Article Text
PDF
Abstract
Objectives This paper investigated sickness presence (SP) among students in academic upper secondary schools (USS). The research question asked was: To what extent do the prevalence of SP and reasons for SP vary between school classes in USS in Norway?
Design A cross-sectional survey was done in the final year of USS. Multilevel modelling was used to estimate school class-level effects.
Setting 71 school classes from 13 Norwegian schools participated in the study in April 2018.
Participants 1955 students in academic education programmes were invited to participate in the survey, and 1511 completed it. The response rate was 77%. 58% of the respondents were female students.
Outcome measures The outcome measures were the number of SP incidents and reasons for practising SP in school. The analysis examined factors that influenced the prevalence of and reasons for SP.
Results 54% of the students reported two or more incidents of SP. The primary reason for reporting SP was attendance pressure, including factors such as not wanting to exceed the absence limit (90%), concerns that high school absences could negatively affect grades (65%) or reduce chances of getting a job (61%). Some students reported positive reasons for SP, such as maintaining social connections (32%). Multilevel modelling indicated that the class-level effect on the prevalence of SP incidents and positive attendance reasons accounted for 5% of the variance. Some of the class-level effect was due to contextual factors (eg, student relations) and less was due to compositional factors (eg, gender distribution).
Conclusions The study discovered significant variations in SP between school classes, suggesting that students’ attitudes and behaviours towards SP can influence their peers. Moreover, attendance pressure factors for SP were more frequently reported than positive attendance factors.
- PUBLIC HEALTH
- Schools
- Risk Factors
- Surveys and Questionnaires
Data availability statement
Data are available on reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
STRENGTHS AND LIMITATIONS OF THIS STUDY
The study had a reasonably large sample (1511 students) and an acceptable response rate (77%).
71 school classes participated, enabling a multilevel approach to analysing sickness presence (SP).
The list of SP reasons was not exhaustive and did not cover all possible factors.
Recall bias could affect how accurately students report instances of SP and reasons for SP.
The sample was limited to schools in a specific geographical area, potentially leading to findings that do not reflect the broader Norwegian population.
Introduction
The study of sickness presence (SP) has become increasingly important in recent decades as it pertains to a large number of employees. SP is often defined as going to work despite being ill,1–4 and its prevalence has been studied in various occupational groups (eg, factory workers, health personnel, academics, teachers)5–8 as well as in national samples.9 10 SP has attracted attention due to its negative effects at the individual, organisational and societal levels.6–11 However, some studies have also found positive outcomes of SP, such as aiding recovery to full health by keeping individuals active and helping them maintain social connections.4 10
The motivations behind the decision to work while ill have been studied previously. The majority of research on this topic has viewed SP as a negative phenomenon driven by demands and pressures to attend work.3 6 7 11 Some studies have found that SP can be driven by positive reasons, such as maintaining social connections and enjoying one’s job.4 6 10
The topic of this article is SP among students in upper secondary school (USS) in Norway. There is limited research on SP among students in secondary schools and universities.12–16 One rationale for studying SP among students is that high SP can lead to negative health outcomes, such as an increased risk of future illness and decreased physical and mental well-being.9–13 17 Additionally, students who engage in SP may be at risk of spreading illness to their classmates.13 14 A second rationale is that SP can reduce the capacity for schoolwork. Prior studies of employees have found that SP can reduce work performance and capacity.6 7 18 19 A third rationale for investigating SP is that attitudes towards absence and presence may carry over from school to the workplace.14 15 20 Fourth, studying SP in Norwegian USS is particularly interesting because of the absence limit introduced in 2016. The absence limit states that students lose the right to a final grade in a subject if the absence rate exceeds 10%. This attendance pressure may influence SP among students.21
A final rationale to study SP in USS is that differences in SP between school classes are an unexplored area. Prior studies have indicated substantial class-level effects when explaining students’ outcomes in mental health, well-being and academic performance.22–24 Variations between classes in terms of SP may be expected due to differences in class composition or context. A compositional explanation examines differences in the student population (individual characteristics and backgrounds), whereas a contextual explanation points to shared organisational, cultural, social and physical factors within the class.22
The present study examines the prevalence of SP among students in USS. The study also examines the occurrence of eleven reasons for attending school while ill, including attendance pressure and positive attendance.25–27 Finally, the study examines whether the prevalence of SP and reasons for SP vary between school classes. Students are nested within classes, and multilevel modelling makes it possible to separate the class effects from individual circumstances.23 28 29
The research question asked is: To what extent do the prevalence of SP and reasons for SP vary between school classes in USS in Norway?
Methods
Participants and procedures
The cross-sectional data on SP used in this paper were part of a larger research project to investigate and gain insights into academic, social and relational aspects of school life among adolescents. The project, titled ‘Life in schools’, was conducted by researchers from the department of education and lifelong learning at the Norwegian University of Science and Technology (NTNU). The author was part of the research team. The project took place from 2015 to 2018 and included four student surveys, as well as group interviews with students. The research covered various topics, such as social life in school, social support, mental health, academic achievements, motivation and dropout rates. Questions about the prevalence and motives for SP were only included in the survey in 2018.
The survey with questions about SP was collected in April 2018 in Trøndelag county, Norway. The student survey included 183 questions on a variety of topics, and this paper was just one finding from the broader survey. At the time of the study, there were 22 USS in Trøndelag county, and 13 of these schools took part in the survey. The survey was administered to eight city schools and five schools from small municipalities. The sample can be described as a convenience sample. All students in the final year of academic studies at the 13 participating schools had the opportunity to take part in the survey. A total of 1955 students were invited to participate in the survey, and 1511 responded, yielding a response rate of 77%. The participants were 18–19 years old.
Convenience sampling often leads to selection bias, as the sample is not randomly chosen. In this survey, the students were from schools located within a 2-hour drive from NTNU. As the participants were from Trøndelag county, their characteristics may not fully reflect those of the broader Norwegian student population, and the generalisability of the findings to other regions with different conditions may be limited. Nevertheless, Trøndelag county can be considered a suitable setting for a Norwegian student survey, as the county includes a diverse mix of urban and rural schools, as well as small and large schools. Additionally, the sample was representative in terms of gender, ethnic background and academic achievement, closely mirroring the proportions observed in the national student population.
The schools and participating students were recruited in the fall of 2017. A pilot study was done among students in USS. The students completed the questionnaires in writing at the school, and it took about 30–40 min to fill out the questionnaire. An information letter about the research project was provided before the students completed the survey at school. The letter presented detailed information about the study’s purpose and procedures. The students were assured that their participation was voluntary and that they could withdraw at any time without any consequences. Students who decided not to answer the survey did other schoolwork.
To minimise human error in data entry for our paper-based survey, we used closed-ended questions with predefined answer choices, we used optical character recognition technology to scan and extract data from physical forms, and the researchers reviewed and verified the scanned data.
Measures
Prevalence of SP was measured by the question: ‘So far this school year, did you go to school despite feeling so ill that you should have stayed home sick?’ The students were given these alternatives: ‘no’; ‘yes, one time’; ‘yes, two times’; ‘yes, three or four times’; ‘yes, five times or more’. SP was defined as attendance of this sort two or more during the previous school year, which corresponds with the cut-off criteria used in previous research.1 4 14 The students who reported SP two or more times responded to a series of eleven statements regarding the reasons they attended school while ill. Of 818 students that reported two or more incidents, 15 did not report on reasons for SP. Thus, the analysis of reasons for SP included 803 students.
The statements regarding reasons for SP were based on a similar study conducted with secondary school students in five European countries. In this European study, statements had been developed after a literature review of reasons for SP among adult workers, combined with group interviews with students to understand attendance patterns in secondary schools.14 Before conducting the new Norwegian survey, a pilot study was conducted with 10 students in USS to identify problems or limitations with the questionnaire and make necessary adjustments. The students in the pilot study received a draft of the questionnaire, and then they met for a group interview. Two important adjustments were made after the pilot study. First, to take the Norwegian context into account, the ‘absence limit’ was added to a statement about attendance rules. In addition, some students commented on the importance of parents’ concern about attendance, and a statement about parental concern was included in the questionnaire.
The question regarding reasons for SP was: ‘Please agree or disagree with the following statements: I went to school even though I was ill because…’:
I enjoy going to school.
I have great interest in what we learn at school.
Going to school was beneficial for my health.
I want to maintain my social network.
High school absence might negatively affect my chances to get a job.
High school absence might negatively affect my grades.
I do not want to exceed the absence limit in one or more subjects.
My pride depends on not being sick from school.
Crucial material/syllabus is explained at school.
My parents are concerned that I go to school.
I do not want to burden my classmates (eg, group work, lab work, etc).
The survey used a six-point Likert scale to measure responses; 1=disagree strongly, 2=disagree, 3=somewhat disagree, 4=somewhat agree, 5=agree and 6=agree strongly.
Statistical analyses
The data were analysed using IBM SPSS (V.27). Missing data was minimal. Still, on the scale variables, missing values were replaced with the mean of the observed values. Results were not weighted.
A principal component analysis (PCA)28 clustered the items into two components. The extraction was based on Eigenvalue, and Kaiser’s criterion of one was used to confirm two components. A scale for positive attendance reasons was constructed by computing four items into one variable and then dividing by four to get the mean score for each respondent (statements 1–4). A scale for attendance pressure was constructed by computing three items into one variable (statements 5–7) and then dividing by three to get the mean score for each respondent. Internal consistency was measured using Cronbach’s alpha, which was 0.84 for positive reasons and 0.70 for attendance pressure. Four items (statements 8–11) were excluded from the analysis because they loaded on both components or lowered the alpha values.
Multilevel models allow researchers to separate contextual effects from intraindividual effects and are a preferred technique for analyse data with a hierarchical structure.23 28 29 The intraclass correlation coefficient (ICC) describes how strongly units in the same group resemble each other, and when multiplied by 100, the ICC gives the percentage of variance attributed to each level. In previous studies, class-level effects varied according to the study group and the type of outcome investigated.22–24 28 29 The variance between classes is analysed both with and without adjustment for student demographics.
An important issue in multilevel models is what constitutes a sufficient higher-level sample size, but there is general agreement among statisticians and researchers that 50 or more units are acceptable in most situations.23 29 The students in this study were from 71 school classes and 13 schools. It was decided to account only for the school class level to meet the generally accepted threshold of 50 or more units, and this would ensure reliable and stable estimates in the multilevel models. It was decided not to separate schools as an analytical level, since estimates for the school level would be less accurate and unreliable due to the sample size of schools.
Four independent variables were included in the multilevel analyses.
Gender: divided between male students and female students. Prior studies indicated that female students more often reported SP and extrinsic motivation for SP.13–15
Mothers’ education: divided between students who have a mother with low educational attainment USS and students who have a mother with high educational attainment (bachelor’s degree or higher). In a previous study, students with highly educated parent(s) reported intrinsic motivation for SP more often.14
Age: divided between students 18–19 years of age and students from 20 to 29 years of age. Prior studies in USS indicated different results with regard to connections between age and SP.13 14
School absence: divided between students with no/low absence (4 days or less), moderate absence (5–9 days) and high absence from school (10 days or more). Prior studies in USS indicated a strong correlation between school absence and SP.13
Patient and public involvement
The students played an important role in the pilot study prior to the full-scale study. First, the questionnaire and outcome measures were adjusted based on insights from the group interviews with students. Second, students spend significant amounts of time together in the class setting, where they are exposed to each other’s attitudes and behaviours. In the group interviews, some of the students reflected on the relationship between social dynamics, the learning environment and attitudes towards school absence and presence. These reflections inspired the research question addressed in the article about variations in SP between school classes. The plan is to share the results through social media and presentations to reach a wider audience, particularly teachers and students in USS.
Results
Table 1 summarises the characteristics of the participating students. The sample comprised 42% boys and 58% girls. 45% of the sample had a mother with a bachelor’s degree or higher. 91% of the sample were 18 or 19 years of age. 51% of the sample had no/low absence, 29% had moderate absence and 20% had high absence from school.
Overview of the characteristics of the respondents (n=1511)
Table 2 displays information about the prevalence of SP incidents. An SP incident can last from a single day to multiple days, but the survey did not ask about the exact duration. 54% of the students responded that they had gone to school on two or more incidents even though it would have been reasonable to stay at home sick. 16% reported high SP (five or more incidents of SP), while 38% reported some SP (two to four incidents of SP).
Distribution of SP incidents in academic upper secondary schools, percentage of students and 95% CI around the percentage (n=1511)
Table 3 presents calculations of means for various statements about SP. 90% of the students reported SP because they did not want to cross the absence limit (mean value: 5.5). 50–65% of the students reported SP because high school absence could affect grades or chances of getting a job, that crucial material was explained at school, or parental concerns (mean values: 4.2–4.7). About 30% of the students reported SP to maintain their social network, not burden classmates, or that they were proud not to be absent (mean values: 3.2–3.4). Finally, less than 20% reported SP because they enjoyed going to school, had an interest in what they learnt, or that SP was beneficial for their health (mean values: 2.6–2.9). For most students, multiple reasons for SP were important.
Distribution of students that agree or disagree with various motives for sickness presence in upper secondary school
A PCA with quartimax rotation examined whether items clustered together, and table 4 shows the final rotated analysis with seven items. Kaiser-Meyer-Olkin values of 0.73 verified the sampling adequacy for the analysis, which is above the acceptable limit of 0.6. Bartlett’s test of sphericity indicated that correlations between items were sufficiently large for PCA.28 The items that cluster together suggest that Component 1 represents positive attendance reasons, such as well-being and health (alpha=0.84). Component 2 represents various forms of attendance pressure, including concerns about poor grades, decreased job chances and the need to uphold attendance obligations (alpha=0.70).
Rotated matrix from principal component analysis with quartimax (n=803)
Table 5 presents the results from the multilevel models about prevalence and reasons for SP. Model I shows that most of the variance in SP incidents was accounted for by individual level factors (95%), though there was also a statistically significant school class-level effect (5%). The inclusion of individual factors reduced the class-level effect (4%). Female students reported more SP incidents compared with male students (p<0.01). Students with moderate and high school absence reported more SP compared with students with low or no school absence (p<0.01). Model II indicates a significant school class-level effect on positive attendance reasons, which remained unchanged after the inclusion of student demographics. Male students reported more positive attendance reasons for SP compared with female students (p<0.05). Model III shows a minimal class-level effect on attendance pressure. Students with high school absence reported attendance pressure factors for SP more often than students with low or no school absence (p<0.05).
Results from multilevel modelling predicting SP incidents, positive attendance reasons and attendance pressure in upper secondary school
Discussion
The findings of this study point to the importance of the school class with regard to SP incidents and positive attendance reasons. The multilevel approach is new to studies of SP in schools, but the impact of class-level factors on SP aligns with previous multilevel research on topics like mental health, behaviour, attitudes and academic performance.22–24 28 The consistency of the results supports the conclusion that the class level has an impact on the prevalence of SP and positive attendance reasons for SP. However, there was a minimal class-level effect on attendance pressure.
The analyses of SP incidents indicated that a small part of the class-level effect was explained by the composition of the student group. Classes with a high proportion of female students and students with high absence often have higher SP than classes with fewer girls and fewer students with high absence. None of the observed class-level effects on positive attendance reasons for SP could be attributed to the composition of the student group. The contextual explanation of class-level effects refers to shared experiences, and that students in a class share courses, teaching staff and belong to the same student group and learning environment.22 24 In the group interviews (pilot study), some of the students reflected that those in a supportive class environment with a high level of well-being are more likely to have positive motivations for SP, such as enjoyment or a desire to maintain social connections. Moreover, they noted that some students’ attitudes towards absence and presence could significantly influence their classmates. For example, if one student has a positive attitude towards attending school and values the importance of being present even when ill, this can encourage others in the same class to adopt a similar attitude. Previous studies on emotional and social contagion have demonstrated that various behaviours and preferences, such as smoking habits, music and movie tastes, emotions such as happiness and loneliness, and attitudes, can cluster and spread within a group.30 31 This underscores the role of social connections in shaping individual behaviour and attitudes, suggesting that positive attendance reasons and SP practices may also be contagious within a class, spreading from one student to another.
Students in USS in Norway experience a variety of ailments, including mental health issues, headaches, abdominal pain, common cold and influenza, allergies, sleep problems and many others.13 32 These ailments cause discomfort or pain, but it is still quite common for Norwegian students to attend school despite feeling unwell. As regards the prevalence of SP, more than half of the students in academic studies reported two or more incidents of SP in a school year. This finding was in accordance with the previous studies of SP among secondary school students and university students.13–15 In correspondence with previous studies at secondary schools, female students reported SP more often than male students.13 14 There was also a significant positive association between sickness absence and SP.13–15
9 out of 10 students who reported SP did so because they did not want to exceed the 10% absence limit from school.21 The absence limit was implemented to make students attend school more often, both for the sake of their own education, for their peers and the school’s learning environment. A breach of the absence limit has serious consequences, and a student will not get a certificate for completing USS if she breaches the absence limit in one or more subjects, and without this certificate, she cannot apply for higher education. This study indicates that an unintended consequence of the absence limit seems to be more students practising SP.
A majority of the students who reported SP were concerned about the negative impact that high absence rates could have on their grades and future job prospects. The reasons for avoiding negative consequences from exceeding the absence limit, protecting academic performance and safeguarding future job opportunities, were grouped under the component ‘attendance pressure’. Previous research from working life showed that SP from attendance pressure could have a negative impact on employee health.1–3 10 25–27 There was a tendency that students with high school absence reported attendance pressure somewhat more often than students with low absence. A probable explanation could be that students with a high absence would opt to attend school when sick because more absence could affect grades negatively or violate attendance requirements.
Rather few students reported SP because of health benefits or their social connections, or because they had an interest in school or enjoyed school. The prior study in USS found that few students reported positive attendance reasons for SP.14 Work satisfaction and work enjoyment were quite often reported reasons for SP in the adult population.10 25 26 The tendency that male students reported positive attendance reasons somewhat more often than female students could be an indication of slight differences in attitudes between female and male students. This gender difference was in line with the previous study in secondary schools in five other countries.14
The present study has some strengths, such as a large sample, high response rate and the opportunity to use the multilevel approach to investigate variations between school classes. Having a large sample size and using a multivariate approach enhance the generalisability and rigour of the results. One limitation, however, is that different students may have different criteria for determining when they should stay home or go to school while they are sick. This variability in criteria could affect the number of SP incidents that they reported. Second, recall bias is another common problem in self-reported data, and it might well be that students had difficulty accurately remembering the exact number of SP incidents and their reasons for practising SP. Third, the list of reasons for SP was not exhaustive, and other reasons such as fear of missing out, feeling a sense of obligation, or workload or time constraints could have been included. Fourth, there may be selection bias due to the sample being restricted to schools in Trøndelag county, which could lead to findings that do not fully reflect the national population.
This article provided insights into the prevalence of SP and reasons for SP among students in USS. Addressing SP and absence in Norwegian schools requires a balanced approach. Setting clear attendance policies has encouraged students to attend school more regularly, but such policies must be balanced, and it is important not to pressure students to attend when they are genuinely ill. SP can have negative consequences such as prolonging recovery time and spreading illness to other students and teachers, and schools and parents should support students in making the right decision to stay home when necessary and provide resources to help them catch up on missed work.
The discovery of significant variations in the prevalence of SP between school classes was an important finding. Such variations can highlight differences in class environments, teacher expectations and peer influences. Still, there is much more to learn about SP in secondary schools. Further research with a quantitative design can help to shed light on the differences in the prevalence of SP and reasons for SP between school classes and student groups. Qualitative methods, such as focus groups, interviews and ethnographic observations, could provide a deeper understanding of the experiences, perspectives and motivations of students as regards SP. Moreover, longitudinal studies are necessary to establish causal relationships between SP and future health and school performance. Studies over time could determine whether SP based on positive attendance reasons has a positive impact on school performance and future health, or whether SP driven by attendance pressure has a negative impact on students’ health and academic outcomes.
Data availability statement
Data are available on reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by Norwegian Centre for Research Data, ID no: 42443. This approval signifies that the research complied with the relevant ethical and legal guidelines, ensuring the protection of participants’ rights and privacy. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
I would like to thank Professor Per Frostad and Associate Professor Per Egil Mjaavatn for giving me the opportunity to take part in the ‘Life in schools’ research project. The research project explored several topics, and questions about sickness presence were included in the student survey in 2018, generating the data used in the present analysis. I would also like to thank all the students who participated in this project.
References
Footnotes
Contributors VJ is the sole author of the manuscript. He contributed to the planning, conduct and reporting of the work described in the article, and he is the guarantor.
Funding The research project was self-funded by the Norwegian University of Science and Technology.
Competing interests None declared.
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; externally peer reviewed.