Characterizing adolescent eating behaviors with Bandura's social learning theory

The purpose of this research was to evaluate the efficacy of Bandura's Social Learning Theory in describing adolescent eating behaviors. Social Learning Theory variables included: outcome expectations for nutrition, dietary self-efficacy, and nutrition knowledge. The criterion variable was quality of diet, as determined by computer analysis and dietary score.

Participants in the study were 136 high school students (93% response rate) with a mean age of 16.1 (SD=7.28). Forty-one percent were female, 51.1% were white and 36.8% black.

Two instruments were created: The Nutrition Self-Efficacy Scale (NSES) and The Food Choices survey. The NSES measured Social Learning Theory components and collected demographic information. Validity and reliability assessments were conducted after instrument review, and pilot testing. The Food Choices Survey consisted of a 24-hour dietary recall.

Students were asked to rank their confidence in choosing healthy foods in 21 situations. This subscale measured dietary self-efficacy. There was a significant difference between mean dietary analysis scores for individuals who believed they could choose healthy foods and those who did not. The students who perceived themselves as more confident in their ability to choose healthy foods scored significantly higher (p < .05) on the DINE analysis (M=3.075 ±·227) when compared to those students who perceived themselves as less confident (M=1.750 ±.629).

Pearson correlation coefficients were used to determine if positive self-efficacy was consistent (generalized) across all food choice situations. There were significant relationships (p < .001) between the self-efficacy ratings across all food choice situations, indicating generalizability for dietary self-efficacy.

There were no significant differences between mean DINE computer scores for students with high and low degrees of importance placed on priority goals, or with strong and weak beliefs that nutrition affected those priority goals.

Participant were placed in three groups according to their level of dietary self-efficacy (high, medium, or low). The mean nutrition knowledge score did not differ significantly for groups with increasing levels of dietary self-efficacy. Dietary self-efficacy was a predictor variables for adolescent food choices in this study; however, the variance for which it accounted was low.

Individuals with higher dietary self-efficacy also achieved a higher dietary quality score. Thus, it appears at least one component of Social Learning Theory can be used to characterize adolescent eating patterns.