consisting of informed consent, a background data sheet, and the EBAI. This study addressed six research questions in order to ascertain the descriptive attributes, factorial validity, concurrent validity, and discriminant validity of the EBAI. First, frequency counts of behavior and attitudes items endorsed at each of the three levels of eating stress were conducted. Then, a frequency count of the total mechanisms selected per individual was performed. Next, an exploratory factor analysis, using a principal components analysis with varimax rotation, was conducted on both the behavior and attitude checklists across three levels of eating stress. Then, a Pearson Product Moment correlation was used to ascertain the strength of the relationship between the participant's BMI and factor scores on the both the behavior and attitude checklists of the EBAI. A t-test was also performed to compare the number of behavior strategies and attitude items employed by majority and minority participants. Finally, a chi square analysis was administered to contrast the endorsement between majority and minority participants for each item. The results show that all items of the EBAI are useful, as each item was endorsed by at least one participant. Also, a four-factor structure model for both the Behavior and Attitude Checklists emerged. The factors were labeled control and monitoring, self-detachment, extreme behaviors and action for the Behavior Checklist and positive emotions, negative emotions, guilt, and motivation to change for the Attitude Checklist. A Pearson Product Moment correlation showed some relationship between participants' BMI score and factor scores. The majority participants endorsed significantly more behavior and attitude items than the minority participants. However, the individual items did not discriminate between race and therefore no bias was present at the item level. Finally, suggestions for further research development and validation of the EBAI and implications for practice are discussed.