When nursing students transition to practice as newly licensed nurses, they face many challenges, including feelings of being unprepared (Draper et al., 2014). Clinical and simulation-based learning experiences have been shown to increase self-efficacy and various clinical competencies, preparing nursing students for nursing practice (Khalaila, 2014; Kim, 2018). Self-efficacy is one's judgment of personal capabilities to execute the actions required to attain a goal (Bandura, 1997). As self-efficacy is essential in facilitating the development of clinical competencies in nursing students, it is essential to be able to measure self-efficacy in clinical and simulation.
BACKGROUND
The Clinical and Simulation General Self-Efficacy Scale (CSGSES) was developed by Dykes (2011) to evaluate nursing students' perceived self-efficacy in clinical and simulation settings. The CSGSES was adapted from the General Self-Efficacy Scale (GSES), developed by Schwarzer and Jerusalem (1995) to measure perceived self-efficacy and predict coping and adaptation associated with daily hassles and stressful events. Dykes (2011) modified items on the GSES to coincide with clinical objectives and used it to measure nursing student performance in a medical-surgical nursing course. The CSGSES is an 18-item scale rated on a 4-point Likert scale (1 = not at all confident, 2 = slightly confident, 3 = moderately confident, 4 = highly confident). Scores are summed and range from 18 to 72, with higher scores indicating greater self-efficacy. Cronbach's alpha was previously reported as .86 to .948 (Burbach et al., 2019; Dykes, 2011).
Further psychometric evaluation of the CSGSES has not been completed. To measure self-efficacy related to clinical and simulation accurately, the instrument should be further evaluated for validity and reliability. This study addresses that gap, which can have implications in nursing education. Therefore, the aim of this study was to psychometrically evaluate the CSGSES.
METHOD
This study employed the use of psychometric testing to evaluate the reliability and validity of the CSGSES in a convenience sample of senior-level prelicensure bachelor of science in nursing (BSN) students in one public university in the Midwestern United States. For an 18-item scale, a sample of 180 was sought (Waltz et al., 2017). Data were collected via an online questionnaire. Human subject's approval was obtained.
The CSGSES was completed by four Certified Healthcare Simulation Educators to examine face and content validity, meeting the minimum recommendations by Waltz et al. (2017). The experts rated each item on a 4-point rating scale for relevance and clarity (1 = not relevant/clear to 4 = very relevant/clear) and provided rationales for a choice of 3 or less. Wording revisions were made based on their feedback. Concurrent validity was tested using the GSES as a criterion, with the hypothesis that the instruments would positively correlate. The GSES was administered to participants at the same time as the CSGSES. Participants then completed the CSGSES two weeks later to examine test-retest stability reliability.
Data analysis was conducted using SPSS 26 (SPSS, Inc., Chicago, IL). Item characteristics were analyzed using means and standard deviations. For content validity, an individual content validity index (I-CVI) was calculated for each item, as well as a scale-level content validity index (S-CVI). Items with an I-CVI score of .75 or greater were retained (Waltz et al., 2017). Construct validity was examined with exploratory factor analysis using principal axis factoring, Promax rotation, and methods per Osborne (2014) were used as the underlying structure of the CSGSES was unknown (Waltz et al., 2017). Factors with eigenvalues of >1 were retained. Items loading <.3 and with double loadings were removed. Pearson product-moment correlation was used to examine the concurrent validity of the CSGSES with the GSES. Internal consistency reliability was examined using Cronbach's alpha, and test-retest correlation was examined using Pearson's correlation coefficient.
RESULTS
One hundred twenty-five participants completed the CSGSES and GSES; 115 participants completed the CSGSES the second time. The majority of the sample (91.2 percent) was female and Caucasian (86.4 percent). Ages ranged from 20 to 32 years, with an average age of 23 years (SD = 3.6). Total scores on the CSGSES ranged from 50 to 76, with an average of 65.6 (SD = 6.2). The S-CVI was .94 for relevance and .92 for clarity. The I-CVI for all items was .75 or greater. Seven items were modified for clarity, and Item 1 was broken into two separate items based on experts' suggestions.
Exploratory factor analysis using principal axis factoring and Promax rotation was used as multivariate normality assumptions were violated in the data, and it was assumed that the scale items would be correlated with self-efficacy (Osborne, 2014). Although the a priori sample size of 180 was not met, Bartlett's test of sphericity (p < .001) and Kaiser-Meyer-Olkin measure of sampling adequacy of .879 indicated the sample size was large enough to perform a factor analysis, and a significant factor analysis model was present. All items loaded with a factor of >.3. The factor analysis revealed three factors with eigenvalues of >1, which was consistent with the scree plot. Items 5, 6, and 15 were removed because they cross-loaded onto two factors with a difference of <.2 (Osborne, 2014). With 16 items, this three-factor model accounted for 49.65 percent of the variance. The three factors were named nursing process (six items), safety (six items), and performance (four items).
Concurrent validity of the CSGSES with the GSES using Pearson correlation coefficient showed a significant medium association (r = .421, p < .001, 95% CI [.261, .586]). Internal consistency of the revised CSGSES with 16 items was completed. Cronbach's alpha of the instrument overall was .875. Cronbach's alphas were .816 for the nursing process subscale, .814 for the safety subscale, and .814 for the performance subscale. For test-retest stability reliability, Pearson's correlation coefficient was r = .668 (p < .001, 95% CI [.435, .806]).
DISCUSSION
The aim of this study was to psychometrically evaluate the CSGSES to provide evidence of its reliability and validity in a sample of senior prelicensure BSN students. This study adds to the limited evidence available on the psychometric properties of the CSGSES. Evidence of content validity was based on review by nursing education experts. The I-CVI for each item and the S-CVI indicated a high degree of agreement among experts on relevancy and clarity of items, meeting the recommend I-CVI of .75 or greater and S-CVI of .9 or greater (Polit & Beck, 2017; Waltz et al., 2017).
The exploratory factor analysis produced a three-factor structure that included the subscales of nursing process, safety, and performance. These factors are consistent with the initial clinical objectives from which the CSGSES was developed (Dykes, 2011). However, it is important to note that the a priori sample size of 180 was not met, indicating the need for further research and confirmatory factor analysis to confirm these findings with the revised 16-item scale.
Concurrent validity of the CSGSES indicated a significant medium positive correlation with the GSES as expected. Results from internal consistency in this study of the overall scale are consistent with previous research reporting Cronbach's alpha (Burbach et al., 2019; Dykes, 2011). This study adds evidence of acceptable internal consistency for the subscales of nursing process, safety, and performance as Cronbach's alpha of .7 or greater is considered adequate (Houser, 2008). The results of the test-retest stability reliability of the CSGSES supported the instrument's reliability as a correlation of .5 or greater is desired (Houser, 2008).
This study's limitations include the largely homogenous sample, and a convenience sample of senior nursing students in one Midwestern public university was used, limiting the generalizability of the study. Furthermore, it should be noted that student self-evaluation does not always equal actual abilities; however, this relationship was out of the scope of the current study. The revised CSGSES would benefit from additional testing with nursing students in both the clinical and simulation settings and validation by nurse educators in the clinical setting as this study used simulation experts to examine content validity. Results of this study indicate the need for further research with a more robust and diverse sample.
CONCLUSION
The revised CSGSES is a 16-item instrument to evaluate nursing student self-efficacy in relation to clinical and simulation. In this sample, the CSGSES showed evidence of face validity, content validity, construct validity, concurrent validity, internal consistency reliability, and test-retest stability reliability. The CSGSES shows promise for use in simulation and clinical as a form of self-evaluation and check of expected nursing competencies for nursing students. This allows nursing students to self-identify development and improvement areas before completing various clinical or simulation experiences and, ultimately, before graduating and entering the workforce. Nurse educators can use the revised CSGSES with students to expand their understanding of clinical and simulation-associated self-efficacy and provide educational experiences to meet nursing students' needs.
REFERENCES