Study design and approach
The study employed a controlled quasi-experimental (pre-post) design with quantitative research approach among 401 randomly selected undergraduate nursing students (Intervention: n = 134 students and the control: n = 267 students) between January and April 2019. The difference of sample between groups is based on the ratio of 1:3. The study believed that nursing students are exposed to different sources of knowledge including media, experts, and or outside school-educated peers. Thus, comparing one student exposed to the intervention with three students in the control group was adequate to discriminate the effect of an intervention over the control.
The study was conducted in higher training institutions in two regions including Dar es Salaam and Dodoma, Tanzania. The distance factor was used to select regions purposively owing to the availability of higher training institutions with nursing programs. One higher training institution was randomly selected by a simple random technique using the lottery method to get two institutions (One higher training institution per region). The institutions were randomly allocated to either be the intervention or control group by a statistician independent to this study. The nature of study locations and the procedures involved in selecting them would minimize selection bias, allocation bias, performance bias and information contamination among undergraduate nursing students.
Inclusion and exclusion criteria
The entrance of students in this study was set voluntarily after they were explained the aim, advantages, and disadvantages of participating in the study. The study included first year undergraduate nursing students enrolled in higher training institutions and who were available and accessible during the study, students who provided informed consent to participate in the study, those who lived in or off-campus, and nursing students who were not recruited in other projects. Students in all training institutions were in the same midterm. On the other hand, the study excluded nursing students with health problems. As shown in Fig. 1, sums of 693 nursing students were eligible to participate in the study. However, 401 (57.8%) consented to join the study whereas 292 (42.2%) were excluded due to different reasons including postgraduate nursing students 67 9.7% (n = 67), decline to participate 14.1% (n = 98) and recruited in other projects 12.4% (n = 86) and having health problems 5.9% (n = 41).
Sums of 134 nursing students were assigned to the intervention arm whereas the other 267 nursing students were assigned to the control group. The interventional group learned the conflict resolution lesson materials through facilitation in a problem-based pedagogy (PBP) while the control group learned the the standard conflict resolution lesson materials through a lecture-based pedagogy (LBP) pedagogy. They were then followed for end-line assessment after an intervention. There was no loss to follow-up students in the study.
The study was conducted through four phases including Phase one (for ethical approval and clearance, development of the conflict resolution lesson materials, consultations and meetings with stakeholders, recruiting and training research assistants, and piloting of the research instruments); Phase two (For identifying and sampling study locations, recruiting undergraduate nursing students, and baseline assessment); Phase three (for intervention implementation), and Phase four (for end-line assessment/follow-up). With the technical and professional support from the consulted experts, the principal investigator developed the conflict resolution lesson materials which were then implemented by research assistants’ by using facilitation in a problem-based pedagogy. The interventions were conducted during the early days of a semester as per the study schedule that was blended with the institution’s academic schedule.
Separate unoccupied classrooms in respective higher training institutions were used for the nursing students to fill the questionnaires and for the researcher and research assistant to deliver the conflict resolution lesson materials among nursing students. Undergraduate nursing students were given a brief descriptions of how to fill the questionnaires then the copies were distributed among them. The researcher and assistants were available throughout the process to supervise and address any queries from the nursing students before collecting all copies of the questionnaires and secure them. A schedule of the next meeting to start the intervention was shared with nursing students before moving from the room.
The research tool that was used for data collection consisted of two parts including the socio-demographic part and the part that assessed self-directed learning readiness among nursing students. Socio-demographic items included Gender, age, marital status, accommodation, interest, and reasons to join nursing programs, satisfaction in learning nursing programs, perceived benefits, and challenges in learning nursing programs. A study adopted a 40 items 5-points Likert scale Self-directed learning Readiness scale (SDLRS) for nursing education and used by several previous studies to measure self-directed learning among nursing students [22,23,24,25,26,27,28,29].
The tool had scales ranging between 1 = strongly disagree and 5 = strongly agree. It consisted of three domains including students’ self-management (13 items), in which the responses were then transformed statistically into “Yes” response for a weight of “1″ score and “No” for the weight of “0″ with sums of scores ranging from 0 to 13. Mean as a central tendency measurement was used to define the endpoint of analysis of this domain whereas scores above mean were considered good self-management in the learning process. Moreover, responses of the desire for learning (12 items), and self-control domains (15 items) were treated the same as the self-management domain. Sums of scores ranging from 0 to 12 defined the desire to learn while a range of 0 to 15 scores defined self-control during the learning process. Scores above means were considered positive for the desire to learn, and good for self-control during the learning process respectively, otherwise not.
The validity and reliability tests were performed first before subjecting the tool to field data collection. It was shared with the subject matter, and statisticians for appropriateness of the items, consistency, and clarity to fit the literacy level of undergraduate nursing students. The comments were addressed accordingly and a tool was re-shared among the consulted research team for their approval to get the final version. The amended tool was then subjected to a pilot study among 40 nursing students from the training institution other than the sampled ones for testing its reliability. The findings of scale analysis from the pilot study revealed a Cronbach’s Alpha of 0.807, which was acceptable as recommended by previous studies [28, 30, 31].
The first part of the tool assessed students’ socio-demographic characteristics profiles including age, sex, marital status, year of study, accommodation status, interest to join nursing, satisfaction in the nursing program, and a reason to join nursing programs just to mention a few. For analytical purposes, the Likert responses were transformed into dichotomized (“Yes-No”) responses of which “Yes” was associated with an action or behaviors to have been demonstrated by the student and it was assigned a weight of “1″ score otherwise not (“No” response = “0″ score). A new dichotomized variable of self-directed learning was computed whereas a score of > 20 of the total item score was considered to be the acceptable level of SDL readiness, otherwise not.
The study adopted the problem-based pedagogy framework by Millanzi et al., [9, 21]. The framework includes some stages to implement problem-based pedagogy in facilitating learning among nursing students. They include self-introduction and group formulation, presentation of paper-based scenario, time for resolving the presented problems, debates and sharing of solutions in the class, guidance and support from facilitators, peer and facilitator assessment, and summary of the study activities. Students learned two units including the unit about the concept of conflict resolutions alongside the strategies to address it. The second unit was all about demonstrations of conflict resolutions in nursing.
Introduction and group formulation
The first day of the intervention was used by the facilitators to introduce the intervention by providing brief descriptions about the duration of the intervention, learning objectives, and in-out-class learning activities. The groups of 5 to 8 members were randomly formed under the supervision of the researcher and assistants and group leaders were identified including chairpersons and secretaries while other members participated in addressing the assigned activities accordingly.
Paper-based scenario presentation, resolving, debates, and sharing (30 min)
The researcher and assistant distributed pieces of papers with a conflicts-based scenario in each group followed by brief descriptions on the ways students were required to address them individually and in their groups. The individual and group learning activities were identified before starting the session and queries from students were addressed by the researcher and assistants accordingly. Learning activities that were assigned to students included the identification of the professional context of the presented scenario, identification of known issues and new ones, issues to know more, and identification of issues to learn from the scenario.
Over a period of 60 to 120 min, nursing students had to clarify, rank, and assign learning tasks among themselves to find appropriate strategies to resolve the presented scenario in their groups. Given the facilitator’s guidance and facilitation, nursing students were supposed to identify and suggest resources necessary in resolving the assigned scenario. After a brief description and assignment of the scenario among groups, students were given one to address the scenario by exploring possible conflict resolution strategies that suit the given problem before being shared and discussed in the entire class during the next scheduled session.
The researcher and assistant were available to receive any calls, queries, and concerns from students either by mobile texts, orally, or in writings through email. This was meant to help them if they would need any help or clarifications about the assigned task. In the next session (after a week) students presented, discussed, and shared what they explored in the entire class. The facilitator’s guided min-debates were allowed whereas groups had to defend their works when other groups challenged and critiqued them. Students were also offered opportunities to share real-world events and other related phenomena related to the topic under study.
Support from facilitators and facilitation of groups
Facilitators organized and managed the learning environment by ensuring that there was an adequate supply of learning materials, opening windows to allow fresh air to enter the room, enough light and enough space that facilitated students’ discussions, free movements, and that which minimized disturbances among students. Given the prescribed roles by the researcher and assistant, group leaders were sometimes used to lead, monitor, control, and evaluate group members.
Peer assessment among students was used to determine students’ interactions and abilities to evaluate one another throughout sessions. The student’s experience inventory developed by the researcher was used to evaluate the implementation of conflict resolution material that was developed in the nature of facilitation in a problem-based pedagogy. Using a random selection method, a researcher and assistant selected 2 to 5 students randomly after each session to share their experiences about the sessions. Moreover, the written posttest was used as a summative assessment to measure self-directed learning among nursing students.
A Statistical Package for Social Sciences (SPSS) software program version 23. Error checking (data screening) test was performed to test for normality before data analysis to identify the type of analysis model to be used. Findings from the normality test indicated data were approximately normally distributed and therefore parametric measurements were opted. All missing data were excluded from the analysis models. Descriptive analysis through chi-square and crosstabulation was performed to establish nursing students’ socio-demographic characteristics profiles and determine the relationship between categorical variables. Independent two-sample t-tests were performed to compare the mean score to establish mean score change and differences among nursing students between groups. Binary and multinomial logistic regression analyses controlled with other co-related factors were performed to determine the association between an intervention and self-directed learning readiness among nursing students. The confidence interval was set at 95%CI with the power of the study set at 80%. A significance level was set at 5% (0.05) of which < 0.05 of the association between variables was considered statistically significant.