Flipped or traditional online teaching? Two different strategies to handle teaching in nursing education during the COVID-19 pandemic

Context

The module examined in this study took place during the second semester of a nursing program at undergraduate level taught at a mid-size university in Sweden during the COVID-19 pandemic 2020. The nursing program as a whole is mostly assessed on competence and practical skills. However, the students in this study were still completing their basic education in medical science and were assessed according to the course learning outcomes. They just completed the program’s first short clinical practice, however, most students had not yet reached the level where they were able to integrate medical science knowledge with nursing theory and clinical practice.

Study design

Approval to conduct the study was received from the Karlstad University’s Institutional Review Board (No: HNT 2019/801). Two different course iterations were included in the study, the spring and fall semesters of 2020, with 104 and 103 students, respectively. One teacher gave lectures in microbiology and, in parallel, another teacher gave lectures regarding the immune system.

The first strategy during the shift to online teaching during the spring of 2020 was to move traditional lectures onto Zoom for the whole course module. The module ended with a summative digital exam covering the immune system and microbiology, which was studied simultaneously. Students completing the whole module served as a control group in this study.

During the fall semester of 2020, the parts of the course regarding the immune system were changed to a flipped classroom approach. The activities regarding the immune system were divided into two major blocks: (1) immunology and (2) asthma and allergy. Before the module started, the students were informed of the upcoming events and what was expected of them during different learning activities. Before their seminars, students were supposed to watch instructional videos prepared by the teacher, however, there was no control that the students had done this pre-class activity. During the seminars, the students were randomly divided into groups and sent to Zoom breakout rooms to work with detailed learning outcomes and investigate case-based problems. The teacher was available during these group sessions. After the group seminars, there was a whole-course seminar on Zoom where the block was summarized and uncertainties were followed up, giving the whole group the same information. During these whole-course seminars, the teacher took a passive role, letting the students take an active teaching-learning role. The number of attending students at each activity was noted.

After the seminars, students had the opportunity to self-assess their learning process regarding the immune system using a summative online quiz, which could be taken as many times as desired. The activities regarding the immune system using the flipped classroom approach constituted the intervention in this study.

In parallel with the flipped classroom approach, students also attended traditional lectures in microbiology given through Zoom in the same manner as during the spring semester. The lectures in microbiology served as adjustment variable. The summative digital exams concluding this module were also given in the same manner as during the spring semester.

To clarify, there were two sets of control groups in this study. First, the entire spring iteration was when all teaching took place through traditional lectures using Zoom. Second, in the fall iteration, the microbiology content was provided through lectures using Zoom. The students in the fall iteration were also exposed to the intervention through the flipped classroom approach regarding the immune system. These students thus constituted their own controls, but they were also compared to the spring iteration which were only subjected to the control procedure entirely through traditional lectures. The course structure for both iterations is in Figure 1.

Figure 1:

Illustration of the course structure for both iterations.

Data collection and questionnaire

An online questionnaire was distributed, inviting the students during the spring and fall semesters of the second year of the nursing program. There were 104 and 103 students respectively with a mean age of 26.9 years. Data were collected between December 3, 2020 and January 13, 2021. In total, 59 students participated. The questionnaire included 10 questions about reasons for not attending seminars, the level of pre-seminar preparation, the preferred form of teaching, and estimates of how the form of teaching had affected their study results. Most students (85%) were women. The questionnaire also included five questions where students could choose from a list of options as well as write their experiences about the flipped classroom in general and the courses specifically. Questions followed a similar format.

Dependent variable

Independent variables

Adjustment variables

Analysis

Logistic regression analysis was used to analyze the correlation between the exam grades in the fall iteration with the number of seminars attended in immunology, asthma, and allergy, i.e., correlation between activity in the flipped classroom and the outcome. In order for the students to benefit from the complete flipped classroom approach, participation in the seminars was an important parameter to investigate. To compensate for the fact that different students can be at different levels before the intervention the model was adjusted for the previous grades in pathophysiology. Logistic regression was also used to analyze the correlation between pathophysiology and the immune system, as well as the respondents’ estimate of the flipped classroom’s impact on their exam result and amount of time invested. The exam scores between these groups were compared using the 2-sided Student’s t-test, and the Mann–Whitney U test was used to compare grades from the previous semester between the courses before the intervention.

The quizzes constituted the main post-class activity in this flipped classroom approach. In order to provide an idea of the effect of the quizzes on academic achievement, the average exam scores of students who made no quiz attempts were compared with those of students who made at least one attempt by using the 2-sided Student’s t-test. The major post-class activity consisted of quizzes, and logistic regression was used to analyze the correlation between exam grades and the number of quiz attempts, both with and without adjusting for previous grades. Logistic regression was also used to analyze the correlation between the number of quiz attempts and attendance at the seminars.

The limit of statistical significance was set at p≤0.05 for all correlations and tests of significance. SPSS (Statistical Package for the Social Sciences) version 26 was used for statistical analyses.

The qualitative data included comments from the open-ended questions on the questionnaire and were analyzed together, using an inductive thematic analysis approach, and the six steps suggested by Braun and Clarke (2006). Thematic analysis is a qualitative method of analysis with the aim of finding common themes among the respondents’ answers. The results were analyzed on a semantic level. In a first step, data were read and re-read, making notes by hand. In a second step, all open-ended questions were coded by the second author. In a third step, themes were conceptualized based on the data and on the research question. A theme, according to Braun and Clarke (2006) seizes something important in the data, following the research question, representing some kind of meaning within the data. Fourth, the themes and codes were reviewed and discussed with the first author by the researchers to reach consensus. In a fifth step, themes were named, and these directed the overall story of the analysis.

Seminar participation

At the four seminars during the fall iteration of the course on immunology, asthma, and allergy, the whole-course seminars were more popular than the group seminars, as seen in Table 1. Attendance also decreased overall from the first occasion regarding immunology to the second occasion regarding asthma and allergy.

The students were divided into three groups, with 13 (12.6%) students having participated in three or four seminars, 27 (26.2%) in one or two seminars, and 63 (61.2%) not participating in any. During the three lectures on microbiology, there were approximately 80, 40 and 20 students attending, respectively.

Pre-intervention comparison

The academic achievements in medical science between courses were compared. The Mann–Whitney U test showed a statistically significant difference between the courses, with the spring course (control group) having an average throughput of 61.7%, and the fall course (intervention group) having 59.4%.

Overall exam scores

When the exam scores were compared between the two course iterations using Student’s t-test, the spring iteration scored significantly higher than the fall iteration, 71.3 and 64.9% respectively (p=0.007) (see Table 2). When the results were divided into subjects, no statistically significant difference was seen between the iterations for the immune system (p=0.3). Microbiology, on the other hand, resulted in a significantly higher exam score for the spring iteration (p=0.001). All scores were normalized to the control (Microbiology) which was set to 100 for easier comparisons.

Association between seminar participation and exam score

Higher seminar participation resulted in higher average exam scores, 71.5 and 62.3% respectively, although not statistically significant using Student’s t-test (p=0.08) for the highest and lowest participation groups. This pattern is also seen when the subjects were reported separately. The immune system seminars consistently resulted in higher exam scores than the microbiology lectures, shown in Figure 2. The more seminars the students attended, the greater the difference in exam scores between the subjects, 81.8 and 62.4% (p<0.000) for the immune system and microbiology respectively for the highest participation groups, and 68.8 and 56.7% (p<0.000) respectively for lowest participation.

Figure 2:

Exam score related to the number of seminars attended. Total score (■), immunology (●) and microbiology (▲). ***p≤0.001 between the immune system and microbiology.

Effect of seminar participation on exam grade

When analyzing the correlation between exam grades and participation in the seminars using logistic regression, the total grade pass with distinction for the exam in the fall iteration was explained by participation in the seminars, shown in Table 3. However, the effect disappeared when the previous course grade in pathophysiology was taken into account. On the other hand, the grade pass was explained by participation in the seminars (OR 2.3 (CI [1.001–5.1])), i.e. 2.3 more likely to pass than fail if you participate) when adjusting for the previous course grade. The previous course grade in pathophysiology had the greatest effect on the grade for both pass and pass with distinction (OR 4.2 (CI [1.77–9.95])) and 20.0 (CI [3.53–112.91])), respectively). As seen in Table 3, the adjusted OR is bigger than the crude OR, but this was probably the result of random variations and small samples.

Effects of quiz attempts on exam score and grade

The voluntary quiz regarding the immune system was used by 42 students. Of these, 17 took it several times in quick succession just before the exam and 14 took it only once. The average number of attempts was 2.3. When comparing exam scores between the students who did not use the quiz with those who made at least one attempt using Student’s t-test, the scores were 60.8 and 71.2% respectively (p<0.000).

When comparing the grades on the exam with the number of quiz attempts, the crude OR was 1.9 (CI [1.06–3.43]) and 3.6 (CI [1.67–7.67]) for pass and pass with distinction, and 1.9 (CI [1.03–3.42]) and 3.8 (CI [1.64–8.85]) when adjusted for the grade attained in an earlier course. When also adjusting for participation in the seminars, the OR was 1.7 (CI [0.94–3.139]) and 3.5 (CI [1.49–8.11]) for pass and pass with distinction, respectively. No correlation was seen between the number of quiz attempts and the number of seminars attended.

Descriptive results from the questionnaire

The respondents preferred to watch video lectures, followed by teacher-led seminars (48%) or group seminars (39%), in contrast to classic lectures online, which only eight (14%) respondents preferred. As seen in Table 4, a majority of the respondents who attended at least one seminar (78%) stated that the seminars were rewarding. Furthermore, 73% of the respondents stated that they felt relatively or very confident before the exam regarding both the immune system and asthma and allergy, whereas 51% felt relatively or very confident regarding microbiology.

The perception of the respondents was that they succeeded better in the exam due to the set-up with films and seminars, but also that they needed to invest more time studying in this scenario compared to classic lectures, 37 (80%) and 30 (65%) students, respectively. No correlation was seen between the respondents’ estimate of the flipped classroom’s impact on their exam result and invested time using logistic regression.

As seen in Table 5, most respondents stated that they were prepared or very prepared before the seminars. The major reason for not attending the seminars was that they thought the recorded lectures were enough. Of the respondents, 29 (49%) and 30 (51%) of them had watched all the video lectures at least once before the seminars in immunology and asthma and allergy, respectively.

Pros and cons with recorded films

It emerged that the recorded films in the flipped classroom approach was mostly appreciated by students and viewed as a pedagogical way of enhancing learning: “I think this approach, when it comes to distance education, is absolutely superb, recorded lectures are something I learn a lot from.” Another student wrote: “During live lectures on Zoom, I usually don’t have time to listen and take notes and ask questions.” However, the negative aspects of the recorded film in the flipped classroom approach were the lack of opportunities to ask the teacher questions directly, to get concepts explained, and just “talk to the teacher”. Additionally, the lack of social interaction was mentioned. It appeared that life situations, such as having small children, also affected students’ perspectives on the flipped classroom approach. Having the opportunity to view lectures later, pause them, and view them more than once was especially liked by students who were parents.

Group seminars – most benefit for the weak?

The students’ opinions of group seminars were partly dependent on previous experience. Group work in general can be difficult for teachers to administer because there might be students who do not do their work or feel very insecure about explaining and talking in front of others. One aspect that emerged was “for me, it felt like I became some kind of extra teacher for the others I got to work with”. Some chose not to participate for this reason. It was, however, a good thing to be required to explain something (a concept, a mechanism) to others.

Strengths and limitations

An important strength in this study is that a rigid control group was used, as the students themselves constituted their own controls as they were compared to both themselves and former students. The students were examined at the same time on elements that belonged to both the flipped classroom intervention and the unchanged traditional lecture control. An additional strength is that the study used a mixed method approach, resulting in a better understanding and interpretation of the results using several data collection methods.

Although this study contributes to the scarce literature on flipped classrooms in nursing education, there are some methodological limitations worthy of mention. First, any causal relationships cannot be claimed due to the cross-sectional nature of the data. Second, quite few students responded to the questionnaire, and it might be that some important aspects from non-participating students having other experiences were missed. However, the low participation rate does not affect any association between variables. Third, there is likely to be a selection bias, with students who enjoy the flipped classroom approach more likely to participate in the survey and write comments in the open-ended boxes. Fourth, the qualitative data from the open-ended questions were not in great depth and interviews may have provided more information. These two above mentioned aspects might have influenced the transferability of the results negatively. Finally, there were two different teachers who taught microbiology and the immune system. However, they were the same during the two semesters, so any differences arising from the teachers should have been present in both the control group and intervention group and thus not be the cause of any differences when the groups were compared. Apart from the intervention, the teaching strategies were kept constant during the two semesters. Overall, due to a small sample size the generalizability of the results is limited. However, the results are consistent with previous literature, which show a positive effect of flipped classroom on academic achievement in the nursing program as well as in the education to other health disciplines such as medicine and pharmacy (Lelean and Edwards, 2020; Youhasan et al. 2021).