The study was conducted at a nursing school at a university in the Western Cape, South Africa. The school offers a 4-year undergraduate degree in nursing. Students were already exposed to a variety of educational approaches, for example, online learning management systems, clinical skills laboratory and traditional face-to-face teaching. But at the time of the study, there was no formal blended learning approach built into the nursing programme.

A one-group pre- and post-evaluation study using pre- and post-surveys using a self-administered questionnaire was conducted to assess nursing students’ perceptions of using a blended learning approach during the pandemic.

The target population was 150 third-year nursing students enrolled in the undergraduate nursing programme for a selected third-year module. All-inclusive sampling was used, and all the third-year students were given the questionnaires to complete at both times of data collection. This group of students was specifically chosen for this study because of their unique position in their academic journey, having completed 2 years of traditional face-to-face classes, prior to the challenges brought about by the COVID-19 pandemic and then having to adapt to a blended learning approach with online lectures and face-to-face clinical training.

A self-administered questionnaire was used to collect pre- and post-group data with a validated and reliable scale on student perceptions of blended learning (Shantakumari & Sajith 2015). The scale has 24 statements covering three domains in blended learning; process, content and ease of use (Shantakumari & Sajith 2015). Each statement was rated on a scale from 0 to 4 ranging from strongly disagree (0) to strongly agree (4). The questionnaire was in English and was piloted with five nursing students, resulting in no changes to the scale and the inclusion of the pretest data. The scale has good internal consistency (α = 0.87) (Shantakumari & Sajith 2015), similarly to this study (α = 0.87), with the three domains also showing good internal consistency, process (α = 0.81), content (α = 0.73) and ease of use (α = 0.51).

The pre-survey data collection was done during the orientation week in February 2021 prior to implementation of the community health nursing module. The post-data were collected using the same questionnaire, post-completion of the module in June 2021, during the last clinical skills session. Completion of the questionnaires was voluntary, anonymous and took 5 min to complete. Written informed consent was obtained from respondents of the study.

Data were captured and analysed in SPSS v28.0 (IBM Corporation, Armonk, New York, US). Students’ perceptions of the three domains were analysed using means and standard deviations as per scale instructions for each of the domains and for the individual items. When analysing the responses of individual items in the questionnaire, a mean score of 2 was used as a reference value to analyse perceptions. A score of 2 or more for an item was regarded as positive, while less than 2 was considered negative (Shantakumari & Sajith 2015). Statements that were negatively phrased were reversed. The overall scale had a maximum attainable score of 96, with higher scores meaning more positive perceptions. Perceptions were classified into three categories that reflect the extent of the positive perceptions related to blended learning: 28–53, i.e. relatively low level of positive perceptions, possibly indicating reservations or concerns; 54–80, i.e. moderate to high levels of positive perceptions, possibly finding value in blended learning and > 80, reflecting a high level of positive perceptions and possibly strong endorsement and satisfaction (Shantakumari & Sajith 2015). Independent samples Chi-squared tests were performed to determine differences between pre- and post-survey demographics. Independent samples Mann–Whitney U tests were performed to determine whether there were differences in the mean scores of the students’ perception of blended learning pre and post-implementation of the module. The significance level was set as p < 0.05.

Ethics approval was obtained from the University of the Western Cape Humanities and Social Sciences Research Ethics Committee (HSSREC), reference number HS21/3/2, and permission to conduct the study was obtained from the university registrar and the director of the nursing school. Following the ethics approval, information about the study was distributed to third-year students. Written informed consent was obtained from those respondents who agreed to participate. Students were free to voluntarily participate or withdraw without penalty. Questionnaires were coded and anonymous. All data were securely stored.

The questionnaire was completed by 128 (85.3%) third-year nursing students prior to the implementation of the module and 95 (63.3%) respondents after the module was completed. There were no significant differences observed in demographics between the pre- and post-group respondents with most of the respondents being female (113, 88.3% pre-group; 84, 88.4% post-group), and the majority were 24 years and younger (105, 82.7% pre-group; 79, 84.0% post-group) (Table 2).

There were no significant differences in technology access during the implementation with more than half of the respondents (70, 54.7% pre-group; 60, 62.2% post-group) reporting that they had a laptop for accessing study modules and most reporting that they owned a cell phone with Internet access (120, 95.2% pre-group; 93, 97.9% post-group) and Internet at home (108, 86.4% pre-group; 78, 83.9% post-group). Email appeared to be the preferred mode of communication (83, 66.4% pre-group; 68, 73.1% post-group) (Table 3).

Before the implementation of blended learning, over two-thirds of the respondents indicated that they would prefer the blended learning approach over traditional face-to-face learning (86, 68.8%). This remained so post-implementation of blended learning with a near significant increase of 10.1% (75, 78.9%; χ² = 2.832, p = 0.092).

There was no significant difference in the overall blended learning score pre and post the blended learning implementation (pre 64.96/95, ± 11.47 vs. post 64.60/95, d = –0.36 ± 11.59; W = –0.11, p = 0.014). Around three-quarters of the respondents’ score were in the 54–80 group, i.e., moderate to high levels of positive perceptions, increasing post-implementation (73, 72.3% pre-group; 68, 81.2% post-group), because of a decrease of 5.6% in the score post-implementation in the > 80 group (12, 11.9% pre-group; 5, 6.0% post-group), i.e., high level of positive perceptions (Figure 1).

Perceptions of blended learning were measured in terms of the blended learning process (8 positive and 2 negative statements, Table 4), blended learning content (9 positive and 1 negative statements, Table 5) and blended learning ease of use (4 positive and 1 negative statement Table 6), with each statement rated out of a possible 4.

All domains had a score above 2, i.e., positive perceptions of the blended learning process (pre-group 2.55, ± 0.58; post-group 2.54, ± 0.63), content (pre-group 2.75, ± 0.52; post-group 2.79, ± 0.52) and ease of use (pre-group 3.07, ± 0.49; post-group 2.99, ± 0.58), with no significant differences in the overall rating score pre and post. However, the blended learning process domain was rated significantly lower on both occasions compared to score for content delivered via blended learning and ease of use, which was rated significantly higher in the pre-measurement (Figure 2).

Overall, the process domain was rated significantly lower than the other domains (Figure 2), though the overall score was above 2. The statement ‘I am in favour of incorporating blended learning to my module’ had the highest level of agreement (pre-group 3.11, ± 0.91; post-group 2.89, ± 1.01; U = –1.71, p = 0.088), though there was no significant difference post-implementation. The statement ‘Blended learning is more effective than traditional in class delivery’ had the lowest level of agreement (pre-group 2.56, ± 1.17; post-group 2.58, ± 1.33; U = –0.06, p = 0.950), with no significant difference post-implementation (Table 4).

In terms of the negative statements, ‘Blended learning is a waste of time’ had the highest agreement ratings from respondents, though significantly lower post-implementation (pre-group 3.13 ± 1.05; post-group 2.43 ± 1.33; U = –3.93, p < 0.001). ‘Incorporation of blended learning made my module more time-consuming’ had a low level of agreement (pre-group 1.59 ± 1.07 vs. post-group 1.52 ± 1.09; U = –0.80, p = 0.422), which did not differ significantly post-implementation (Table 4).

Overall, the blended learning content domain was rated significantly higher than the process domain (pre- and post-implementation) but lower than the ease-of-use domain, significantly so pre-implementation (Figure 2). In terms of the positive statements, ‘The online activities on the learning management system were related to the module objectives’ had the highest agreement rating both pre and post-implementation (pre-group 3.27, ± 0.73 vs post-group 3.14, ± 0.69; U = –1.49, p = 0.136). The statement, ‘The activities of blended learning helped me to learn more and interact with my colleagues on the given topics’, had the lowest agreement ratings pre and post-implementation, though the post ratings were higher though not significant. All ratings were above 2 indicating positive perceptions of the content taught using blended learning (Table 5).

In terms of the negative statements, ‘The entire module was difficult to follow’ had a high level of agreement (above 2) but significantly lower post-implementation (pre-group 2.41 ± 1.16; post-group 2.05 ± 1.19; U = –2.19, p = 0.028). The statement ‘The online activities were of long duration’ had low level of agreement (below 2) and decreased post-implementation though not significantly (pre-group 1.48, ± 0.96 vs. post-group 1.29, ± 0.92; U = –1.49, p = 0.135) (Table 5).

Overall, the ease-of-use domain was rated higher than the other domains, significantly more so before the implementation (Figure 2) with all the positive statements rated over 3 (Table 6). The statement ‘I found the learning management system easy to use’ was rated the highest pre (3.61, ± 0.65) and post-implementation (3.44, ± 0.67; U = –2.16, p = 0.030), though significantly lower post-implementation. Except for two items, ‘I was able to learn to use the learning management system quickly’ (pre-group 3.56 ± 1.16; post-group 3.34 ± 0.74; U = –2.17, p = 0.030) and ‘I was able to access the online content without any problems’ (pre-group 3.29 ± 0.81; post-group 3.01 ± 0.94; U = –2.19, p = 0.029), which both decreased significantly post-implementation, there were no significant differences in pre- and post-survey ratings for these items (Table 6).

The negative statement, ‘I felt my knowledge regarding using the learning management system was limited compared to my peers’ was rated significantly lower post-implementation, with the agreement dropping significantly post-implementation (pre-group 2.34, ± 1.36 vs. post-group 1.93, ± 1.34; U = –2.15, p = 0.032).

Though the blended learning process was rated significantly lower than the other domains, all the statements were positive about the process. There was some concern prior to the implementation that incorporating blended learning could potentially lead to time wastage (3.13, ± 1.05) though this dropped significantly post-implementation (2.43, ± 1.33). This was similar to concerns regarding the use of blended learning reported by other studies (Almahasees, Mohsen & Amin 2021; Lu 2021; Siraj & Maskari 2019; Wu et al. 2020), where it was hypothesised that perceived time wasting might be affected by factors such as the online structure of components, first-time user difficulties, lengthy resources such as videos (Siraj & Maskari 2019; Su et al. 2023; Swaminathan et al. 2021). Suggestions to address this perception and to improve comprehension of the content while also making the best use of the time spent by students include shorter videos, engaging activities (Armellini, Teixeira Antunes & Howe 2021; Becker et al. 2017; Siraj & Maskari 2019; Su et al. 2023; Swaminathan et al. 2021) and streamlining the content to ensure clear, engaging and goal-oriented educational resources (Almahasees et al. 2021; Su et al. 2023). In addition, asynchronous and flexible learning schedules allow students to engage with content at their convenience, reducing time waste and enhancing inclusion and accessibility in blended learning together with detailed timetables, targets and reminders (Almahasees et al. 2021; Armellini et al. 2021; Becker et al. 2017; Lu 2021; Su et al. 2023).

Overall, the content delivered via blended learning domain was rated positively with the design of the module and respondents’ comprehension of the resources influencing their perception of blended learning. The alignment of online exercises with module objectives received mostly positive feedback pre- and post-implementation (pre-group, 3.27 ± 0.73; post-group, 3.14 ± 0.69, ns). This was like findings in blended learning literature where participants expressed an interest in learning through online content, which allows them to access materials as needed, and the given content should be appealing and simple to grasp (Alexander & Boud 2018; Larson & Lockee 2019; Lu 2021; Swaminathan et al. 2021).

The respondents found the module difficult to follow (2.41, ± 1.16) pre-implementation, which reduced significantly post-implementation (2.05, ± 1.19), though this remained high with an agreement level over 2. This aligns with similar findings in another study (Langegård et al. 2021), which attributed this finding, to the nurse educators’ limited knowledge of technology, learning management systems and their ability to create comprehensive online content. A lack of experience might lead them to rely on methods traditionally used in face-to-face settings, such as presenting a slideshow lecture, without fully utilising the features of the learning management system or incorporating a more diverse range of online resources. Similar findings from other studies suggest that educators working in online environments face similar difficulties (Dailey-Hebert 2018; Johnson, Mejia & Cook 2015; Larson & Lockee 2019).

These findings suggest there is a need for clearer instructional design and communication strategies to improve content accessibility and understanding (Langegård et al. 2021; Lu 2021; Singh et al. 2021). This can be done by refining the module design to prioritise clarity and engagement. In addition, offering additional resources or support mechanisms can enhance participants’ understanding (Khan et al. 2021). Actively seeking and integrating student feedback on materials in the module is essential for ongoing improvement and alignment with students’ needs (Khan et al. 2021). In addition, nurse educators require consistent training opportunities and skill development if blended learning is to continue to be effective (Almahasees et al. 2021; Jowsey et al. 2020; Singh et al. 2021; Siraj & Maskari 2019).

The ease of use was measured in terms of the use of the learning management system, and this had high levels of agreement of the ease of use of the learning management system, pre- and post-implementation. However, concerns were expressed regarding self-perceived knowledge limitations compared to peers (pre-group 2.34, ± 1.36; post-group 1.93, ± 1.34) pre the module, which reduced significantly post-implementation. This suggests a potential gap in students’ confidence and online competence prior to the implementation of blended learning, which can play an important role in effectively using an online learning management system (Nwamu & Ni 2023; Singh et al. 2021). This also highlights the importance of ongoing support and training initiatives (Rasheed, Kamsin & Abdullah 2020; Siraj & Maskari 2019) prior to implementing blended learning. Training and preparation to ensure students have necessary skills and knowledge to sidestep difficulties, as was done in this study, can enhance the overall experience of using blended learning (Divjak et al. 2022; Khan et al. 2021; Singh et al. 2021). Other recommendations for training should also include accessible support materials (Langegård et al. 2021; Mukasa et al. 2021; Singh et al. 2021), and interactive online learning community where students can exchange experiences and assist each other to navigate the platform to create confidence and friendships (Divjak et al. 2022; Langegård et al. 2021; Singh et al. 2021). By combining these strategies, the perception of online learning management systems among students can improve, resulting in a more enjoyable experience.

The methodological approach was limited to a single university, and the participants were from a specific country. Additionally, convenience sampling was used, and the knowledge of the students was very specific.

It is essential to provide comprehensive training and familiarisation of students with blended learning prior to embarking on offering a previous face-to-face module in a blended learning format. Additionally, investing in the professional development of nurse educators is crucial to ensure that they can develop effective online learning activities. Lastly, the importance of user-friendly and intuitive technological platforms is essential to enhance the overall student experiences.