Design and sample

For this research, a cross-sectional descriptive study has been carried out in a representative sample of Spanish teachers. The National Institute of Statistics (Instituto Naciona de Estadística (^{INE), 2020}) censused the population of Spanish teachers at 847,713. According to the INE data, estimating a sample heterogeneity of 50%, a confidence level of 99% and a margin of error of 5%, it would be necessary to reach a sample size of 663 teachers. However, for this work, a sample of 965 teachers of different educational levels from all the ACs of Spain has been obtained.

The questionnaires were completed virtually during the month of August 2020. An email with the protocol and instructions was sent to all practicing teachers in Spain through a database of emails from public, concerted and private educational centres in Spain. Thus, a sample of 965 teachers was obtained who indicated their agreement to participate in the study voluntarily and asked the different questions on the form sent by email. In the same way, in the questionnaire itself it was indicated to all participating teachers that the principles of the International Code of Ethics in Humanities and Social Sciences of the Center for Research Ethics & Bioethics (^{European Commission, 2021}) and Law 15/1999 on the Protection of Data from Spain (^{Gobierno de España, Jefatura del Estado, 1999}) were followed. In turn, the research protocol was approved by the EDUCA Ethics Committee (Code 62020).

Instrument and variables analyzed

The instrument used for this study has been an ad hoc questionnaire that collected information on the teachers’ perception of the anti-COVID-19 protocols designed by the central government and the Autonomous Communities, to return to the classroom in Spain. The questionnaire was prepared based on other related research on this topic (^{Chang & Yano, 2020}; ^{Feito Alonso, 2020}) and was validated through the judgment of experts (^{Lynn, 1986}). To do this, we enlisted the help of seven specialists in education and health. In this way, a first draft of the questionnaire was raised and we sent it to the group mentioned above, sending us their comments and suggestions. With this information we re-elaborated and modified the questionnaire that was sent again, so that they could verify the suitability of the questions of the modified instrument. Finally, these seven specialists issued their assessment, showing an agreement level of 85%, so it was assessed that the questionnaire was correct to be applied to the 965 participating teachers, according to the criteria of ^{Polit et al. (2007}). The survey administration protocol and reliability were determined by standardizing it for all participants.

The questionnaire was made up of a total of 38 items aimed at knowing the perception and feelings of the teaching staff about the measures imposed by the national and regional administrations and the return to the classrooms in the 2020-2021 academic year. For this publication we have focused on the study of the following items directly associated with the teachers ‘perception and feelings at the beginning of the course: Do you think that the administration is managing well the planning of the students’ return to the classroom? (P1), Do you feel supported by this new situation by the state and regional educational administrations? (P2), With the beginning of the new school year, do you feel safe teaching in person at the centre? (P3), From your point of view, should the working day be reduced? (fewer hours at the centre, 4-day shift ...) (P5), Do you have enough training and / or preparation to face virtual or blended teaching? (P6), In case of having to face virtual classes, do you have enough resources to face this non-face-to-face teaching? (Q7), Do you think the human resources are sufficient to face this return to school? (P8). All of them had three answer options: Yes, No and Don’t know / no answer. Derived from item P3, it was specifically asked Which of the following options worries you the most? (P4): (a) Contaminate my students; (b) Being infected by my students; (c) I am equally concerned about infecting or being infected when interacting with my students. If they answered negatively to item P8, they were specifically asked to select the category or categories of insufficient human resources (P9): (a) Teaching staff; (b) Health personnel; (c) Maintenance and service personnel; and / or (d) Administrative staff.

Two items were also included whose response options were those of a five-point Lickert scale (in which zero was the most unfavourable option and four the most favourable option): According to your point of view, indicate on this scale the level of security perceived upon returning to class (P10) and According to your point of view, indicate on this scale if the return to school seems correct (P11).

Finally, the following socio-demographic variables were included: gender; age; years of dedication to teaching; the area of knowledge of their university education (art and humanities, social and legal sciences, health sciences, science or engineering and architecture); the academic level in which they carried out their teaching activity (Infant, Primary, or Secondary Education, Baccalaureate, Professional Training, University Degree or others such as Official Language Schools, Superior Conservatories of Music or Schools of Performing Arts); and the ownership of the work centre (public, concerted or private).

Statistic analysis

For the analysis of the results, the sample was divided into four age groups: under 30 years (n = 128), participants between 31 and 40 years (n = 247), participants between 41 and 50 years (n = 324) and participants older than 51 years (n = 269). Similarly, the sample was divided into four groups according to their teaching experience (Table 1): participants with less than 10 years of teaching work (n = 371), with between 11 and 20 years (n = 311), with between 21 to 30 years (n = 200) and with more than 31 years of professional experience (n = 86).

Descriptive measures were used to characterize the sample, such as: frequencies, percentages, mean and standard deviation. Also, the chi-square test of proportions and Cramer’s V statistic were used to verify the equality of proportions of the groups in the database according to the different age groups, teaching experience and ownership of the center.

The t tests and Cohen’s d statistic were used to determine the differences by gender, and the ANOVA test and the eta-square statistic were used to analyse the questions of the instrument with a continuous response option (from 0 to 5), according to the definition indicated in the sample groups.

All statistical analyzes were performed with Stata version 12 (StataCorp., United States) and statistical significance was always established at a value of p <0.05.

Results

The characteristics of the sample are presented in Table 1. The analysed sample consisted of 965 teachers, 74.3% female and 25.7% male. Among them, significant differences were obtained in their age (p <0.001; d = -0.31) and in the years of teaching experience (p <0.01; d = -0.19).

The differentiated analysis of the sample according to gender obtained significant differences in the area of knowledge of the participant’s university career (X2 = 56.812; p = 0.001; V = 0.24), with the majority of women being students of social and legal sciences; and in the level of education in which they carried out their teaching activity (X2 = 32.979; p = 0.001; V = 0.18), with the first levels of education (Infant Education, Primary Education and Secondary Education) being the majority in women.

Table 1: Descriptive statistics of the samp 

Note: Result of the t-test between genders *p < 0.01; **p < 0.001.

Result of the X2 analysis between genders: #p = 0.001.

Prepared by authors.

The results obtained in the questionnaire used are shown in Table 2. Significant differences were identified according to gender in the questions: P2 (X2 = 6.294; p = 0.04; V = 0.08), P6 (X2 = 11.141; p = 0.004; V = 0.11) y P8 (X2 = 6.111; p = 0.04; V = 0.08). In the three items, the proportion of affirmative responses was significantly higher in the male gender.

Table 2: Results obtained in the questionnaire used (data: frequency (percentage)) 

Note: * Result of the X2 between sexes: *p = 0.01.

Prepared by authors.

The responses obtained in item P4 are represented in Figure 1, without having identified differences between both sexes in it. The differentiated analysis according to age groups identified significant differences in the P1 items (X² = 15.08; p = 0.02; V = 0.09), P2 (X² = 14.748; p = 0.02; V = 0.1), P3 (X² = 15.818; p = 0.01; V = 0.1) y P5 (X² = 21.481; p = 0.002; V = 0.11). In all of them, the proportion of affirmative responses was significantly higher in the older age groups and the proportion of negative responses in the younger age groups. In item P4 the results showed that among the youngest teachers the fear of contagion was significantly higher and among the older age groups, being infected: infecting their students was the fundamental concern in 10.9% of the teachers under 30 years and 2.6% of teachers older than 51 years. On the contrary, being infected was the main concern of 9.4% of teachers under 30 years of age and 28.2% of teachers over 51 years of age.

Note: Prepared by authors. Values in percentages.

Figure 1: Teachers’ contagion concerns 

However, the differentiated analysis of the participants according to their experience as a teacher identified significant differences in items P2 (X² = 14.41; p = 0.03; V = 0.09), P3 (X² = 13.238; p = 0.03; V = 0.08), P8 (X² = 14.225; p = 0.03; V = 0.09) y P9 (X² = 14.129; p = 0.03; V = 0.09). In all of them, an increase in affirmative responses was observed as the teaching experience of the participants progressed (and the consequent reduction in negative responses).

The contrast analysis between subgroups by ownership of the centre showed statistically significant results in items . In item P3, the proportion of affirmative answers was significantly higher in teachers who carried out their work in private centres (45.5% of affirmative answers in this group in contrast to 24.7% of teachers in public centres). In item P4, it was observed that the concern of being infected was significantly higher in public education teachers (87.3% of this subgroup of participants selected this response option). On the contrary, in private school teachers, the majority answer option in this item was that the concern was the same about infecting and being infected (67.6%) compared to 23.6% who were more concerned about infecting their students and 8.8% who stated that they were more concerned about being infected.

The human resource needs expressed by the respondents identified that more than 66% of the participants recognize the need for more teaching, health and service and maintenance personnel (Figure 2). These needs were significantly different among workers in educational centres of different ownership in relation to the need for more teachers (X2 = 11.853; p = 0.003; V = 0.11).

Note: Prepared by authors. Values in percentages.

Figure 2: Teachers’ concerns based on the type of centre 

In relation to the responses obtained in items P10 and P11, the perceived level of security was, on average, 1.5 ± 1.1 points and the return to the classroom in September was perceived as correct, on average, with a score of 1.9 ± 1.3 points. In neither of the two items were significant differences identified in the responses obtained according to gender (p> 0.05) but there were in relation to the ownership of the centre in Q10: in which it was found that the perceived level of security was significantly lower in public centres compared to concerted schools (p = 0.001) and private ones (p = 0.03), but it was not different between the latter two. The correlation analysis identified significant relationships with the age of the participants in item P10 (r = 0.8; p = 0.002) and P11 (r = 0.7; p = 0.04). On the contrary, years of experience as a teacher were not correlated with any of these items (p> 0.05).