The purpose of this study was to design the Abstinence Motivation Scale-Sex (AMS-Sex), a scale to assess adolescent motivations for sexual abstinence using the self-determination theory (SDT) motivation continuum as a framework. Previous sexual abstinence research, SDT measurement studies, pilot study data, and expert panel feedback were used to develop the initial 77-item AMS-Sex. The sample consisted of 695 adolescents (aged 15 to 18; M age = 16.49, SD = 1.08; 50.8% male) recruited online from the across the U.S. Participants completed an online survey. The sample was randomly split into two samples. Independent samples t-tests indicated that sample 1 (N = 351; M age = 16.53, SD = 1.09; 48% male) and sample 2 (N = 344; M age = 16.46, SD = 1.07; 51% male) did not significantly differ by age, gender, parent education, annual household income, sexual behavior, or social desirability. Sample 1 data was used to assess scale reliabilities and run exploratory factor analyses (EFA) to examine the factor structure of the 77-item AMS-Sex. The 77-item AMS-Sex had good internal consistency (α = .99), as did the four subscales. As anticipated, the EFA resulted in a four-factor structure. However, these factors did not represent the anticipated SDT motivation types. Rather, the autonomous motivation items loaded on a single factor; the controlled motivation items loaded onto three factors by source or reference (i.e., parents, peers/others, and self). A second EFA was examined, constraining the model to two factors (i.e., controlled, autonomous). The 77 AMS-Sex items were reduced to a final 24-item AMS-Sex measure with six items per motivation type through a rigorous process using confirmatory factor analysis (CFA) with modification indices and information from CFA loadings, item means, standard deviations, skewness, correlations with social desirability, EFA loadings, and item wording and clarity. Sample 2 was then used to validate the final 24-item AMS-Sex using CFA and structural equation modeling (SEM). CFA was used to assess the four-factor, three-factor, and two-factor models. Model fit indices indicated that the four-factor and three-factor models had acceptable model fit, but the two-factor model did not. SEM was used to assess the relationships between the factors (i.e., types of motivation) and age, gender, social desirability, and sexual behavior for the four-factor, three-factor, and two-factor models. The SEM models indicated that older teens tended to be less motivated towards abstinence; girls were more motivated towards abstinence than boys; and sexual behavior was negatively correlated with all four motivation types, but more strongly with the autonomous motivations; and, in general, autonomous motivations were negative predictors of sexual behavior, while controlled motivations positively predicted sexual behavior. EFAs with the final 24 items were assessed, resulting in a three-factor solution: external, introjected, and autonomous motivations.



College and Department

Family, Home, and Social Sciences; Psychology



Date Submitted


Document Type





abstinence, motivation, adolescents

Included in

Psychology Commons