1 Introduction

Online education, pushed by the coronavirus, went global in 2020, affecting 1.5 billion students worldwide [1]. However, despite some attempts to label online education during COVID-19 as a “great equaliser”, it emphasised digital inequities among students. [2]. The lack of preparation of all actors involved in online education, like students, teachers, and parents, was one of the main hardships in implementing online schools [3]. In such challenging circumstances, computer science or informatics knowledge and skills are essential for students and teachers. Since there was no time to prepare for online schooling, pedagogical nor technical, computer science teachers became essential factors in implementing and administrating online tools and as technical support for other teachers and students in Croatia [4]. Such a role, besides preparing online classes and other actions related to teaching, contributed to overburn felt by computer science teachers which was more emphasised at the elementary school level.

But how much is computer science (or informatics) represented in schools as a subject, and what are the qualifications of respective teachers? The answers to posed questions could reveal the less-noticed importance of Computer science as a discipline.

Computer science or informatics (hereinafter CS), as a compulsory school subject from the elementary school level, has been considered necessary for at least the last decade. In terms of digital equity, all students should have “access to hardware and connectivity; autonomy of use; digital and literacy skills; availability of technical and social support; and access to educators skilled in the use of information and communication technologies.” [5]. Despite many worldwide initiatives promoting and popularising programming and informatics in general and formal coalitions aiming at establishing Informatics as a fundamental discipline for all students in school [6, 7], many countries still need to implement Informatics as a separate school subject. Appropriately educated and skilled informatics teachers are “a vital requirement” for implementing CS in formal K-12 education [8]. Countries planning to implement CS in the curriculum have a problem with the lack of CS teachers [9] educated to teach fundamental domains like computational thinking, programming, operating systems, computer networks, hardware, data processing, and digital agility. Therefore, besides basic pedagogical knowledge common to all teaching areas, CS teachers should possess knowledge of didactics of CS and broader CS-related knowledge [10,11,12]. The recent lockdown during the pandemic of 2020. additionally highlighted the need for formally educated Informatics teachers.

CS teachers have been educated at the University of Split, Croatia, since 1988. [13]. At the same time, CS, as an independent school subject, has been implemented in elementary (K-12) and high school (K-16) curriculums. As in many countries, in Croatia whole education system went online from March 2020 to the end of the school year 2019/2020 [14]. The role of CS teachers during online education was decisive. CS teachers, besides teaching, had two critical roles in their respective schools, (1) they actively participated in implementing tools for online teaching, and (2) they acted as a "helpdesk" for their colleagues, students and their parents in coping with the technical issues during online schooling [4]. The pandemic showed us the importance of CS teachers in teaching CS and supporting school administration in technology-related changes. Furthermore, CS as a separate school subject has also been emphasised as a crucial factor in adopting online technologies by students, thus minimising digital inequity [5] among students and teachers.

This paper deals with attitudes and roles of CS teachers during online schooling the first lockdown and two years after at elementary (K-12) and high school (K-16) levels. The first part of the research deals with the CS teachers' attitudes towards online education during the lockdown. The second part is focused on the K-12 informatics teachers' perspective of online education two-years-after the COVID-19 period when teachers and students were accustomed to online surroundings. Semi-structured interviews were conducted to determine the effects of fully online education during the COVID-19 lockdown on traditional education.

1.1 The prerequisites for online education in Croatia

CS is a separate school subject from the 1990s at the K-12 level in Croatia. Simultaneously, CS teachers have been educated following specific university curricula to acquire CS-related and pedagogical qualifications [4, 11]. Croatian Qualifications Framework (CROQF) represents the main framework for reforms in education in Croatia. Recently CROQF defined CS teacher occupation standards, which can be found in the CROQF register [11]. The standards align with the core topics defined by the informatics reference framework [12]. The e-learning area is one such qualification for CS teachers and part of the CS curriculum for all students. As a result, CS teachers in Croatia are generally proficient in managing and utilizing e-learning platforms, and most students are knowledgeable about using e-learning environments.

Furthermore, several national public institutions are operating within the Ministry of Science and Education for handling the technical and software infrastructure of science, elementary and high schools, and higher education. The main institutions are the Croatian Academic and Research Network [15], The University Computing Centre [16] and the Authentication and Authorisation Infrastructure of science and higher education in Croatia [17]. The Croatian Ministry of Science and Education has an agreement on the lease of the use of Microsoft software products for public elementary and higher education institutions and public scientific institutes, which provides a license for all actors in the educational system for Microsoft Office tools [18]. Additionally, as part of the national curricular reform in 2018, every school in Croatia was equipped with internet access, every elementary school provided tablet devices to students in at least the sixth grade, and every teacher underwent training sessions on online teaching using the Loomen (CARNET's adaptation of Moodle) LMS. Every teacher and student also had an AAI account which granted access to Loomen and registration for Microsoft Office tools, among others. Thus, all parties involved in the Croatian educational system, including teachers and students, had the necessary resources for transitioning to online learning during lockdowns.

According to all mentioned above, it is not surprising that during the lockdown from March 2020 in Croatia [14], the most used platform for asynchronous learning was Loomen (Moodle), and the most used platform for synchronous learning was MS Teams. The most used e-learning instruction delivery was the combination of asynchronous and synchronous [4, 18]. Furthermore, the teachers and students were not unprepared for online education since all teachers attended more than one online course for using LMS in pre-pandemic time, every school had internet infrastructure, and students from fifth grade had informatics classes in school. Therefore, we can say that there is digital equity [5] among K-12 students and teachers in Croatia. However, they also encountered many problems. Some problems are investigated, and the results are presented in the Methodology section.

1.2 Literature review

During the COVID-19 lockdown, teachers and students had to go online with no time for preparation. The main problem in coping with online education can be attributed to the need for teachers' and students' technical knowledge to use tools for online learning [19]. While in-person, or face-to-face (F2F) education offers more advantages than online education, particularly for younger students in the K-12 range [20] prior to COVID-19, there was no widespread and full online education at the K-12 level globally. Consequently, literature concerning online schooling at the K-12 level remains limited.

A search of the WoS database using the query "COVID-19 AND K-12 AND education AND ("computer science" OR "informatics")" yielded a mere 13 papers published after April 2020. A closer examination of titles and abstracts revealed that only three of these papers relate to K-12 education during the pandemic. Employing a snowball methodology identified additional seven papers relevant to this research.

The overall state of distance education preparedness during COVID-19 was notably deficient, with a preference for blended learning designs [21]. Significant factors impacting students' motivation were their attitudes towards ICT, internet usage for educational purposes, and teachers' positive perspectives on e-learning [22]. Moreover, K-12 students express workload and stress with school tasks and lack of social contact with classmates and teachers [23]. However, no matter what kind of online tools were used, complete online education during COVID-19 highlighted digital inequities among students at the K-12 level in both devices used and digital literacy, especially in lower-income families [24].

Technical infrastructure emerged as a major hurdle in implementing online education, both for students and teachers. In Germany, insufficient digital equipment for students posed a substantial challenge during lockdowns. Remarkably, in 2018, 36% of 8th graders from disadvantaged backgrounds possessed no more than one necessary digital device [25]. Poland faced comparable challenges to Croatia, as they conducted a recent reform mandating informatics education in primary schools [26].A study [26] revealed that majority of Polish primary school students were fully supplied by ICT devices for online learning during the COVID-19, which increased students’ CS learning outcomes in extracurricular settings.

Many teachers working in public schools did not have adequate prior experience using online tools, thus they were struggling to shift to online modes in Cyprus. The research indicated that teachers who have a positive attitude towards distance learning are more likely to frequently use digital teaching strategies [27]. Furthermore, it is revealed that training and support in a school are significant predictors of teachers’ technological pedagogical content knowledge and perceived usefulness [28]. In Turkey, K-12 teachers and students used social media such as Twitter for online education discussions. Among eleven categories of themes, demonstration of sample distance education activities took 44.02% of all themes, followed by information related to planned/ongoing distance education processes at 34.14% [29]. Such findings are important for teachers’ readiness for adopting online learning elements into regular teaching in post-COVID time.

In comparison to non-CS teachers, educators with computer science backgrounds exhibited more positive attitudes towards the shift to online teaching and learning across educational levels and contexts [30]. Although many school teachers did not have enough experience with online learning and teaching [31], during the crisis, all teachers had to quickly adopt tools for online learning which even highly skilled teachers found time-consuming and challenging due to a lack of training for such task [32]. Therefore, teacher education should be reshaped to be more flexible for learning new technologies [32] by investigating prepared teachers' experiences for online learning for future online learning [31].

As mentioned earlier, a gap in knowledge persists regarding the implementation and maintenance of online tools in schools during the initial lockdown, alongside the provision of technical assistance to fellow educators and students. In Croatia, these responsibilities primarily fell upon computer science teachers, resulting in increased workloads.

2 Methodology

The research was conducted in two phases. The first phase was conducted during the first lockdown caused by the COVID-19 pandemic in March 2020. Such a situation was new to all actors in the educational system. Therefore, the idea was to detect the coping of CS teachers in such circumstances. Since switching to online started CS teachers. The second research phase was conducted two years later in "normal" circumstances. In the meantime, there were several partial lockdowns due to the COVID-19 outbreaks. Teachers and students were accustomed to online surroundings this time. To gain insight into online education's positive and negative consequences, the author conducted semi-structured interviews with CS teachers in elementary schools.

2.1 Research questions

Since there are known problems with implementing informatics as a school subject available to all students and the problems in online education caused by COVID-19, the purpose of this paper is to detect the impact of CS as a school subject and the role of qualified CS teachers in coping with switching to online education.

RQ1—What were Computer Science teachers' attitudes toward online education during the first lockdown in Croatia?

RQ2—What were the main Computer Science teachers' roles during the first lockdown in Croatia?

RQ3—What are the effects of online schooling on education after the lockdown period?

2.2 Research context

The first research phase was conducted during the first lockdown caused by the COVID-19 pandemic. Schools in Croatia went completely online from March 16, 2020, and remained online until the end of the school year 2019/2020. Such a situation was new to students and teachers as well. Since there was no time for technical and pedagogical preparation, there were no strictly defined online tools for conducting online education. Instead, decisions for choosing online tools and their maintenance are left to schools. CS is a separate school subject in Croatia in elementary and high schools, and in every school, works at least one CS teacher. Therefore, CS teachers mostly actively participated in the technical preparation of online tools and supported other teachers using them. Because online education lasted for months, such a role could be led to overload and frustration.

The Google Form survey was designed to gain insight into teachers coping with the new situation. Before conducting the study, the Croatian Ministry of Science and Education approved using a survey for teachers. From May 18 to June 29, 2020, the survey was conducted among 538 elementary and high school CS teachers across Croatia. The surveys were distributed by e-mail to CS teachers. Participation was voluntary and anonymous.

The second research phase was conducted in June 2022, two years after the first lockdown. In the meantime, there were several partial lockdowns, but schools are expected to be prepared for that. Nevertheless, this "new normal" education impacted teachers and students. To explore CS teachers' perspectives on the "new normal" and the impact on education from the first lockdown, the author conducted face-to-face semi-structured interviews with eight elementary school teachers. Before the interviews, all participants were informed about the research purpose and voluntarily agreed to the interview. The anonymity and confidentiality of participants are guaranteed. Therefore, teachers' identities are hidden. The author coded teachers and schools to hide their identities in the paper.

2.3 Participants

During the first phase of the research, participants were elementary and high school CS teachers.

Among 538 CS teachers surveyed, 351 were elementary school and 187 of high school teachers. The ratio by gender in elementary school is 73.22% female and 26.78% male teachers. In high schools, 65.24% were female, and 34.76% male teachers. Even though the computing sector is considered a "men" area, in the educational field, women are more represented.

Since CS teachers have been educated in universities in Croatia since the 1988, we also wanted to explore the teachers' teaching experience. Figure 1 shows the distribution of teachers' work experience in school.

Fig. 1
figure 1

Distribution of participants by teachers' working experience in school

Full size image

More than 75% of teachers have more than six years of working experience as school teachers. Therefore, most of them are skilled teachers.

During the second phase of the research, the author interviewed eight teachers working in elementary schools in Split, Croatia. The decision to focus exclusively on elementary school teachers is attributed to the following reasons: (a) all teachers within this group instruct informatics according to the national elementary school curriculum, whereas informatics curriculum in high schools and the duration of informatics education vary based on the type of high school; (b) elementary schools exhibit a notably higher concentration of CS teachers compared to high schools.

There are 28 elementary schools in Split and ~ 70 CS teachers teaching from 5 to 8th grades in the town of Split. The schools vary in the number of students and teachers. Therefore, the author picked a representative sample of eight interviewed teachers working in five elementary schools, by the most relevant varieties: school size and teacher’s background. Such sampling fits into non-probability purposive sampling [33]. Interviews were conducted face-to-face individually. Table 1 shows the school’s details.

Table 1 Schools’ details
Full size table

All teachers, except T4 (English teacher), are CS teachers by vocation, and they actively participated in implementing online environments and providing technical support for other teachers. The reason for interviewing T4 was to acquire insights from a non-CS teacher regarding the role of CS teachers during the lockdown.

2.4 Assessment instruments

For the first phase of the research, the survey was used for collecting data. The author designed and conducted semi-structured interviews for the second phase of the research.

Survey for teachers. The pilot survey was conducted from May 11 to May 17, 2020, among 11 elementary (n = 6) and high school teachers (n = 5). Based on teachers' feedback, some questions were preformulated. Additionally, some questions related to students' evaluations are added based on CS teachers' suggestions. The final survey was dispatched to teachers by e-mail. The teachers could take the survey from May 18, 2020. to July 1, 2020. The survey contained 35 questions grouped into three sections: general data, online classes, and IT-related administrative work. Appendix A shows the questions. This paper's author processed questions about attitudes toward online education and CS-related administrative jobs. Such questions (n = 12) were Likert-scale questions (1–5) and had a Cronbach alpha of 0.839, which indicates a good level of internal consistency [33].

Semistructured interviews. In semistructured interviews, the researcher defines the topic and the main questions, while depending on the answers, the new questions are open-ended [33]. Therefore, the author found that such an instrument is most appropriate for this research. The survey results provided insight into the roles of CS teachers during the lockdown. Accordingly, the author designed a semi-structured interview with three main questions related to the informatics teachers’ role during the lockdown and in the next two years. One question was related to CS teachers’ role during online education in the first lockdown, the second was related to adopting elements of online education in regular classes, and the third was related to CS teachers’ role in schools two years after the lockdown. Some other questions were raised during the interviews. The author conducted the interviews from June 1, 2022, to July 1, 2022. The interviews lasted for ~ 60 min. The author took notes during the interviews, which were documented as transcripts immediately after the interviews. Afterward, the author conducted a content analysis [33] emphasizing the keywords used during participant interviews.

3 Results

The author describes the research results based on the proposed research questions in this section.

3.1 RQ1 What were Computer Science teachers' attitudes toward online education during the first lockdown in Croatia?

Factor analysis is used to reduce a large amount of data into a smaller and more manageable set. Factors are listed according to factor loadings or how much variation in the data they can explain. Exploratory factor analysis is used to explore the data and identify factors without prior theory or hypothesis about factors [34].

The exploratory factor analysis was conducted for 12 variables and 538 CS teachers to determine the main factors related to CS teachers' roles and attitudes in online school during the first lockdown. Therefore, the amount of data for factor analysis is satisfied with the ratio of > 30 cases per variable. Furthermore, the high Kaiser–Meyer–Olkin value (KMO = 0.796, Bartlett's test p = 0.000) indicates that the sample is highly adequate for factor analysis. The Principal Component Analysis (PCA) using varimax rotation was used to identify the underlying factors of the questions. The analysis showed four factors explaining 62.846% of the variance. Table 2 shows the result of a varimax rotation (factor loadings =  > 0.30).

Table 2 The result of a varimax rotation
Full size table

Out of 12 items, factor analysis yielded three factors. Five items loaded onto the strongest first factor (31.903% of variance). All five items are associated with the frustration of CS-related administrative work related to setting up an online environment for their schools, feeling overloaded and underrated for such work. Therefore, the factor is labeled "Frustration with CS-related work in school".

Four items were loaded onto the second factor (14.815% of variance) related to personal satisfaction with online classes. These items are associated with satisfaction with online classes, students' feedback, and the availability of online tools and online multimedia materials for students. The second factor is labeled "Satisfaction with Informatics online classes".

Finally, three items were loaded onto the third factor (10.117% of variance) related to the frustration in online classes. The items are associated with frustration with online classes. Thus, the third factor is labeled "Frustration with online education".

3.2 RQ2 What were the main computer science teachers' roles during the first lockdown in Croatia?

Based on the survey’s results, the interview questions were designed to gain more in-detail insight into informatics teachers’ roles in online education during the first lockdown. The interview question related to RQ2 was “What was the role of Computer Science teachers in schools during the first lockdown?”.

All (8) interviewed teachers agreed that the role of CS teachers during the first lockdown was "enormous". Besides setting up online environments in their schools, CS teachers had the most significant role in technical support for other teachers. T1 said that she was online "all the time", which was even helpful because it was a lockdown period, so it helped to "keep the mind busy". Similar statements had also T3, T5 and T8. T4 (English teacher) said that she felt ready for Loomen because of mentioned Loomen courses for preparation for the school reform. Though, all (7) CS teachers noted that the main frustration was that they felt like a "helpdesk" for school administration which was forwarding all jobs related to online school to them. Besides, other teachers often contacted them for help with the online tools but students and parents also because sometimes they had technical issues and needed help. Usually, they were approached by phone, which was very often, thus time-consuming. All CS teachers got the impression that, as T4 confirmed, the lockdown forced other (non-CS) teachers significantly improve digital literacy because they were forced to use it, which they would probably want.

3.3 RQ3 What are the effects of online schooling on education after the lockdown period?

In semistructured interviews, the author posed three main questions about the effects of online schooling on education after the lockdown period. One question is related to adopting elements of online education in regular classes, the second is related to CS teachers’ role in schools two years after the lockdown, and the third one was related to students’ habits after the first lockdown. In the following sections the questions are presented along with the analysis of the answers.

3.4 Did schools incorporate the elements of online education into regular teaching?

Undoubtedly, all interviewed teachers uniformly confirmed that online education during the COVID-19 lockdown contributed to teachers' digital literacy. Non-CS teachers, especially older ones, resisted using online tools for teaching. However, since the lockdown forced them to use online tools, they got used to it, with the significant support of CS teachers. The result is that schools are now ready to go online instantly. For example, T1 remembered that once, in the current school year, classes in S1 could not be organised due to national exams. Before, classes would be dismissed, but since all teachers are now used to online school, they quickly switched online for that day, with no need to make up for missed school hours. It would not be possible without online alternatives.

S2s' teachers (T3, T4) highlighted the advantage of hybrid teaching when some students were in isolation. With the help of online tools, such students could participate more actively in regular classes.

However, T5 and T8 said that next school year (2020/2021), some teachers continued to contact students on MS Teams, although classes were face-to-face in S3 and S5. Some teachers were sending teaching materials and tasks on MS Teams which was a problem because students did not check their MS Teams classes because online school was over. In the end, the school administration in S3 instructed teachers to use MS Teams only for students in isolation but not for regular classes. S3 did not usually use MS Teams for real-time classes.

However, all teachers agreed that face-to-face teaching is not replaceable with any online tool for long-time periods. However, online elements could help in the short term, and some aspects of online school could be used regularly, such as placing teaching materials, examples, quizzes, etc.

3.4.1 The reasons for not using real-time online tools

During interviews, another issue was raised about using real-time online tools. Most interviewed CS teachers (6) noticed that a significant number of other teachers showed resistance to using online tools in a real-time. In Croatia, in the school year 2020/2021, schools could conduct face-to-face classes in model A, but only if a school could satisfy epidemiological prerequisites in several physical classrooms. Therefore, schools with many students and not enough classrooms could conduct classes in model B, a "switch-mode", a combination of face-to-face and online school changing turns every week. Because of that, only S3 dominantly used MS Teams for real-time lectures, while such occurrence was rare in other schools. The reason is that S3 was the only school holding classes by model B during the school year (2020/2021). As T1 and T2 reported, even in S3, most teachers did not hold real-time lectures at the start of the school year. After some time, because of model B school mode, parents started to press the school administration for real-time classes because their students were not active enough in asynchronous mode and were losing their working habits. So, most of the teachers started to hold real-time lectures but still not all of them. In other schools, most teachers, in case of switching online (due to a partial short lockdown), which were mainly two-week periods, held their lectures in asynchronous mode. There was apparent resistance to real-time lectures, despite being much closer to face-to-face classes than asynchronous online classes. All teachers addressed similar reasons: (1) "fear" of being publicly exposed to students' household members and social networks, and (2) insecurity in using the online tool (teachers did not want to). Nonetheless, the most dominant reason seems to be fear of public exposal.

3.5 Did CS teachers' role in schools change after the first lockdown?

T3 noted that her role as a "service activity" for technical issues in school changed, not because of her different formal status but because of her attitude. Although before the lockdown, she had a similar role, the "CS helpdesk" jobs were acceptable because it was occasional and mostly related to the end of the school year when other teachers needed help with the e-school administration. However, during the lockdown, she was overloaded with CS-related administration jobs. The problem is that some teachers refuse to improve their digital literacy. Instead, they were looking for the informatics teachers' help because "it is their field", culminating during the lockdown. The result is that now it is easier to reject such requests because it is not a formal obligation of informatics teachers. For example, a biology teacher is not obligated to care for the school garden, a Croatian teacher to lecture other teachers' texts, and so on. T4 said that the online experience stimulated her digital literacy because if it were not for online education, she probably would not force herself to learn to use online tools. Therefore, T4 was one of the non-informatics teachers who did not reject using online tools. Moreover, she feels that she forced herself to embrace technology elements in teaching.

3.6 Did students change their habits related to their school obligations during the last two years?

The question related to students’ habits was raised during the interviews. Since even in the first interview, the teacher expressed an opinion about changing students’ habits, the topic is mentioned in all interviews.

At the beginning of the first lockdown, students were mainly very active and eager to actively participate in an online school. However, as time passed, students' enthusiasm for online school decreased. The decrease in students' activities was very noticeable during the 2020/2021 school year, which was conducted in "mixed mode", a combination of face-to-face and online school. Students started to experience online school as a "vacation" and invested less effort in learning, which was reflected in their grades and working habits. Some schools, like S1, had to organise a school in the B model due to many students and limited physical classroom capacities. Schools working in the B model organised school as shifting from one week online and the other week face-to-face.

Additionally, there were several two-week periods when all schools went online due to the high rates of COVID-19 infections. Therefore, students in S1 spend at least half of the school year online. S1's teachers agreed that students constantly decreased their working habits, which became concerning. On the other hand, their grades did not drop; just the opposite happened. Some teachers awarded students with higher grades due to a stressful situation during the lockdown. Moreover, students' parents probably did a lot of homework and tests during online school instead of students, and teachers could not prove it. Overall, such a period resulted in a decrease in students' engagement in school obligations and increasing in their grades.

T3 and T4 agreed that students were not awarded with higher grades after the lockdown, but there is an impression that teachers lowered the expectations and criteria. By lowering criteria, students put in less effort to achieve set expectations. In addition, there is a feeling that students are getting less concentrated. They decreased their school-related working habits, not directly due to online education but rather due to an "overuse" of mobile phones and information overload caused by the internet.

4 Discussion

In many countries worldwide, schooling at all levels faced emergency remote teaching (ERT). Such an emergency shift to online schooling brought forth a lot of pedagogical and technical issues. Non-computer science teachers probably encountered more problems, but these emergencies also had a significant impact on them, enhancing their readiness for ERT [35]. Furthermore, based on the literature review, professors faced several critical challenges: (1) limitations on internet access and appropriate communication infrastructure, (2) lack of access to computer equipment and hardware, and (3) low levels of familiarity and necessary training for teachers and students in working with virtual education systems [36]. Another important issue in ERT was the implementation and administration of online tools for online schooling. In Croatia, computer science (CS) teachers have been educated to acquire CS-related and pedagogical qualifications [4, 11]. Moreover, the Croatian Qualifications Framework (CROQF) has defined CS teacher occupation standards [11]. Therefore, CS teachers in Croatia were technically and pedagogically prepared to cope with not only technical issues but also pedagogical challenges in helping their colleagues and students which is in line with the study that revealed that CS practitioners teaching on all educational levels and context had significantly more positive attitudes towards the move to online LT&A compared to non-CS teachers [30]. Nonetheless, the significant role of CS teachers in ERT in Croatia also raised several problems for them. Table 3 shows the research questions and answers related to CS teachers' roles and attitudes towards online education raised during data analysis.

Table 3 CS teachers’ roles and attitudes towards online education detected during analysis
Full size table

Table 3 shows categorised CS teachers’ roles and attitudes towards online education detected during analysis. RQ1 categories are based on the factor analysis. RQ2 is based on the teachers’ comments left in the survey in open-ended questions and interviewed teachers’ comments confirming data collected by the survey. RQ 3 is based on the interviewed teachers’ comments. While RQ1 and RQ2 are related to the past RQ3 is important for the future. Since non-CS teachers were insecure in using online tools, they had to adapt to new circumstances which impacted their digital literacy. Such finding is in line with previous studies also revealing that many non-CS teachers did not have much experience with online learning [27, 31] but they had to quickly adopt online tools during the pandemic [32]. Even experienced teachers found this challenging and time-consuming since they had not received training on how to use these tools. Studies have shown that teachers who are more positive about distance learning are more likely to use digital teaching strategies regularly [27]. Additionally, research has shown that training and support from schools can help teachers develop their technological pedagogical content knowledge (TPACK) and perceived usefulness of online learning [28].

These findings are important for teachers' readiness to incorporate online learning elements into their regular teaching after the COVID-19 pandemic. Therefore, teacher education should be updated to be more flexible and adaptable to new technologies. This can be done by studying the experiences of teachers who are already prepared for online learning so that we can better prepare future teachers for this type of teaching.

4.1 Limitations of the study

The first phase of the research was limited by several factors. The questions in the survey were not utilised by the previous scale because of the unique circumstances during the first lockdown. Since there was no previous evidence of the impact of complete online education on teachers and students the questionnaire is designed during the lockdown. However, pilot testing was conducted, and the survey was updated accordingly. Furthermore, reliability is proved by Cronbach alpha coefficients approving the internal consistency level of the instrument. Another limitation is a lack of precise information about the total number of CS teachers in Croatia and the response rate.

The main limitation of the second phase is a lack of interviews with the high school teachers. The author conducted interviews with elementary school teachers for several reasons. Since interviews are time-consuming and sampling is difficult, the author interviewed elementary school teachers because quantitative analysis showed that they felt more overloaded by additional work related to implementing and maintaining online tools during the first lockdown. Additionally, sampling of high school teachers is difficult due to the diversity among them. For example, there are several types of high schools where computer science teachers have different roles and computer science school subject is not the same for all high schools. In elementary schools, all computer science teachers have the same roles and follow the same curriculum.

5 Conclusions

Online education from March 2020, caused by the COVID-19 pandemic, made drastic changes for all actors in educational systems. Such a sudden shift warned us that we must improve digital literacy to keep up with the digital age. One of the problems at the beginning of online education during the first lockdown in schools was the setup and administration of online tools. In Croatia, CS has been a separate school subject in elementary and high schools for the last three decades. At the same time, CS teachers have also been educated at the university level to gain pedagogical and CS-related knowledge. Consequently, the responsibility of implementing online tools for education during the lockdown mostly fell on CS teachers.

Since every elementary and high school in Croatia has at least one CS teacher, besides their CS classes, they mostly took assignments for setting up and administering online environments in their schools. At the beginning of the lockdown in March 2020, nobody knew it would last until the end of the school year in June 2020. In May 2020, the survey was conducted among 538 elementary and high school CS teachers across Croatia. The exploratory factor analysis showed three factors among teachers: (1) Frustration with CS-related work in school, (2) Satisfaction with CS online classes, and (3) Frustration with online education. Two years later, although there weren't such long-period lockdowns, the online education elements are still present in the "new normal" surroundings. Thus, the author conducted semi-structured interviews with eight elementary school teachers to determine the impact of online education on the "new normal". Teachers during interviews confirmed that CS teachers had a significant role in implementing online schools, from technical and supportive roles to other teachers. The results of the interview analysis were consistent with the three factors mentioned above. Still, the "new normal" formal education is affected by online schooling, and some online practices are now adopted into regular teaching.

Although the role of CS teachers in Croatia in the "new normal" is not formally changed, the educational environment has been altered in many ways. For example, the digital literacy of other teachers is improved, and they feel less fear of technology. Furthermore, since all teachers had to use different online environments, sudden short-term switching to online is much easier. Anyway, in the “new normal”, all teachers should be digitally literate, while educated CS teachers can have a significant and formal role in setting up digital educational environments in emergency circumstances, such as online education during the COVID-19 lockdown.