European Journal of Public Health, Vol. 19, No.

5, 554560
The Author 2009. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.
doi:10.1093/eurpub/ckp032 Advance Access published on 25 March 2009
................................................................................................
Staff reports of psychosocial climate at
school and adolescents health, truancy
and health education in Finland
Marianna Virtanen1, Mika Kivimaki1,2, Pauliina Luopa3, Jussi Vahtera1,
Marko Elovainio3, Jukka Jokela4, Minna Pietikainen3

Background: Psychosocial climate at school reported by school personnel may be related to pupils
reports of being heard at school, depression, physical and psychological symptoms, truancy and received
health education. Methods: Cross-sectional study combining data from two independent surveys
conducted between 2004 and 2005 in 136 public schools in Finland. The pupil sample comprised 11 583
boys and 12 706 girls from the eighth and ninth grades of lower secondary school and first and second
grades of upper secondary school. A personnel survey (n = 1946) was used to measure psychosocial
climate indexed by trust and opportunity for participation, support for innovation, orientation towards
high-quality work and accepted and clear goals. Results: After adjustment for individual and school-
level covariates, multilevel analyses revealed odds for pupils opinion of not being heard were higher in
schools where personnel reported poor trust and opportunity for participation (OR = 1.33) and low

Downloaded from eurpub.oxfordjournals.org by guest on April 18, 2011

support for innovation (OR = 1.37). Poor trust and opportunity for participation and unclear work goals
among staff were associated with high truancy among pupils (ORs = 1.54 and 1.39). Poor trust and
opportunity for participation among staff were also related to pupils depression [cumulative odds ratio
(COR = 1.14)], and physical and psychological symptoms (COR = 1.17). Unclear goals among staff were
associated with pupils opinions that health education was insufficient (OR = 1.40). Conclusions: school
climate is associated with adolescents health, wellbeing and received health education.

Keywords: adolescents, depression, health education, school climate, truancy

................................................................................................

Introduction association between the climate seen by school personnel and

ecent research has highlighted psychosocial environments
R as important determinants of positive youth develop-
ment.15 Psychosocial environments have been characterized
by three basic dimensions: connection, regulation and
autonomy across multiple settings, including families, schools
and neighbourhoods. These three aspects have been expanded
to refer to other requirements for healthy adolescent develop-
ment, for example, physical and psychological safety, adequate
and appropriate supervision, supportive relationships and
support for efficacy and opportunities for skill development.3
In school settings, pupils perceptions of positive psychoso-
cial climates have been associated with positive developmental
outcomes such as good mental health, and with a lack of health
risk behaviours, violence and delinquency.2,613 Furthermore,
although school truancy has been shown to be associated with
pupils poor school performance and feelings of unsafety
at school,14 little is known about the effect of psychosocial
climate at school on pupils truancy.14,15 Positive school
climate, as reported by teachers themselves, has been associated
with the teachers perceptions of a successful implementation
of the health promotion programme16 and HIV/AIDS
education17 at school, but there is no evidence about the
1 Finnish Institute of Occupational Health, Topeliuksenkatu 41 a A,
00250 Helsinki, Finland
2 Department of Epidemiology and Public Health, University College
London Medical School, 1-19 Torrington Place, London WC1E
6BT, UK
3 National Institute for Health and Welfare, 00531 Helsinki, Finland
4 Faculty of Sport and Health Sciences, University of Jyvaskyla,
Finland
Correspondence: Marianna Virtanen, Finnish Institute of
Occupational Health, Topeliuksenkatu 41 a A, FIN-00250 Helsinki,
Finland, tel: +358 40 7450106, fax: +358-30-474 2552,
e-mail: marianna.virtanen@ttl.fi
pupils perceptions of health education. However, because
these studies have relied on either pupils or teachers self-
reported perceptions of both the school characteristics and the
measured outcomes, the findings are open to common method
bias in which individual-related factors (for example, response
style, negative affectivity) artificially inflate associations by
affecting both the perception of the school climate and
outcome measures.
In the present study, we eliminated problems related to
common method variance by linking school personnel
evaluations to pupils responses in two independent surveys.
We used a four-factor model of psychosocial climate at school
and hypothesized that pupils health and wellbeing are more
affected by those factors that are connected to psychological
dimensions of school climate (trust and opportunity for
participation and support for innovation among staff) whereas
their perceptions of health education at school are more
affected by performance-oriented dimensions (orientation
towards high-quality work and accepted and clear goals
among staff).18 Trust and opportunity for participation refers
to a non-threatening and supportive atmosphere that
encourages involvement and creates feelings of trust. When
trust and opportunity for participation is high, all members
feel sufficiently safe to contribute ideas, share information and
challenge existing ways of working without fear of being
criticized. The support-for-innovation dimension concerns
whether suggested innovations are taken seriously, tried out
and given adequate time and resources, and whether the
members of a work unit work together to actualize new ideas.
Orientation towards high-quality work means that the work
unit members have a shared concern about performance
quality and constantly aim to improve the quality of their
work, and therefore follow systems of evaluation, modification,
control, and critical appraisal. Accepted and clear goals refer to
a vision; a valued outcome shared by all work unit members,

which represents a higher-order goal and motivating force at
work, which is achievable and negotiable. In earlier research,
these dimensions of work unit collaboration have predicted
organizational commitment,19 work-related stress and sickness
absence20 and depression21 among employees. In the present
study, we examine whether the psychosocial climate at school
reported by school staff is associated with various pupil
outcomes, including health, school perceptions, truancy and
health education.
Methods
The data were obtained from two ongoing studies: the
Finnish School Health Promotion Study22,23 focusing on
adolescent health and health behaviours, and the 10-Town
study focusing on health of local government personnel,
including schools.24,25 The nationwide Finnish School Health
Promotion Study is a classroom survey which has been carried
out every second year since 1995. The study covers virtually
all eighth and ninth grades of lower secondary schools (14- to
16-year-old pupils) and the first and second grades of upper
secondary school (16- to 18-year-old pupils). The study has
been approved by the Ethics Committee of Tampere University
Hospital. Either in May 2004 or May 2005 students voluntarily
responded to the survey on their health and lifestyle habits
(n = 25 879, response rate 84%).
In the 10-Town Study, local government personnel of
the participating towns responded to a voluntary-basis
questionnaire between October 2004 and January 2005
(response rate 65%; school personnel response rate 54%).
The study was approved by the Ethics Committee of the
Finnish Institute of Occupational Health. The number of
secondary schools participating both to the Finnish School
Health Promotion Study and the 10-Town study was 150 (83%
of the secondary schools in the 10 towns). Of them, five
schools were excluded because they had less than a required
minimum of four respondents among personnel. Nine more
schools were excluded because they had <30 participating
pupils, resulting in a total of 136 included schools. The total
number of school personnel respondents was 1946 (495 men,
1451 women), the mean number of personnel respondents per
school being 14 (range 445, SD = 6.9). Of them, 1856 (95%)
were teachers (84 head teachers). The remaining 5% were
mainly administrative staff and canteen workers. In the 136
schools, the number of pupil respondents was 25 879, for 94%
(24 289) of whom there were complete demographic data.
The mean number of pupil respondents per school was 190
(range 32531, SD = 76.2).
Psychosocial climate among the school personnel was
measured in the 10-Town study by using the short 14-item
version of The Team Climate Inventory26 developed from
the original 38-item version.27 The inventory measures
four dimensions of team climate: (i) trust and opportunity
for participation (four items, Cronbachs  = 0.90; example of
item: Everyone feels that they are both understood and
accepted); (ii) support for innovation (three items,  = 0.88;
example of item: In our workplace we take the time needed to
develop new ideas); (iii) orientation towards high-quality
work (three items,  = 0.79; example of item: In your work
place, do you pay attention to weaknesses in your work to
improve quality?); (iv) accepted and clear goals (4 items,
 = 0.84; example of item: Do you accept the goals set at your
work place?). The participants evaluated all statements on
a five-point Likert-type scale (1 = totally disagree to 5 = totally
agree). School-level averages were calculated and linked to the
pupil survey.
All pupil variables were obtained from the Finnish School
Health Promotion study. Self-reported depression was
School climate and adolescents well-being 555
measured using a shortened 12-item version of the Beck
Depression Inventory.28,29 The 12 items consist of statements
showing increasing intensity of depressive emotions and
cognitions, each scored on a scale ranging from 0 to 3. A
sum score for self-reported depression was then calculated and
categorized as none (04 points), mild (57 points), moderate
(815 points) and severe depression (1636 points).29
The psychometric properties of the 12-item scale corresponds
with those of the original 13-item scale.30 Frequent physical
and psychological symptoms (neck or shoulder pain, low back
pain, stomach pain, tension or nervousness, irritability or
tantrums, sleeping problems, headache and fatigue) were asked
about separately under the query: During the past six months,
have you had some of the following symptoms and how often?
with response alternatives 1 = never; 2 = about once a month;
3 = about once a week; 4 = almost every day. A dichotomous
variable (1 = never to once a week; 2 = almost every day) was
calculated for each symptom. The scale of frequent physical
and psychological symptoms was formulated as the number
of daily symptoms: 0, 1 or 28 symptoms.
Pupils perceptions of being heard at school were assessed by
Downloaded from eurpub.oxfordjournals.org by guest on April 18, 2011
three statements (Cronbachs  = .59), such as: Teachers
encourage me to express my opinion in the classroom, with
response alternatives 1 = totally agree; 2 = agree; 3 = disagree;
4 = totally disagree. A sum score from dichotomized items
(0 = totally agree or agree; 1 = disagree; 2 = totally disagree) was
calculated, and pupils scoring more than three points were
identified as having an opinion that pupils do not get heard
at school.
Absenteeism was asked about in the following questions:
How many schooldays have you been absent from school
during the past month for the following reasons: (i) sickness,
(ii) truancy, (iii) other reason? For each reason, the response
alternatives were 1 = not at all; 2 = one day; 3 = 23 days; 4
= > 3 days. The overall absence rate was calculated and used as
a logarithmic transformation of the mean of the three items.
High absence due to truancy was dichotomized as follows:
1 = 03 days; 2 = > 3 days.
Sufficiency of health education was asked about in a single
question as follows: According to your opinion, is the
frequency of health-related education at your school 1 = too
much; 2 = appropriate; 3 = too little?
Frequency of classroom conversations or lessons about
health risk behaviours was asked about with the questions
During this school period, have you had classroom conversa-
tions or otherwise been taught about the following subjects:
(i) smoking, snuffing; (ii) alcohol; (iii) drugs; (iv) sex and
relationships between boys and girls? with response alter-
natives 0 = no; 1 = in one class; 2 = in several classes. From this
information, a total score was calculated by summing up the
points in each subject category (range 08 points) and divided
into five groups (1 = 0 points; 2 = 12 points; 3 = 34 points;
4 = 56 points; 5 = 78 points, that is, all subjects been taught
several times during this school period).
The sex and age of the pupils were derived from the survey
responses. Each pupils socioeconomic background was
ascertained from two questions asking separately about the
mothers and fathers education: What is the highest
education your parents have performed? (1 = comprehensive
school/comprehensive school and vocational school; 2 = upper
secondary school/upper secondary school and institute;
3 = college or university). If both the mothers and fathers
education were reported, the highest education level between
them was chosen.
To assess pupil socioeconomic composition at school,31
proportion of pupils with low parental education was
calculated using the measure of individual responses about
the parental education. Pupils with low parental education
556 European Journal of Public Health

were those whose parents education level was vocational
school or less (range between schools varied from 3.7% to
67.9%). The schools were then put into one of three categories
based on this percentage: <20% of pupils with parents with
low levels of education; 2039.9%; 40%).
We used the SAS 9.1 GLIMMIX procedure in the multilevel
analyses to assess the school-level and individual-level effects
on pupil outcomes. For dichotomous outcome variables, we
used multilevel random intercept models in binary logistic
regression models and calculated odds ratios (ORs) and their
95% confidence intervals (CIs). For outcome variables with
three or four categories, we used multilevel random intercept
models with a multinomial logistic regression procedure for
calculating cumulative odds ratios (CORs) and their 95% CIs.
Dimensions of school psychosocial climate were divided into
quartiles to indicate four levels for each scale. We made
adjustments for pupils sex, age, parental education, the
proportion of pupils at school having parents with low levels
of education, and the proportion of non-teacher respondents
among school personnel. Because a pupils own absenteeism
may affect his/her perception of health education received,
adjustment for parental education had little effect on the
results (model 1 results not shown). None of the other staff-
reported dimensions of psychosocial climate at school were
related to pupils health.
Table 4 presents the relationship between psychosocial
climate assessed by the school staff and pupils truancy and
perceptions of being heard. Poor trust and opportunity
for participation among staff were associated with pupils
perception of not being heard (OR = 1.33) and with high
truancy (OR = 1.54). Low support for innovation was also
related to pupils perceptions of not being heard (OR = 1.37).
If the personnel thought that goals were not accepted and clear
at school, pupils were 1.39 times more likely to have high
absence rates due to truancy. As in table 3, adjustment for
parental education had little effect on the results.
Table 5 shows the association between the staffs perceptions
of the psychosocial climate at school and pupils perceptions
of received health education. Poor clarity and acceptance of
goals were related to pupils perceptions of insufficient health
education (COR = 1.40). Adjustment for parental education

Downloaded from eurpub.oxfordjournals.org by guest on April 18, 2011

had little effect on the results.
we adjusted the models concerning health education also for
pupils overall absenteeism.
Table 1 Descriptive characteristics of the pupils
Results Pupil characteristics n (%)a/Mean (SD)
Table 1 presents descriptive characteristics of the pupils and Age (Mean, SD) 15.6 (1.1)
schools. Of the pupils, 48% were girls and 52% were boys. Sex
Twenty-seven percent of the pupils had parents with low Boy 11 583 (48)
education and 46% had parents with college or university level Girl 12 706 (52)
education. Depression scores indicated that 78% had no Socioeconomic background (parental education)
Comprehensive school/vocational school 6667 (27)
depression, 11% were mildly depressed, 8% were moderately Upper secondary school/upper secondary 6433 (26)
depressed and 2% were severely depressed. Two or more school and institute
frequent physical and psychological symptoms were reported College or university 11 189 (46)
by 15% of the pupils, 26% thought they were not heard at Depression (number of symptoms)
school, 5% had been absent from school due to truancy > 3 04 18 549 (78)
days during the past month and 25% thought health education 57 2592 (11)
815 1979 (8)
is insufficient. Twenty-two percent reported no classroom 1636 576 (2)
conversations or lessons about any of the health behaviours Frequent physical symptoms (number)
asked about, whereas 19% had had conversations or lessons 0 15 923 (68)
about all of the topics. 1 3979 (17)
Descriptive characteristics of the participating 136 schools 28 3636 (15)
Opinion about whether pupils have voice at school
are presented in table 2. Largest variance was found for pupil
Yes 17 812 (74)
composition, that is, proportion of pupils with low parental No 6168 (26)
education. The high variation on pupil composition may relate One-month absenteeism due to truancy
to the differences in participating schools of which a part were 03 days 22 004 (95)
secondary (compulsory) schools and a part were upper 4 days 1157 (5)
secondary (non-compulsory schools). Frequency of health education at school
Too much 953 (4)
Table 3 shows the relationship between staff experiences of
Sufficient 17 167 (71)
the psychosocial climate at school and pupil-reported depres- Too little 6048 (25)
sion and physical and psychological symptoms. Poor trust and Frequency of classroom conversations about health risk behaviours
opportunity for participation among staff was associated both All subjects several times 4548 (19)
with pupils depression and with physical and psychological 56 points 3700 (16)
symptoms (CORs = 1.14 and 1.17, respectively, in the final 34 points 6771 (28)
12 points 3607 (15)
models). To examine the relative contribution of school None 5190 (22)
variables on pupil outcomes, we made analyses before and after
adjustment for pupils parental education. However, a: Figures are numbers (percentages) unless otherwise stated

Table 2 School characteristics

Mean (SD) Variance Minimum Maximum

a
Proportion of pupils with low parental education (%) 28.60 (12.47) 155.60 3.66 67.86
Trust and opportunity for participation (school personnel mean) 3.70 (0.38) 0.15 2.63 4.50
Support for innovation (school personnel mean) 3.30 (0.39) 0.15 2.20 4.57
Orientation towards high-quality work (school personnel mean) 3.35 (0.31) 0.10 2.22 4.22
Accepted and clear goals (school personnel mean) 3.92 (0.25) 0.06 3.34 4.63

a: Parental education consists of comprehensive school/comprehensive school and vocational school.

 School climate and adolescents well-being 557

Table 3 Association between personnel evaluations about school psychosocial climate and pupils health problems

Depression Physical and psychological symptoms

School personnel reports of the
psychosocial climate at school COR (95% CI)a P-value COR (95% CI)a P-value

Trust and opportunity for participation

I highest 1.00 1.00
II 0.95 (0.841.09) 0.484 1.04 (0.921.16) 0.541
III 1.00 (0.881.14) 0.990 1.04 (0.931.17) 0.454
IV lowest 1.14 (1.011.29) 0.042 1.17 (1.051.31) 0.006
Support for innovation
I highest 1.00 1.00
II 0.98 (0.851.12) 0.739 0.97 (0.861.09) 0.763
III 1.08 (0.951.24) 0.228 1.08 (0.961.22) 0.431
IV lowest 1.09 (0.961.25) 0.194 1.05 (0.931.18) 0.560
Orientation towards high-quality work
I highest 1.00 1.00
II 1.00 (0.881.15) 0.970 0.93 (0.831.05) 0.224
III 0.96 (0.851.10) 0.586 1.02 (0.911.14) 0.734
IV lowest 1.09 (0.961.24) 0.189 1.06 (0.941.19) 0.352
Accepted and clear goals
I highest 1.00 1.00
II 1.00 (0.871.14) 0.957 1.03 (0.921.16) 0.603

Downloaded from eurpub.oxfordjournals.org by guest on April 18, 2011

III 1.03 (0.901.18) 0.647 1.06 (0.951.20) 0.306
IV lowest 0.98 (0.851.12) 0.761 1.01 (0.901.14) 0.870

a: Adjusted for pupils sex, age, parental education, the proportion of pupils with low parental education at school and for the
proportion of non-teacher respondents among school personnel

Table 4 Association between personnel evaluations about school psychosocial climate and school perceptions and behaviours
among pupils

Pupils school perceptions and behaviours

Pupils are not heard High truancy

School personnel reports of the
psychosocial climate at school OR (95% CI)a P-value OR (95% CI)a P-value

Trust and opportunity for participation

I highest 1.00 1.00
II 1.04 (0.831.31) 0.744 1.13 (0.861.49) 0.430
III 1.23 (0.981.53) 0.076 1.48 (1.131.93) 0.004
IV lowest 1.33 (1.061.66) 0.012 1.54 (1.182.00) 0.002
Support for innovation
I highest 1.00 1.00
II 1.30 (1.031.63) 0.028 0.90 (0.681.20) 0.482
III 1.39 (1.101.75) 0.006 1.14 (0.861.51) 0.371
IV lowest 1.37 (1.091.73) 0.008 1.18 (0.901.57) 0.230
Orientation towards high-quality work
I highest 1.00 1.00
II 0.88 (0.701.11) 0.290 1.10 (0.831.47) 0.495
III 1.05 (0.831.32) 0.684 1.23 (0.931.63) 0.146
IV lowest 1.07 (0.851.35) 0.576 1.23 (0.931.63) 0.140
Accepted and clear goals
I highest 1.00 1.00
II 1.36 (1.071.72) 0.011 1.27 (0.961.68) 0.089
III 1.26 (0.991.60) 0.059 1.47 (1.121.94) 0.006
IV lowest 1.26 (0.991.61) 0.058 1.39 (1.051.83) 0.023

a: Adjusted for pupils sex, age, parental education, the proportion of pupils with low parental education at school and for the
proportion of non-teacher respondents among school personnel

in the lowest quartile of trust and opportunity for participation

Discussion
In this large study of over 24 000 Finnish adolescents, we found
a relationship between staff-reported psychosocial climate
at school and various pupil-reported outcomes. We found
that among the different dimensions of psychosocial climate
at school, trust and opportunity for participation was the
only significant predictor of pupils health and the strongest
predictor of absenteeism due to truancy. Pupils from schools
compared to those in schools in the highest quartile had a
1.14-fold risk of depressive symptoms, a 1.17-fold risk
of numerous physical and psychological symptoms and a
1.54-fold risk of truancy. Other dimensions of psychosocial
climate at school were more weak or not at all associated with
pupil health and truancy. To our knowledge, this is the first
study to report the association between the staff-reported
psychosocial climate and the health of the pupils. An
association has been found between organizational climate as
558 European Journal of Public Health

Table 5 Association between personnel evaluations about school psychosocial climate and pupils reports of health education at
school

Pupils reports of health education

Low frequency of classroom conversations

Health education is insufficient and lessons about health risk behaviours
School personnel reports of the
psychosocial climate at school COR (95% CI)a P-value COR (95% CI)a P-value

Trust and opportunity for participation

I highest 1.00 1.00
II 1.05 (0.801.38) 0.710 1.17 (0.901.54) 0.237
III 1.24 (0.951.62) 0.107 1.03 (0.791.35) 0.802
IV lowest 1.30 (1.001.69) 0.054 1.06 (0.821.38) 0.651
Support for innovation
I highest 1.00 1.00
II 1.07 (0.811.41) 0.616 1.06 (0.811.39) 0.647
III 1.27 (0.971.68) 0.086 1.17 (0.891.53) 0.258
IV lowest 1.15 (0.871.52) 0.316 1.14 (0.871.49) 0.348
Orientation towards high-quality work
I highest 1.00 1.00
II 0.98 (0.751.30) 0.912 1.17 (0.901.53) 0.233
III 1.06 (0.811.40) 0.659 0.96 (0.741.24) 0.739

Downloaded from eurpub.oxfordjournals.org by guest on April 18, 2011

IV lowest 1.02 (0.771.35) 0.885 1.27 (0.981.66) 0.075
Accepted and clear goals
I highest 1.00 1.00
II 1.28 (0.981.68) 0.065 0.78 (0.601.01) 0.061
III 1.36 (1.041.77) 0.026 0.98 (0.751.27) 0.853
IV lowest 1.40 (1.061.83) 0.017 1.05 (0.801.37) 0.739

a: Adjusted for pupils sex, age, parental education, the proportion of pupils with low parental education at school and for the
proportion of non-teacher respondents among school personnel

seen by the personnel in youth mental health clinics and
patient outcomes.32 Earlier research has also shown that
pupils own perceptions of their school climate is associated
with their health, wellbeing and problem behaviours.2,413
As expected, our results indicate that trust and opportunity
for participation at school may be important determinants of
adolescent mental health and problem behaviours. When
people feel that they can participate in decision-making and
feel safe to contribute ideas, share information and challenge
existing ways of working without fear of being criticized. Thus,
trust and opportunity for participation refer to a situation in
which the members of the work group perceive the
interpersonal environment as non-threatening. High trust
among adults at school is likely to increase the chances that
pupils will confide in school staff, which in turn enables
early detection and prevention of antisocial behaviour and
violence.7 Truancy has frequently been considered part of
a general deviance or problem behaviour syndrome in which
poor school performance, mental health problems, aggression,
substance abuse and family difficulties are clustered.33
We found a relationship between personnel evaluations of
two dimensions of psychosocial climate at school, i.e. trust,
opportunity for participation and support for innovation and
pupils opinions of whether they are heard at school or not.
These associations suggest that there is a shared perception of
school collaboration between pupils and adults. Performance-
oriented dimensions (orientation towards high-quality work
and accepted and clear goals) were not significantly related to
pupils opinions of whether they were heard at school or not.
We also found some, although modest evidence of an
association between the psychosocial climate at school and
pupils reports of health education received at school. Accepted
and clear goal among staff were related to pupil reports of
more sufficient health education. Confounding by pupils high
absence rates was not possible because overall absence rates
were controlled for in the analyses. Health education is a
relatively new subject at schools in Finland and it is possible
that clear concepts and models of how to organize this subject
have not yet been developed in many schools. Our results
suggest that accepted and clear goals, meaning a valued
outcome which represents a higher-order goal and motivating
force at work, and an orientation towards high-quality work,
meaning work-unit members continuous attempts to improve
the quality of their work, may predict the frequency of health
education and lessons about health risks at school. It is possible
that in schools where the goals are clear and well debated there
might more likely to be a shared opinion that health
promotion is a valued goal, and personnel in these schools
may have a greater motivation to teach health-related subjects
to adolescents. In the future, it would be important to examine
whether this aspect of psychosocial climate at school also
predicts commitment and success in school-based interven-
tions to promote healthy lifestyle among adolescents.16,17
The main strength of this study was the use of information
obtained from two independent sources, school personnel and
pupils, which minimizes the bias related to common method
variance. We adjusted for the most likely confounders in our
analyses, that is, pupils sex, age, absence rate and parents
education level as well as the pupil composition at school.
Other characteristics, such as attachment to the family,
have been shown to affect adolescent health and problem
behaviour,2 but this was not adjusted for in the present
analysis. However, attachment to the family may act as
a confounding factor only if it is also related to school staff
perceptions of psychosocial climate at school.
The response rate in the pupils survey was high (84%),
whereas the response rate among the school personnel was
relatively low, 54%. The low-response rate may have biased the
results if non-response was non-random in relation to school
climate. This is unlikely at least in relation to pupil outcomes,
as the surveys were completely independent.
Because our study was cross sectional, we cannot draw
conclusions about the temporal order between the variables. It
is possible that health problems and problem behaviour among

adolescents affect staff perceptions of school psychosocial
climate. However, pupils reports about insufficient health
education cannot easily be explained by pupils passivity or
problem behaviour only. Selection bias cannot be ruled out
either. Personnel with low work motivation and poor
teamwork capacity may be employed by schools where
pupils also have health-related and behavioural problems.
Prospective studies are therefore needed to confirm the
direction of the causality. Pupil socioeconomic composition
is an important factor potentially related to the psychosocial
climate at school. However, a limitation in our study is that
our measurement of pupil composition was based on pupils
report of their parents education which may not the best one
to catch pupil socioeconomic composition at school. However,
we analysed correlations between psychosocial variables and
pupil composition at school and found that none of them were
significantly correlated with pupil composition (P > 0.30),
indicating that this limitation is unlikely to be a major source
of bias in our study.
In summary, the psychosocial climate at school as reported
by school staff is associated with adolescent school perceptions,
truancy, health symptoms and received health education.
Trust and opportunity for participation among the school staff
seem to be especially important for pupils health problems
and truancy, whereas accepted and clear goals may be
important for pupils perceptions about the sufficiency of
health education. The present findings suggest that the
psychosocial climate at school is an important contributor
to the promotion of adolescent health and wellbeing.
Funding
Academy of Finland (#117604, #124322, and #124271).
Conflict of interest: None declared.
Key points
 The psychosocial climate at school reported by school
staff is associated with adolescent reports of their
health, school attitudes, truancy and received health
education.
 Trust and opportunity for participation at school seem
to be especially important for pupils health problems
and truancy.
 Accepted and clear goals among school personnel may
be important for the sufficiency of health education
received at school.
References
1 Sellstrom E, Bremberg S. Is there a school effect on pupil outcomes?
A review of multilevel studies. J Epidemiol Community Health
2006;60:14955.
2 Resnick MD, Bearman PS, Blum RW, et al. Protecting adolescents from
harm. Findings from the National Longitudinal Study on Adolescent Health.
JAMA 1997;278:82332.
3 Youngblade LM, Theokas C, Schulenberg J, et al. Risk and promotive factors
in families, schools, and communities: a contextual model of positive youth
development in adolescence. Pediatrics 2007;119:54753.
4 Konu AI, Lintonen TP, Rimpela MK. Factors associated with school-
childrens general subjective wellbeing. Health Educ Res 2002;17:15565.
5 Haavet OR, dalen I, Straand J. Depressive symptoms in adolescent pupils are
heavily influenced by the school they go to. A study of 10th grade pupils in
Oslo, Norway. Eur J Public Health 2005;16:4004.
School climate and adolescents well-being 559
6 Simons-Morton BG, Davis Crump A, Haynie DL, et al. Student-
school bonding and adolescent problem behaviour. Health Educ Res
1999;14:99107.
7 Greene MB. Reducing violence and aggression in schools. Trauma Violence
Abuse 2005;6:23653.
8 Brookmeyer KA, Fanti KA, Herrich CC. Schools, parents and youth
violence: a multilevel, ecological analysis. J Clin Child Adolesc Psychol
2006;35:50414.
9 Bonny AE, Britto MT, Klostermann BK, et al. School disconnectedness:
identifying adolescents at risk. Pediatrics 2000;106:101721.
10 Voisin DR, Salazar LF, Crosby R, et al. Teacher connectedness and
health-related outcomes among detained adolescents. J Adolesc Health
2005;37:337.
11 Mc Neely C, Falci C. School connectedness and transition into and out of
health risk behaviour among adolescents: a comparison of social belonging
and teacher support. J School Health 2004;74:27283.
12 Resnick MD, Harris L, Blum R. The impact of caring and
connectedness on adolescent health and well-being. J Paediatr Child Health
1993;29:S39.
13 Andersen A, Holstein BE, Due P. School-related risk factors for drunkenness
Downloaded from eurpub.oxfordjournals.org by guest on April 18, 2011
among adolescents: risk factors differ between socio-economic groups. Eur J
Public Health 2006;17:2732.
14 Henry KL. Whos skipping school: Characteristics of truants in 8th and 10th
grade. J School Health 2007;77:2935.
15 Mattison RE. School consultation: A review of research on issues unique
to the school environment. J Am Acad Child Adolesc Psychiatry
2000;39:40213.
16 Parcel GS, Perry CL, Kelder SH, et al. School climate and the
institutionalization of the CATCH program. Health Educ Behav
2003;30:489502.
17 Matthews C, Boon H, Flisher AJ, Schaalma HP. Factors associated with
teachers implementation of HIV/AIDS education in secondary schools in
Cape Town, South Africa. AIDS Care 2006;18:38897.
18 West M. The social psychology of innovation in groups. In: West M, Farr J,
editors. Innovation and creativity at work: psychological and organisational
strategies. Chichester, UK: John Wiley & Sons, 1990, 30933.
19 Jalonen P, Virtanen M, Vahtera J, et al. Predictors of sustained organisational
commitment among nurses with temporary job contracts. J Nurs Admin
2006;36:26876.
20 Kivimaki M, Sutinen R, Elovainio M, et al. Sickness absence in hospital
physicians 2-year follow-up study on determinants. Occup Environ Med
2001;58:3616.
21 Ylipaavalniemi J, Kivimaki M, Elovainio M, et al. Psychosocial work
characteristics and incidence of newly diagnosed depression: a prospective
cohort study of three different models. Soc Sci Med 2005;61:11122.
22 Hakala P, Rimpela A, Salminen JJ, et al. Back, neck, and shoulder
pain in Finnish adolescents: national cross sectional study. Br Med J
2002;325:743.
23 Kaltiala-Heino R, Rimpela M, Marttunen M, et al. Bullying, depression, and
suicidal ideation in Finnish adolescents: School survey. Br Med J
1999;319:34851.
24 Vahtera J, Kivimaki M, Pentti J, et al. Organisational downsizing, sickness
absence, and mortality: 10-town prospective cohort study. Br Med J
2004;328:555.
25 Vahtera J, Poikolainen K, Kivimaki M, et al. Alcohol intake and sickness
absence: a curvilinear relation. Am J Epidemiol 2002;156:96976.
26 Kivimaki M, Elovainio M. A short version of the Team Climate Inventory:
development and psychometric properties. J Occup Organ Psychol
1999;72:2416.
27 Anderson N, West M. The Team Climate Inventory: manual and users guide.
Windsor: NFER-Nelson, ASE Press, 1994.
28 Beck AT, Beck RW. Screening depressed patients in family practice. A rapid
technic. Postgrad Med 1972;52:815.
29 Chinet L, Plancherel B, Bolognini M, et al. Substance use and depression.
Comparative course in adolescents. Eur Child Adolesc Psychiatry
2006;15:14955.
560 European Journal of Public Health
30 Kaltiala-Heino R, Kosunen E, Rimpela M. Pubertal timing, sexual behaviour
and self-reported depression in middle adolescence. J Adolesc
2003;26:53145.
31 Virtanen M, Kivimaki M, Elovainio M, et al. Neighbourhood socioeconomic
status, health and working conditions of school teachers. J Epidemiol
Community Health 2007;61:32630.
32 Glisson C, Landsverk J, Schoenwald S, et al. Assessing the organisational
social context (OSC) of mental health services: implications for research and
practice. Adm Policy Ment Health 2008;35:98113.
33 Miller P, Plant M. Truancy and perceived school performance: an alcohol
and drug study of UK teenagers. Alcohol Alcoholism 1999;34:88693.
Received 30 April 2008, accepted 17 February 2009
Downloaded from eurpub.oxfordjournals.org by guest on April 18, 2011