Psychometric testing of the Chinese version of the Contraceptive Behavior Scale: a preliminary study
Article first published online: 19 JUL 2011
© 2011 Blackwell Publishing Ltd
Journal of Clinical Nursing
Volume 22, Issue 7-8, pages 1066–1072, April 2013
How to Cite
Wang, R.-H., Jian, S.-Y. and Yang, Y.-M. (2013), Psychometric testing of the Chinese version of the Contraceptive Behavior Scale: a preliminary study. Journal of Clinical Nursing, 22: 1066–1072. doi: 10.1111/j.1365-2702.2011.03785.x
- Issue published online: 11 MAR 2013
- Article first published online: 19 JUL 2011
- Accepted for publication: 23 March 2011
- Contraceptive Behavior Scale;
- female adolescent;
- psychometric testing;
- sexual health
Aims and objectives. To test the psychometric characteristics of a five-item Chinese Contraceptive Behavior Scale (CBS) among female adolescents in Taiwan.
Background. Improving contraceptive practice is important for reducing adolescent pregnancy. A scale for measuring contraceptive behaviour of sexually active female adolescents will help nurses promote contraceptive practices to sexually active female adolescents.
Design. A cross-sectional design.
Methods. Based on literature review and interview with female adolescents, a five-item CBS was developed. Using convenience sampling, 525 sexually active female adolescents were invited to complete the (anonymous) CBS. Exploratory factor analysis, confirmatory factor analysis (CFA), contrasting group validity and concurrent validity were tested to examine the validity of CBS. A multiple-group analysis was used to test the cross-validation of the structure of CBS in two groups. Cronbach’s α and test–retest reliability were used to examine the reliability of CBS.
Results. One factor was produced by exploratory factor analysis. Each item significantly loaded on the CBS by CFA. A multiple-group analysis indicated that the structure of CBS was reliable across two different samples. The overall chi-square and model fit indices were χ2 = 13·554, df = 8, p = 0·094, normed fit index = 0·992, comparative fit index = 0·994 and root mean square error of approximation = 0·036 (90% CI = 0·001, 0·069). The scores of CBS significantly correlated with subscales of perceptions of benefits to contraception (r = 0·47, p < 0·001) and subscales of perceptions of barriers to contraception (r = −0·29, p < 0·001). CBS mean scores significantly differed between the participants who used and did not use contraceptives at first sexual intercourse (t = 10·00, p < 0·001). Cronbach’s α for the CBS was 0·81. The intraclass correlation coefficient was 0·94.
Conclusions. The CBS had satisfactory validity and reliability.
Relevance to clinical practice. Nurses can use CBS to initiate discussions with sexually active female adolescents about their contraceptive practices and provide them with useful strategies to improve their contraceptive practices.