To determine the reliability of the study questionnaire, it was completed by all the participants. Next, the Cronbach ’ s alpha coefficient was computed for the BCKT instrument, the CRHBMS questionnaire, and its subscales. The alpha value for the BCKT and CRHBMS questionnaires were 0.75.2 and 0.82.2, respectively. In addition, the alpha value of the CRHBMS subscales for the current study ranged from 0.75 to 90 (Table 1). Therefore, the reliability of the study instruments was also confirmed. Statistical analysis To analyze the study data, the researchers applied de- scriptive and inferential statistical methods. A chi-square test was conducted to assess the relationship between two categorical variables. In addition, a multivariate logistic regression was applied to determine the factors affecting behavior using SPSS software version 16.0 (SPSS Inc., Chicago, IL, USA). In this study, a P value less than 0.05 was considered significant in all the analyses. Results Demographics and BSE behavior In total, 366 eligible subjects were recruited into the study and were asked to fill out the questionnaire. Thirty-two questionnaires were excluded due to either the participant not following the instructions provided by the researchers (20 cases) or incompleteness (12 cases). Hence, statistical analyses were performed on 334 questionnaires. The obtained results indicated that 82 out of 334 par- ticipants (24.6%) performed BSEs. To determine the probable predictors of BSE, demographic factors were Table 1 Subscale scores of CRHBMS a regarding breast cancer and breast self-examination Subscale No. of items Cronbach ’ s α Mean SD b Perceived susceptibility 5 0.749 2.64 0.61 Perceived severity 7 0.797 2.95 0.75 Perceived benefits 6 0.846 3.51 0.67 Perceived barriers 5 0.880 2.10 0.85 Health motivation 7 0.766 3.65 0.66 Perceived self-efficacy 11 0.897 2.85 0.75 a CRHBMS = Champion ’ s Revised Health Belief Model scale; b SD = Standard Deviation Didarloo et al. BMC Public Health (2017) 17:861 Page 3 of 8
compared between the groups that performed and did not perform BSE behavior. In the group with BSE behav- ior, 85.4% of the participants were undergraduate/gradu- ate students, and in the group without BSE behavior, 66.3% were undergraduate/graduate students. Academic level had a statistically significant association with BSE behavior ( P < 0.001, Table 2). Academic year showed a significant association with BSE behavior. We found that 20.7% of the participants in the group with BSE behavior and 10.3% in the group without BSE behavior were in the fourth year and higher ( P = 0.045, Table 2). Approximately 62.2% of the subjects in the group with BSE behavior and 59.5% of participants in the group without BSE behavior were of Turkish ethni- city ( P = 0.892, Table 2). A significant association was ob- served between the grade point average of the participants and their BSE behavior. In the group with BSE behavior, over 67.1% of the participants had good or excellent grade point averages, and in the group without BSE behavior, 49.2% of the participants had good or excellent grade point averages ( P = 0.028, Table 2).
- Fall '13