Design
A nonrandomized pretest–posttest quasiexperimental research design was used, and students were nonrandomly assigned to an experimental group (Moodle) and a nonexperimental group (non-Moodle) based on free will. The Moodle group was exposed to learning methods involving a Moodle e-learning environment, textbooks, an iPad, and the Internet, among other resources. Nursing students in the Moodle group could choose any learning method at any time, but use of the Moodle e-learning environment was required. Those in the non-Moodle group could use textbooks, an iPad, and the Internet, among other resources, without using the Moodle e-learning environment to complete the course. Once the study was completed, the students in the non-Moodle group were free to use the Moodle e-learning environment. The study framework used in this study is shown in Fig. 1.
Participants
Heppner and Heppner (2004) suggested that college student recruitment should be approved by the director, dean, and experts of the institution and that participation should be voluntary. G*Power software (latest ver. 3.1.9.7) with one-way ANOVA (F = 0.5; α = 0.05; groups = 2) indicated that the sample size should be at least 54. In this study, 105 third-year nursing students enrolled in a pediatric nursing course in a medical college in northern Taiwan were recruited. Eighty-nine nursing students were nonrandomly divided into two groups (Fig. 2). The Moodle group used Moodle e-learning environment technology, while the non-Moodle group completed the course in the traditional classroom environment and did not use Moodle e-learning environment technology until the completion of the study. Students completed the course following their own study plan based on their time schedule. Participating students chose to join the Moodle group or the non-Moodle group on their own. Because this course was part of nursing students’ compulsory curriculum, random grouping and blind testing were impossible, which was the reason for the unequal sample groups. Hence, only volunteers could join the Moodle group and complete the course through the e-collaborative learning environment created by Moodle e-learning technology. To control the differences, the two groups were matched based on major (nursing students), grade (third-year students), and learning subject (pediatric nursing course), and all students took a pretest.
Data collection
Instruments
Data were collected using an electronic Webropol® questionnaire (Fig. 3), which could be accessed through a link in Moodle. There were 25 questions in the questionnaire, among which four gathered basic information and 21 were related to two survey scales. The following basic information was collected: age, gender, educational background, and field of study. The student satisfaction scale (eight questions in total) measured student satisfaction with the course and its impact on learning. A collaborative learning scale (13 questions in total) was used to measure students’ e-collaborative learning in three aspects, i.e., promoting learning, the role of the teacher, and the role of the student. A five-point Likert scale (5 = completely agree, 4 = partially agree, 3 = neither agree nor disagree, 2 = partially disagree, and 1 = completely disagree) was used to measure student perception.
The student satisfaction scale included two subdimensions: (1) learning satisfaction (five questions) and (2) e-learning environment satisfaction (three questions). Questions related to learning satisfaction in an e-learning environment were developed and used based on previous literature (Virtanen et al., 2017). The content validity of all questions was assessed by the same expert teacher group. All questions were pretested by eight nursing students to ensure that the questions could be properly understood and interpreted. After the content validity assessment and pretest, no changes were made to the questions (Ritter and Sue, 2007). Because the student satisfaction scale combined questions for two different dimensions, the construct validity was verified (Table 1) (Männistö et al., 2019).
Exploratory factor analysis (EFA) indicated a two-factor model for the scale. The eigenvalue of the first factor (learning satisfaction) was 4.38, and the total explained variance in questions was 54.8%. The eigenvalue of the second factor (e-learning environment satisfaction) was 1.06, and the total explained variance in questions was 13.3%. These results indicated that this scale has good construct validity. Cronbach’s alpha for the first and second factors was 0.84 and 0.83, respectively. The internal reliability of the instrument was evaluated using Cronbach’s alpha for both factors separately and for the whole instrument. The alpha for the whole instrument was 0.944. This result indicated that this scale has good internal consistency.
In this study, the collaborative learning scale developed by Vuopala (2016) was used to evaluate collaborative learning in an e-learning environment. This scale includes three subdimensions: (1) promoting collaborative group work (six questions); (2) the teacher’s role in the collaborative learning environment (four questions); and (3) students’ role in collaborative learning (three questions). The content validity of all questions was assessed by the same expert teacher group. All questions were pretested by eight nursing students to ensure proper understanding and interpretation. After the content validity assessment and pretest, no changes were made to the questions (Ritter and Sue, 2007). EFA revealed a three-factor model. The first factor (promoting group collaboration) had an eigenvalue of 5.58, and the total explained variance in questions was 43.0%. The second factor (teacher’s role in the collaborative learning environment) had an eigenvalue of 2.70, and the total explained variance in questions was 20.8%. The third factor (the role of students in the collaborative learning environment) had an eigenvalue of 1.03, and the total explained variance in questions was 7.9%. Therefore, this scale has good construct validity. In addition, Cronbach’s alpha for the first, second, and third factors was 0.92, 0.83, and 0.76, respectively (Männistö et al., 2019). These results indicated that the scale has good internal consistency (Table 2).
Finally, students’ study achievements were assessed. At the beginning of the course, the assessment criteria for study achievements were explained to students in both groups. At the end of the 18-week course, students in both groups were assessed on the knowledge, content, methods, challenges, and learning performance in the pediatric nursing course. Study achievement was divided into six levels from 0 (failed) to 5 (excellent). Students were asked to answer the questions in a question-and-answer format. The Moodle group answered the questions online, while the non-Moodle group answered the questions in the classroom.
In this study, collaborative learning took place in a Moodle e-learning environment. To explain how students utilized the collaborative learning modules, the figures below show the e-learning environment (Moodle) interfaces. The students could answer or ask questions related to the class, and data related to testing in the collaborative learning environment were collected for analysis.
Participants had access to the e-learning environment (Moodle) in the time period after the pretest and before the posttest. The operation interface was preset as a platform with different operation menus (e.g., main menu and discussion topics). After choosing the course from “My courses,” students could enter the virtual classroom (Figs. 4 and 5).
Environment setup
This study was conducted with a teaching environment setup that was the same as that for a general college class. The students met in the classroom on the first day of the study and received course information before beginning the study. Subsequently, all the students took the pretest in the classroom. The posttest was also held in the same classroom. Students were familiar with the environment because they often attended classes in the classroom. After the pretest, students were nonrandomly assigned to the Moodle group and the non-Moodle group based on their own choice. Students in both groups received guidance in the classroom on how to use the Moodle e-learning environment for this study.
The differences between the educational intervention measure, i.e., an e-learning environment, and traditional classroom-based face-to-face teaching were compared for the pediatric nursing course. The course content and study objectives were the same for the Moodle and non-Moodle groups, and students in each group took the course during the same 18-week period. The objective of the course was to learn multidisciplinary concepts related to all aspects of pediatric nursing. Both groups were taught by the same teachers and used the same textbooks. Educational interventions were developed based on a systematic literature review (Männistö et al., 2020) and evidence from the learning theory of social constructivism (Chan and van Aalst, 2018). Study achievements were assessed via written examinations before and at the end of the course. Students in both groups were given a pretest and a posttest. The college lecturer and his teaching assistants supervised the exams and collected and graded the pretest and posttest exams.
Experimental procedures
This study was conducted with students enrolled in a pediatric nursing course in a medical college. The person responsible for the research plan gave students a PowerPoint course syllabus and explained to the students the goals of the e-learning environment (Moodle) and e-collaborative learning in the pediatric nursing course. After answering any related questions from students, the researchers recruited student volunteers to participate in the study, and the students who volunteered to participate were required to sign a consent form. Subsequently, the participating students took a pretest. They were given a paper exam and sufficient time to complete the questions.
At the beginning of the first class, the course objectives, content, and teaching methods, along with group learning activities and tasks, were introduced to the students. The Moodle group was divided into five small groups, the members of which remained unchanged throughout the online course. Students were allowed to form groups on their own. After the first class, students opened their Moodle account and completed their weekly assignments, online discussions, comments, self and peer reviews, material searches, reports, and various exams with others in the same group on the Moodle platform. One week before each class, five case scenarios under the study subject on the course schedule were assigned to each group. Each group used the Moodle e-learning environment to study the case scenarios assigned. Through online discussion, comments, self and peer reviews, and material searches, each group completed a report after e-collaborative learning and uploaded and published it in the Moodle discussion room before the deadline for sharing and discussion. In addition, each group was asked to provide comments on and suggestions for other groups’ reports. Taking the study subject of asthma as an example, the teacher assigned five asthma case scenarios to the five groups and posted the assignment on Moodle one week before the class. All groups were required to publish their report, which was created after e-collaborative learning, in the Moodle discussion room two days before the class (Supplemental data 1).
In the course, students constructed new knowledge via research, sharing with others, and studying the given subjects. By studying in groups, students benefited from continuous social interactions, facilitating the development of problem-solving and decision-making skills. Through sharing coconstructed knowledge and continuous social interactions, students were guided to learn together. The students in the non-Moodle group received traditional face-to-face instruction that consisted of 33 hours of subject-centered teaching. Face-to-face teaching occurred in the form of lectures (PowerPoint). The Moodle group was also given additional teaching materials, while the non-Moodle group used the Moodle platform only for course material storage rather than for e-collaborative learning. During the course, students were asked to conduct group discussions related to different subjects presented by the teacher without recording student responses and how they responded. The primary purpose of the discussions was to prompt students to think independently and broaden their understanding of the subject. All students in the non-Moodle and Moodle groups had 18 weeks to learn and complete the pediatric nursing course before the posttest. The posttest was given in the same manner as the pretest after 18 weeks of study, and all test results were e-mailed to the researchers by the teacher (Supplemental data 2).
Ethical considerations
The reliability and validity of the research were considered for every step in this study. This study was approved by the Institutional Review Board (IRB) of XX Hospital. Consent forms were voluntarily signed by all participants. All data were stored in a laptop protected by a password set by the investigator. Printed copies of the data collected were stored in a locked file cabinet that could be opened only by the investigator with a key.
Data analysis
Inferential statistical analysis of quantitative data collected from the pretest and posttest was performed using the social science statistical software SPSS. First, the data were converted to frequencies, percentages, means, and standard deviations. Second, paired t tests were performed to compare the mean study achievement scores for each group for the pretest and posttest and to test whether there was a significant difference in the pretest score between the Moodle and non-Moodle groups. Finally, one-way analysis of variance (ANOVA) was used to test the study hypothesis and compare the mean study achievement scores on the posttest after adjustment between the two groups. The following question was answered: Did the posttest score for the Moodle group increase more than that for the non-Moodle group? To identify any skewed datasets, the data were examined for the presence of interactive relationships between variables (i.e., internal consistency). Statistical analysis results may be biased when greater than 10% of data are missing (Bennett, 2001). Therefore, questionnaires with missing data that exceeded 10% of the total were discarded. Multiple imputation methods were used to handle missing data.
Validity and reliability/rigor
All teaching methods followed the basic pedagogical principle of collaboration and were continuously supported by teachers (Miyake and Kirschner, 2014). The collaborative tools used in the course were multimedia, videos, pictures, textbooks, writing materials, and quizzes related to theoretical knowledge (Kollar et al., 2018).