Development and validation of a condom self-efficacy scale for college students
Brafford, L.J.; Beck, K.H.
Journal of American college health J of ACH 39(5): 219-225
1991
ISSN/ISBN: 0744-8481 PMID: 1783705 DOI: 10.1080/07448481.1991.9936238Document Number: 318775
The methods and validation results of the development of a condom self-efficacy scale (CUSES) for college students is reported. Development of the scale was based on previous literature, an expert panel, and input form 183 college students. 60 items were included in the initial questionnaire and 15 factors were identified as influencing college students; self-efficacy toward condom usage. 28 specific self-efficacy items were then created with a scale ranging from strongly agree to disagree. The 15 factors were 1) personal experience with condoms, 2) embarrassment at purchase, 3) alcohol use, 4) not wanting to offend with the implication of uncleanliness, 5) reduction in excitement, 6) loss of spontaneity, 7) breaking the mood, 8) not being prepared, 9) unsure of partner's feeling about condoms, 10) embarrassment, 11) communication with partner, 12) embarrassment about talking about condoms, 13) afraid of partner's refusal of condom use, 14) ability to maintain an erection, and 15) fear of reputation. The entire survey tested had 124 questions: the 28-item new scale, Brown's Attitude Toward Condom Use (ATC) scale, Levinson's Contraceptive Self Efficacy Scale for women (CSE), a short form of the Marlow-Crowne Social Desirability scale, and 3 outcome measures. Changes were made after 60 students pretested the questionnaire for readability, understanding, and anonymity. Validity scale measures are reported for the other scales used. 803 students were administered the questionnaire during class, but not sex education classes, and 2 weeks later. The demographics are given and were similar to the pretest. The mean score of CUSES was 75.69 )SD=15.50). Cronbach's alpha with 768 cases was .91 and after 2 weeks .81 and are comparable to other reliability coefficients. CUSES correlated significantly with the ATC scale (r=.51, p.001) and the CSE (r=.55, p.001) with evidence of convergent validity. It was significantly correlated with behavioral intentions (r=.40, p.001) and outcome efficacy score of condom users was M=81.31, n=260, and nonusers was M=74.91, n=407, and users mean was significantly greater. Those who had used a condom had significantly higher CUSES scores than those who had not, and those who had a prior sexually transmitted disease (STD) approached a significant difference. There was no significant difference between men and women. A limitation is that the sample was not random, and the measures of behavioral intention and past behavior are self-reports. Although behavioral intention is the best predictor of future behavior, it may not be an accurate measure. The sample was not ethnically or sexually diverse. The confidence with which students felt about some items is discussed and it may be helpful to educators to focus on condom purchase, negotiating condom use with a new partner, and homosexuality and STDs.