This study used a non-equivalent control group design with pre-post non-synchronized intervention.
Participants and setting
This study targeted junior nursing students from baccalaureate programs at a university located in Seoul, Korea. The participants were chosen by convenience sampling. A total of 92 students agreed to participate, and 91 students were included in the final analysis (one questionnaire had missing items). Data from 47 subjects in the experimental group and 44 subjects in the control group were used. The sample size was calculated using the G*Power 3.1.9 program. The effect size was set to 0.58 based on the results of a previous study that verified the effects of a self-directed learning program for Korean nursing students on self-confidence and satisfaction . For two tail independent t-test between the two groups, the minimum sample sizes were each calculated as 48 for the experimental group and the control group for an effect size of 0.58, significance level of .05, power of .80, and allocation ratio N2/N of 1. The number of samples in this study was fairly small.
We developed the SDL_BC based on Garrison’s self-directed learning model [12, 13], which consists of three dimensions: motivation, self-management, and self-monitoring (Table 1). We designed the program to include motivation by pre-clinical activities, development of self-management skills during practice, and conduction of self-monitoring activities after practice. Each stage consisted of both online and offline activities. Coaching strategies were based on the strategy model of self-regulated learning by Zimmerman & Martinez-Pones . Coaching included not only direct instruction but also feedback about discussions and assignments, review of daily learning objectives and content, and encouragement of activities. All coaching processes prioritized support and interaction with students. The detailed components of the SDL_BC are presented in Table 1. The intervention was implemented by a professor, and students were organized into teams of five to six. During the 2-week intervention, participants performed day and evening shifts.
The included variables of the students were age, self-evaluation of previous semester academic achievement, and satisfaction in nursing.
Self-directed learning competency
We used the 45-item Self-Directed Learning Competency Questionnaire, which was developed for undergraduates/adults by Lee et al.  and includes three subscales: planning (20 items), implementation (15 items), and evaluation (15 items). Items are rated on a 5-point Likert scale (1 = almost never to 5 = almost always). The total score ranges from 45 to 225, with higher scores indicating higher levels of self-directed learning competency. The internal reliability coefficient in the Lee et al.  study was 0.93. In this study, the reliability coefficients were 0.89 (pretest) and 0.90 (posttest), and the subscale internal reliability ranged from 0.72 to 0.86.
Perceived clinical competency
Perceived clinical competency was measured with a questionnaire developed for nursing students by Lee et al. . This is a 45-item, self-report instrument that assesses five subscales: nursing process (11 items), nursing skill (11 items), degree of cooperation (8 items), personal relations/communication (6 items), and professional development (9 items). Items are rated on a 5-point Likert scale (1 = almost never to 5 = almost always). A higher score indicates a greater degree of perceived clinical competency. The internal reliability coefficient in the original study was 0.96. In this study, the reliability coefficients were 0.96 (pretest) and 0.95 (posttest).
Clinical practice satisfaction
A numeric rating score (NRS) was used to measure clinical practice satisfaction. Participants were asked to evaluate their degree of satisfaction with the whole clinical practice, homework, content, and self-reflection on a scale of 0 to 10. A higher score indicates a greater degree of clinical practice satisfaction.
We used a non-synchronized design to prevent diffusion of treatment. In the control group that was trained using conventional self-directed learning program based on offline teaching, participants were recruited from September 1 to October 25, 2015, which were the beginning and end dates of the clinical practice period. In the experimental group that was trained using SDL_BC, participants were recruited from August 29 to October 28, 2016. Students were unaware whether they belonged to the control or experimental group.
Data were gathered in two phases, approximately 2 weeks apart. Participants completed self-report questionnaires. Pre- and post-intervention data were collected before and after the pediatric nursing clinical practice, respectively, and included demographic characteristics, self-directed learning, and clinical competency. The assessment of clinical practice satisfaction was performed immediately after completing the clinical practice. The questionnaires were completed in 15–20 min in both phases.
Written informed consent was obtained from all participants before inclusion in the study, which was previously approved by the Sungshin Women’s University Institutional Review Board (SSWUIRB 2015–056, SSWUIRB 2016–022). As for the ethical aspects of the procedure, participants were informed of the purpose of the study and their right to refuse to participate at any time. Participants were informed that there would be no penalties for refusing to complete the survey. Researchers then distributed and collected the questionnaires onsite. Students used a self-chosen code to reconcile their pre- and post-intervention questionnaires, thus assuring data anonymity and confidentiality.
Data were analyzed using the Statistical Package for the Social Sciences (SPSS), version 20 (SPSS Inc., Chicago, IL, USA). A Shapiro-Wilk test was performed to verify the normal distribution of the study variables, and parametric tests were subsequently used. Descriptive statistics, such as mean, standard deviation, frequency, and percentage, were calculated to assess the general characteristics of participants. The pre-intervention test of homogeneity was analyzed using an independent t-test and Fisher’s exact test. The pre-intervention test of homogeneity for self-directed learning and clinical competency between the experimental group and the control group was analyzed using the independent t-test. ANCOVA analyses that were controlled for pretest scores as confounders were performed to examine the effect on self-directed learning and clinical competency. An independent t-test was also used to examine the effects on the clinical practice satisfaction score in the experimental and control groups . The significance level is based on P < .05.