Policies and opportunities for physical activity engagement in Austrian schools: a census survey

Ethics and study design

The Carinthian Schools’ Physical Education Report (CSPER) comprised a structured survey administered to all principals of schools in the Austrian state Carinthia. The study, which followed the guidelines for Good Practice in the Conduct and Reporting of Survey Research (Kelley et al., 2003), was reviewed and approved by the Ethics Council of the University of Klagenfurt (ref.no. 2022-075). All participants provided implied and informed consent in the online survey.

Sample

To evaluate the availability of infrastructural PA opportunities and the schools’ policies with regard to PA, we applied a census sampling method, inviting all 342 principals heading primary schools, middle schools, high schools and vocational schools in Carinthia. Schooling in Austria is mandatory for 9 years, from age 6 to 15 years (first to ninth grade). The first 4 years of education are completed in primary school at age 6 to 10 years. From age 10, children may attend a middle school or a high school, comprising four educational levels. After this period (4 years), children may a) stay in high school until 18 years with the aim of achieving the general school-leaving examination or b) choose a higher vocational school with a higher qualification in profession or occupation and school-leaving examination at the age of 19.

In collaboration with the local school authority, the principals were emailed with a link to the online survey on 16th January 2023. To increase response rate, a reminder email was sent 2 weeks later, on 3rd February 2023 (Sammut, Griscti & Norman, 2021).

Instrument

The CSPER questionnaire was generated using a multi-stage expert consensus process. Initially, all authors gathered ideas and possible survey items which were then grouped, edited and reduced to produce a first version of the instrument. The resulting draft was then sent to the state’s department of education for further external feedback. After revising and adapting the questionnaire, it was passed on to three members of the target population (one principal of a primary, middle, and high school each) for face validation. All reported high comprehensibility and clarity but provided some minor suggestions for improvement.

The final instrument comprised a combination of single- and multiple-choice questions relating to the activity friendliness of schools. The first part addressed basic sociodemographic data such as the school type, environment of the school (urban/rural) or the number of students. The second part of the CSPER questionnaire had two main focuses. The infrastructure section assessed the availability of a school yard, a gymnastic and swimming hall as well as the availability of other facilities for PA promotion (e.g. meadow areas, sports fields, playgrounds, ping pong tables, basketball hoops, beach volleyball courts, climbing walls, skate parks and gyms). The PA policy section asked questions on break duration, the educational level of PE teachers, cancellation of and student participation in PE classes. In addition, we assessed the schools’ involvement in PA promotion projects, extracurricular PA programs, sports events (e.g., school championships, sports badges), cooperations with local sports clubs, and performance of fitness tests. The questionnaire was concluded by Likert-scaled questions used to capture satisfaction with school infrastructure (very dissatisfied = 1, rather dissatisfied = 2, rather satisfied = 3, very satisfied = 4) and the school’s activity friendliness (strongly disagree = 1, tend to disagree = 2, agree somewhat = 3, agree completely = 4).

Statistical analysis

The normal distribution of data was tested by means of the Shapiro–Wilk test. Descriptive statistics were then applied to analyze the obtained data (means/medians) as appropriate. Interval/quasi-interval scaled data (e.g. number of students, duration of the longest break) and dichotomous data (e.g., PA facilities, additional sport activities) were presented as absolute (n) and relative (%) values, respectively.

In addition to descriptive analyses, we examined differences between school types for the mean number of PA facilities (ANOVA), availability of individual PA facilities, use of specialized PE teachers, participation in PE lessons, cancellation of PE classes, PA policies, rated activity friendliness (all chi squared test), school environment (urban/rural, Mann-Whitney-U test) and extracurricular PA programs (yes/no, Mann-Whitney-U-test). Pearson and Spearmen correlations were performed to reveal potential associations between the number of PA facilities and the number of students, break length, as well as satisfaction with PA facilities and school’s activity friendliness. The significance level was set to α = 0.05. Data analyses were performed using JAMOVI, version 2.3 (https://www.jamovi.org).

Sample characteristics

A total of 130 principals completed the survey which corresponds to a minimal response rate of 38%. The majority of the respondents (54%, n = 71) were employed at primary schools, while the remainder directed middle schools (21.5%, n = 28), high schools (10.8%, n = 14) and vocational high schools (13.1%, n = 17). About one third (37.7%, n = 49) of the schools were located in urban areas; 62.3% (n = 81) were from rural areas. School size ranged from 11 (primary school) to 1,294 (high school) students (median: 132).

Infrastructure and PA facilities

School yards (at least 4 m² per student) were reported to be available by 75.6% (n = 96) of the principals. Almost all schools (87.4%, n = 111) had a gymnastics hall (median size: 235 m²), but only one third (37%, n = 47) had access to a swimming hall.

The mean number of PA facilities per school (excluding school yard and gymnastic/swimming hall) was 4.2 ± 2 (min: 0, max: 9). Significant differences were observed between school types (p < 0.05) as primary schools had less facilities (3.5 ± 1.7) than others (high schools: 5.1 ± 2, middle schools: 4.9 ± 1.9, vocational high schools: 4.5 ± 2.3).

The most frequent PA facilities were meadow areas, sports fields and playgrounds. In contrast, gyms and skate parks were least prevalent (Table 1). Similar to the total number, the individually available facilities differed between school types (Table 2): Meadow areas were most prevalent in primary and middle schools (p < 0.05); the same applied to playgrounds (p < 0.05). In contrast, beach volleyball courts were more often found in high schools and vocational high schools (p < 0.05). Athletics facilities were frequent in all types of schools, except for primary schools (p < 0.05, Table 1).

No difference in the number of PA facilities was found for the school environment (urban vs. rural, p > 0.05). However, correlation analyses revealed a significant positive association between the total number of PA facilities and the number of students (p < 0.001, r = 0.22, Table 2).

Two thirds of the principals (66.1%) were satisfied with the PA facilities of their school but satisfaction did not correlate with the number of facilities (p = 0.65). Likewise, no association was found between the number of PA facilities and satisfaction with the general activity friendliness of schools (p > 0.05, Table 2).

PA policy

The median duration of the longest break was 15 min (interquartile range/IQR: 5). High schools reported the lowest duration (10 min, IQR = 5) while vocational high schools had the longest breaks (30 min, IQR = 25). No association was found between the number of PA facilities and break length (p > 0.05).

With regard to the PE teachers’ educational level, three quarters of the schools reported appointment of qualified teachers only, while almost one quarter had classes taught by both partly-qualified and non-qualified (no academic degree in PE) personnel (Table 3). Middle schools demonstrated the lowest percentages of PE classes taught by qualified PE teachers (42.8%), while high schools had the highest percentages (100%, p < 0.05).

In most schools, PE classes were cancelled less than 20% of the time due to other events or timetable shifts, but cancellations were highest in vocational high schools (p < 0.05). More than four in five schools had at least 80% of students participating actively in PE lessons. Vocational high schools showed the lowest participation rate (52.9%, p < 0.05). Lower participation rates were associated with higher PA facility counts (p = 0.02, rho = −0.19).

With regard to PA promotion strategies, about half of the schools (45.2 %) reported being part of a PA project and likewise, more than half (54%) of the respondents stated offering extracurricular PA programs and courses (Table 4). The majority of the schools confirmed both participation in sports events (66.9%) and cooperation with a local sport club (55.6%). One fifth of the schools reported regular performance of fitness tests.

PA promotion strategies were significantly different between school types except for the performance of fitness tests (p = 0.49). While primary schools were most frequently part of a PA promotion project and collaborated most frequently with local sports clubs (p < 0.05), high schools had the highest participation in sports events and the highest number of extracurricular PA programs (p < 0.05, Table 4). Schools offering more extracurricular PA opportunities had higher PA facilities (u = 0.36, p < 0.001).

Almost all principals (94.4%) would welcome professional support for PA promotion at their school and this wish was not different between school types (p = 0.28). More than 80% of the principals agreed their schools were activity-friendly (Table 4) and complete agreement with this statement was highest in primary schools (p < 0.05).

Limitations

Some methodological aspects merit discussion. A particular strength of our study is that we invited all principals of Carinthia to participate in the survey. The resulting response rate of 38%, at first glance, is at best satisfactory (Wu, Zhao & Fils-Aime, 2022). Yet, it needs to be considered that it represents a conservative estimate. First, although the questionnaire was distributed by the education directorate, some dead email addresses may have been included in the sending list. Second and more importantly, not all invited participants may have seen, opened and read the emails, particularly because we did not use a reading confirmation. The distribution of school types and school sizes in our sample corresponded strongly to the distribution of all schools in Carinthia. However, although our participants seem to reflect the true population and despite the the likely higher true response rate, sample representativeness of the entire Carinthia principal population cannot be safely assumed. Nor can this sample be suggested to represent principals in other states in Austria or in other countries.

Another issue relates to the source of the information. We decided to approach school principals as the leaders who would have the most “whole school” perspective. PE teachers may have more granular-level knowledge about what happens in their classes and elsewhere with PA but this could also have skewed their responses about the school facilities and policies (positively or negatively). We also acknowledge that in some cases, principals may not have had sufficient insight into the micro-level PA promotion (e.g. in the PE classes). Therefore, future studies may conduct similar surveys with other stakeholders such as teachers, students, and parents.

Finally, it may have been intriguing to assess more background variables such as the socioeconomic background of the community in which schools are located or the social status of the families. As it has been shown that social inequality is related to PA behavior, it is critical to know which social mechanisms enable or prevent the participation of children and adolescents in PA (Andersen & Bakken, 2019; Rittsteiger et al., 2021; Wijtzes et al., 2014).