Content uploaded by Kirsten J Hancock
Author content
All content in this area was uploaded by Kirsten J Hancock on May 26, 2020
Content may be subject to copyright.
Does the reason matter? How
student-reported reasons for school absence
contribute to differences in achievement
outcomes among 14–15 year olds
Kirsten J. Hancock
a,b,
*, Michael A. Gottfried
c
and
Stephen R. Zubrick
a,b
a
Telethon Kids Institute, The University of Western Australia, Perth, Australia;
b
The Graduate School of Education, The University of Western Australia, Perth, Australia;
c
Gevirtz Graduate School of Education, University of California Santa Barbara,
California, USA
While an emerging body of research has examined the effects of school absences on student out-
comes, there is comparatively little research examining the different reasons contributing to school
absence, how common these reasons are, and the extent to which different types of absences are dif-
ferentially associated with achievement. To address these gaps, we used data from the Longitudinal
Study of Australian Children to examine the reasons for school absence as reported by 14–15 year
olds and how these reasons relate to achievement outcomes in Year 9. Only 7% of 14–15 year olds
indicated they had been absent in the previous six months without parental consent, of which 46%
indicated the most recent absence was due to problems at school. Of the 90% of students who had
been absent with parental consent, only 6% said the most recent absence was due to problems at
school. After controlling for student, family and school characteristics and Year 7 achievement,
Year 9 achievement was most strongly associated with absences related to student- or family-level
reasons. While schools typically bear the responsibility for monitoring and responding to absen-
teeism, the drivers of absence may not be related to factors that schools can realistically address. For
schools, addressing absenteeism requires a dual approach of preventing avoidable absences and mit-
igation strategies for when either avoidable or unavoidable absences occur.
Keywords: absences; achievement; high school
Introduction
Students with larger quantities of school absences face greater risks of a range of
adverse educational outcomes. These include lower achievement (Gottfried, 2009;
Silverman, 2012), poorer grade retention (Neild & Balfanz, 2006) and increased like-
lihood of early school dropout (Rumberger, 1995). Poor attendance patterns can start
as early as kindergarten or Year 1 (Chang & Romero, 2008; Hancock et al., 2013)
and are associated with long-term cycles of diminishing attendance, poor engagement
*Corresponding author. Telethon Kids Institute, The University of Western Australia, PO
Box 855, West Perth, WA 6872, Australia. E-mail: kirsten.hancock@telethonkids.org.au; Twitter:
@kirstenhancock8.
©2018 British Educational Research Association
British Educational Research Journal
Vol. 44, No. 1, February 2018, pp. 141–174
DOI: 10.1002/berj.3322
and lower achievement outcomes throughout a student’s years of schooling (Alexan-
der et al., 1997). Higher absences are also associated with longer-term outcomes
beyond school. For example, of students missing more than 15% of available school
days at age 15 in England, 31% were not in education, employment or training at age
18, compared with 12% of students who attended school more regularly (Depart-
ment for Education, 2011). Additionally, higher absences have impacts beyond the
individual student: recent evidence from the USA suggests that in 2014–15, publicly
funded schools in California did not receive $1 billion in funds (allocated based on
daily attendance rates) as a result of student absences (Office of the California Attor-
ney General, 2015).
Student absences are therefore a key policy issue for schools and education bureau-
cracies in most Western countries. In the USA, chronic absenteeism (defined as miss-
ing more than 10% of school days for any reason) has shifted from a simple
administrative task to becoming a key accountability metric for schools and commu-
nities under the Every Student Succeeds Act.In the UK, responses to student
absences have become increasingly punitive for parents and caregivers. Since 2013,
the range of acceptable reasons for student absences has become restricted, for exam-
ple, an in-term holiday is no longer a valid reason for missing school. In April of
2017, the UK Supreme Court determined that a child must attend school on every
day that the school requires them to do so. Parents who do not ensure attendance
without a valid excuse may be fined up to £1,000. This penalty can apply for short
absences, like holidays, even if the student otherwise attends school regularly. In
2017, the Press Association obtained data under the Freedom of Information Act to
find that 20,000 individuals were taken to court in 2015 for failing to ensure a child
went to school, an increase of 61% since 2011 (Yusuf & Bowcott, 2017). In three-
quarters of cases the individual was guilty, and in 77% of these cases fines were
issued. Eight people received jail terms and 553 were given community sentences.
In Australia, the setting for the current study, attendance policies vary across states.
Typically, escalation procedures apply, where increasing levels of unexplained
absences are followed up with phone calls, letters, parent meetings, family action
plans and only in extreme cases and as a last resort, a recommendation for prosecu-
tion by the state. Very few cases are prosecuted, and where a case reaches this stage, a
fine of up to A$3,000 can apply. There is no evidence to suggest that a fine or court
order mandating school attendance results in improved attendance for individual stu-
dents who have reached this point. In 2009, the Australian government trialled an
intervention in 44 disadvantaged schools, where social security payments would be
suspended for families where children had unsatisfactory attendance and would be
restored when attendance returned to satisfactory levels. While the trial included sup-
port from social workers to help families address the barriers to school attendance,
student attendance did not improve at the trial sites (Department of Education,
Employment and Workplace Relations, 2014). Beyond this trial and the prosecution
of extreme cases, and unlike the policy setting in England, currently there are no
financial penalties for caregivers of students with irregular attendance in Australia. In
most states, schools are encouraged to work collaboratively with students and parents
to develop and implement strategies for improved attendance, including the engage-
ment of external services and agencies to assist with the development of strategies or
142 K. J. Hancock et al.
©2018 British Educational Research Association
the provision of support services. Given the differences in policy contexts across
countries, it is unclear how a study focused on Australian students will generalise to
other settings.
It is well recognised that the drivers of school absences need to be understood in
order to develop effective interventions for reducing absences. In this context, it is
unclear that financial penalties for parents are an effective preventive mechanism for
addressing the vast array of reasons that students have for missing school. Certainly
the risk factors for higher absence rates are well known and include poverty and disad-
vantage (Chang & Romero, 2008; Silverman, 2012; Tobin, 2014), chronic illness
(Shaw & McCabe, 2007) and mental health problems (Lawrence et al., 2015),
amongst others. However, a simple per capita count of absences assumes that all
absences impact upon achievement and other outcomes in similar ways. While both
authorised and unauthorised absences have been associated with lower achievement
levels, and particularly so for unauthorised absences (Gottfried, 2009; Hancock et al.,
2013), there is limited evidence about how the association between absences and
achievement varies according to more precise reasons for absences. Details about this
variation might be useful to identify which types of absence interventions are
required, for how many students, and if such interventions would be effective at
improving outcomes other than attendance. A government report from England, for
example, showed that increasing levels of most authorised absence types, including
illness, were associated with lower achievement among Grade 6 students. However,
the gradient varied by absence type, and authorised absences for family holidays were
not associated with lower achievement at all (Department for Education, 2011).
From these data it is unclear why the attendance policy in England would focus on
family holidays as a type of absence that should attract a fine.
A taxonomy of absence
The few studies that have examined the variation between reasons for being absent
and student outcomes have been based on large administrative datasets which have
predominantly examined the effects of authorised versus unauthorised absences
(Gottfried, 2009; Hancock et al., 2013; Aucejo & Romano, 2014; Gershenson et al.,
2017). However, even within these broad categories there are multiple reasons under-
lying both authorised absences (e.g. illness, cultural events) and unauthorised
absences (e.g. truancy, term-time family vacations). Kearney (2008) defined a num-
ber of different types of absences from school that go beyond the authorised versus
unauthorised classification (conceptualised in Figure 1).
In Figure 1, authorised or unauthorised absences are first differentiated, consistent
with typical reporting levels of school districts. Generally speaking, authorised
absences are those that are accepted by the school, including short-term absences such
as illness. Authorised absences may nonetheless be excessive, where the reason under-
lying the absence is excusable but the duration may still be associated with significant
educational consequences without intervention or support (e.g. chronic illness).
Unauthorised absences, where the reason is either not acceptable or simply not
provided to the school, reflect a much broader array of factors and behaviours. They
can be further categorised as being student-driven (school refusal) or parent-driven
Does the reason matter? 143
©2018 British Educational Research Association
(withdrawal) factors related to logistics, personal circumstances, engagement or atti-
tude. Parent-driven absences may be relatively innocuous (e.g. short family vaca-
tions), but may also relate to limited transport options, caring requirements,
separation anxiety or a limited value placed on education.
Kearney (2008) suggests that student-driven absences can be further categorised
into one of two domains. The first relates to students pursuing positive reinforce-
ment, for example, to get attention from parents, or finding activities outside of
school more enjoyable. The second domain relates to students avoiding negative rein-
forcement, such as avoiding school-related stimuli that promote anxiety/depression,
or avoidance of aversive situations like bullying. This classification of absences relates
to students of all ages, however, the prevalence of different types of disorders will dif-
fer for younger and older students, to reflect developmental differences between
students.
Understanding which of these absences are more common than others, and which
ones are most strongly associated with student outcomes, is critical for determining
what sort of resources may be required for addressing absenteeism. For example,
Kearney (2008) suggests that problematic absenteeism often occurs alongside a men-
tal health problem, and that increasing the availability of mental health services for
children and adolescents may be effective at reducing absenteeism. Yet if only a very
small proportion of students are missing school specifically because of a mental health
problem, then it may not be the best use of school resources to employ a school coun-
sellor or psychologist, where referrals to other service providers may be more
appropriate.
School
absence
Authorised
Short-term
(typical)
Excessive
e.g. chronic
illness
Suspension
Unauthorised
Student-driven
absence
(refusal)
Pursuing positive reinforcement
e.g. to get attention from parents, or activities
outside of school are more enjoyable (e.g. drug
use)
Avoiding negative reinforcement
e.g. avoiding school-related stimuli that
promote anxiety/depression, or to avoid
aversive situations
Parent-driven
absence
(withdrawal)
e.g. Child needs to assist parent at home,
economic purposes, logistics, poor engagement
with education, protect from school-based
threat
Figure 1. Conceptual overview of the drivers of student absence.
144 K. J. Hancock et al.
©2018 British Educational Research Association
Why does the reason matter?
It is also important to understand how different reasons for absence relate to differ-
ences in students’ outcomes, so that particular student needs can be prioritised. On
face value, any distinction between reasons for absence and achievement outcomes
may be irrelevant. If the relationship between absence and achievement is causal and
based on students missing instructional time, the reason for any absence would not
matter. However, previous research has demonstrated that the association between
absence and achievement outcomes is much larger for unauthorised than authorised
absences (Gottfried, 2009; Hancock et al., 2017), indicating that not all absences are
equal. While unauthorised absences may act as a proxy for other unobserved charac-
teristics, such as parent engagement, there are other processes at play. For example,
students who miss school for a family holiday may not fall behind if the majority of
parents ensure their child makes up for lost time. On the other hand, students who
skip school because they are disengaged may see no incentive to catch up on any
material they’ve missed, and when they do attend school may not be ‘present’ in class.
These types of absences may therefore be more strongly associated with achievement
outcomes than other types of absences. So far, these different relationships remain
largely undocumented.
Reasons for absence vary by age and gender
When examining the prevalence of different reasons for absence and how these
absences relate to student outcomes, consideration must also be given to the age and
gender of the student. Australian data shows that while the total absence rate of gov-
ernment school students increases from 8% to 14% between Years 6 and 10, the pro-
portion of absences that are unauthorised increases from 33% to 42% (Hancock
et al., 2013). The increase in both authorised and unauthorised absences in high
school likely reflects a myriad of changes that accompany adolescent transitions,
including an increase in the prevalence of chronic illness (Australian Institute of
Health and Welfare, 2005), the onset of mental disorders (Lawrence et al., 2015),
social problems like bullying (Finkelhor et al., 2015) and decreasing engagement with
school (Darr, 2012; Wylie & Hodgen, 2012), all of which are associated with higher
absence rates. These factors are all in addition to those that contribute to absences
among younger students which also persist into adolescence, such as socioeconomic
disadvantage. In this way, parent-driven absence may be more common among
younger students, whereas student-driven absence may become more salient for older
students. From an intervention perspective, schools may be better equipped to inter-
vene in student-driven absences that arise from peer- or school-related issues,
whereas family-based interventions would be more appropriate for younger students.
For this reason, the current study focuses on the reasons for absence as reported by
adolescents.
The emergence of issues like chronic illness, bullying or mental health in adoles-
cence can also differ by gender. Previous research has shown few differences in overall
rates of authorised and unauthorised absence by gender, or in the association with
achievement outcomes for boys and girls (Gottfried, 2009; Hancock et al., 2013).
Does the reason matter? 145
©2018 British Educational Research Association
However, these studies of overall absence rates may mask differences that boys and
girls have for missing school. A UK study found that among secondary students the
proportion of boys admitting to truancy was higher than girls, though this pattern var-
ied by the race of the students and the racial composition of the schools they attended
(Malcolm et al., 2003). Boys and girls were equally as likely or unlikely to skip school
when they were bullied, disliked a teacher or were under pressure from home. How-
ever, girls are more likely than boys to report being bullied (Messias et al., 2014) and
to have less conflict with teachers (Baker, 2006; Hughes et al., 2006), again suggest-
ing that the reasons boys and girls have for missing school will be different.
The extent to which mental disorders relate to school absence may also differ by
gender. Adolescent girls tend to have higher 12-month rates of mood, anxiety and
eating disorders than boys, whereas boys have higher rates of behaviour and sub-
stance disorders than girls (Zahn-Waxler et al., 2008; Kessler et al., 2012; Lawrence
et al., 2015). The average number of days out of school in the previous 12 months
due to symptoms of these disorders also varies: anxiety (20 days), major depressive
disorder (23 days), ADHD (9 days), conduct disorder (17 days) and oppositional
problem behaviours (9 days; Lawrence et al., 2015). These patterns suggest that the
number of absences that could be attributed to mental health problems will likely vary
by student gender, and in ways that may also differentially relate to learning. For
example, boys with conduct disorder or oppositional defiance disorder may have
more issues with teachers that result in suspensions and difficulties learning while in
the classroom, and may find it more difficult to catch up on missed school days.
Previous studies examining reasons for absence
The literature examining different reasons for absence, beyond authorised or unau-
thorised, is limited. As discussed earlier, the Department for Education in England
published descriptive data about the prevalence of different absence codes, how the
prevalence varied by student characteristics, and how these absences related to
achievement outcomes of Grade 6 and 11 students in 2009/10 (Department for Edu-
cation, 2011). Close to 60% of all absences were for illness, and 21% were for other
authorised circumstances. Approximately 17% of absences were unauthorised, com-
prised of other circumstances (11%), unexplained absences (3%) and unauthorised
family holidays (2%). All absence types, excluding authorised family holidays, were
associated with lower achievement.
Rothman (2002) reported on the distribution of reasons for absence using adminis-
trative records of a population of students in South Australia, based on predefined
codes against which schools could record each absence. Rothman found that around
half of absences in primary school were recorded as an illness, which decreased in
high school as the proportion of unexplained absences increased. However, there was
very little contextual information accompanying these population figures, nor was
there any analysis provided of how these absence codes varied by student or school
characteristics, nor how these codes related to differences in achievement outcomes.
deJung and Duckworth (1986) conducted a study across six high schools in the
USA asking students themselves about the reasons they skipped particular classes.
They reported that 20% of students stated that they had other things to do, whereas
146 K. J. Hancock et al.
©2018 British Educational Research Association
illness, personal problems, homework and being bored each accounted for 10% of
responses. However, as these responses were in relation to ‘skipping class’, they are
unlikely to be an accurate reflection of the reasons contributing to all absences.
Finally, Malcolm et al. (2003) interviewed students, parents and teachers in the
UK to ascertain reasons for truancy among high school students. Approximately 16%
of the 528 high school students surveyed admitted to truanting from school. Of these,
the authors suggested that the main reasons reported were based on school-related
factors such as boredom, problems with lessons and teachers, anticipation of trouble
and frustration at school rules. A survey of parents identified similar issues contribut-
ing to truancy as the students. In contrast, teachers believed that parent-related
issues, such as low education values, negative attitudes towards the school, long work
hours or issues at home—like domestic violence—were the main causes of truancy.
The different perspectives of students, parents and teachers about school absence
highlights another gap in the literature relating to the extent to which school absence
is endorsed or permitted by parents: parent-sanctioned absence from school does not
necessarily equate with absences being excused by the school. For example, a parent
may take a student out of class for a holiday during term time, which may not be
excused by the school. Alternatively, a student may be ill and remain home with par-
ental permission, but the parent never informs the school and the absence is recorded
as unauthorised or unexplained. To determine the extent to which different types of
absences are associated with differences in achievement levels, student reports on
why they are absent from school and if their absences are parent-sanctioned are there-
fore highly informative.
The current study
To summarise, there is limited research about how different reasons for absence con-
tribute to variation in achievement outcomes. The current study aims to expand on
the literature by examining these issues using broadly representative survey data from
an Australian longitudinal cohort study. We address the following key questions: (1)
To what extent are absences among adolescents sanctioned by parents? (2) Within
parent-permitted and non-permitted absences, what are the most common reasons
for missing school? (3) Which characteristics are associated with these different rea-
sons? (4) How do the reasons provided relate to achievement outcomes for boys and
girls and across different learning domains?
Methods
Data source and participants
This study uses data from the Longitudinal Study of Australian Children (LSAC), a
nationally representative and multidisciplinary survey of Australian children and their
families. The study commenced in 2004 with two cohorts of children, including a
5,103 infant cohort (0–1 year olds) and a 4,983 kindergarten cohort (‘K-cohort’, 4–5
year olds). The same cohorts of children and families have been followed up every
two years, with six waves of data now available up to when the study children are
Does the reason matter? 147
©2018 British Educational Research Association
10–11 and 14–15 years of age. We use data from the K-cohort at Wave 6 when the
study children were 14–15 years of age, as this was the first time that the study
children were asked to report on their own absence behaviours. Table 1 provides a
summary of the data collection schedule, age range and sample size for the K-cohort
at each wave. Sample retention to Wave 6 was 71% for the K-cohort. Multiple
imputation procedures were used to address missing data concerns and are described
below.
Survey design. The LSAC survey methodology has been detailed extensively else-
where (Soloff et al., 2005). In brief, the LSAC sampling frame was based on the
Medicare Australia enrolment database. Using a two-stage clustered sample design, 1
in 10 Australian postcode areas were first randomly selected from the database, and
eligible children were then sampled within postcodes in the second stage. The initial
response rate at Wave 1 was 47.0% for the K-cohort. When compared with the 2001
Census data, the initial sample was broadly representative of the Australian popula-
tion of families with children in the relevant age group, though slightly under-repre-
sentative of families who were single-parent, non-English-speaking, living in rental
properties or in remote areas (Soloff et al., 2006). In subsequent waves of data collec-
tion, these same characteristics were over-represented in the families who dropped
out of the study (Sipthorp & Misson, 2009). Of the families still participating at Wave
6, 95.7% of the adolescents also completed the self-report. Multiple imputation was
used to address non-response and attrition bias. The details of this process are
detailed in the Analytic approach section.
At Wave 6, data collection methods included computer-assisted parent interviews
in the home, a computer-assisted self-interview for the study child, mailed teacher
surveys, physical measurements and interviewer observations. Data from this study
are primarily drawn from the child’s self-report, with family and demographic factors
drawn from the parent interview.
Survey methodology and content development is managed by the Australian Insti-
tute of Family Studies in partnership with the Department of Social Services and the
Australian Bureau of Statistics. A Consortium Advisory Group consisting of
leading Australian researchers provides advice and expertise to the study. Members
of the Australian researcher community and government staff can access the datasets
for a small licensing fee.
Table 1. Age range, sample size and study retention, K-cohort, Waves 1–6
Wave 1
(2004)
Wave 2
(2006)
Wave 3
(2008)
Wave 4
(2010)
Wave 5
(2012)
Wave 6
(2014)
K-Cohort
Age (years) 4–56–78–910–11 12–13 14–15
Sample size 4,983 4,464 4,331 4,169 3,956 3,537
Sample retention (%) –89.6 86.9 83.7 79.4 71.0
Child self-report 3,386
Child self-report rate 95.7%
148 K. J. Hancock et al.
©2018 British Educational Research Association
Measures
Absence behaviours. Adolescents were asked how many times in the last six months
they were late for school, had cut or skipped classes, were absent from school without
parental permission, were absent from school with parental permission, or had got
into trouble for not following school rules. Responses for each statement included
never, 1–2 times, 3–6 times, 7–9 times, and 10 or more times. The options in this sur-
vey response set are consistent with other national-level surveys, including the US-
based Early Childhood Longitudinal Study, developed by the US Department of
Education.
Adolescents were then asked about the main reason for their most recent
absence. For parent-permitted absences, the available responses included stress,
anxiety or depression, tiredness, other illness or medical condition, medical, dental
or other specialist appointment, bullying, problems with friends, problems with
teachers, to avoid school work, to complete school work, caring for another family
member, illness of a family member, out-of-school activities, family events or other
reasons. For absences without parental permission, the same responses above were
provided, with an additional option for ‘pressure from friends to do other things’.
As several of the categories had few responses, some were collapsed to form broader
categories, as follows: school or social problems (bullying, problems with friends,
problems with teachers, to avoid school work); family health or caring reasons
(caring for another family member, illness of a family member); and other (other,
to complete school work).
Achievement outcomes. Achievement was assessed using test scores from the National
Program of Literacy and Numeracy (NAPLAN), which are linked to the LSAC data-
set for families who consented to data linkage at Waves 3 and 4. The NAPLAN is a
suite of standardised tests of numeracy, reading, spelling and writing, and has been
administered to all Australian students in Years 3, 5, 7 and 9 each year since 2008.
Tests are administered on the same day across the country, towards the end of May.
We use the Year 9 numeracy, reading and spelling scores to assess achievement out-
comes, and use the corresponding Year 7 scores to account for prior achievement.
The Year 9 scores broadly correspond with the approximate age of adolescents at
Wave 6 (14–15 years), however, we are unable to determine if adolescents undertook
the assessment before or after the LSAC interview or self-report. In most Australian
states, Year 9 is the third year of secondary school, and the second year of secondary
school in other states.
NAPLAN scores are scaled by the administering agency (the Australian Curricu-
lum and Reporting Authority, ACARA) to range from 0 to 1,000, so that student pro-
gress can be compared between students and within students over time. For this
study, and to assist with interpretation, numeracy and reading scores in Years 7 and 9
were standardised to have a mean of 0 and a standard deviation of 1. Scores were
standardised to population norms using the Australian-level means and standard
deviations. Higher Year 9 NAPLAN test scores are significantly associated with
higher university entrance exam scores and access to higher education (Houng & Just-
man, 2014).
Does the reason matter? 149
©2018 British Educational Research Association
Demographics. Student-related demographic variables included age in years at the
time of the NAPLAN test, gender and parent-reported measures of whether the ado-
lescent had a condition or disability that had lasted at least six months prior to the sur-
vey, including but not limited to sight, hearing or speech problems, learning
difficulties, other physical problems or mental illness. Also included were measures of
whether the adolescent had ever repeated a year, and whether they spoke a language
other than English at home. These variables were collected during the parent
interview.
Social and emotional wellbeing problems were assessed using the Strengths and
Difficulties Questionnaire (SDQ; Goodman, 2001). The SDQ is a 25-item scale
comprising five subscales covering peer relationships, conduct problems, hyperactiv-
ity, emotional problems and pro-social behaviour. Total scores were calculated by
summing scores on the peer, conduct, hyperactivity and emotional problems sub-
scales. Scores could range from 0 to 40, with higher scores representing poorer func-
tioning. We used the recommended cut-points for the youth self-reports to identify
adolescents in the borderline or abnormal range of the SDQ (total score ≥16), who
were then classified as having social or emotional wellbeing (SEWB) problems
(Goodman et al., 2000).
Parent demographic variables—previously documented in the literature as being
related both to absence and to achievement outcomes—included the involvement of
mothers and fathers in education, the highest level of parent education, family struc-
ture, family income and maternal psychological distress.
Parents’ highest educational attainment was collected in the parent interview, and
assessed the highest level of education achieved for each parent living with the adoles-
cent. Responses were dichotomised to identify parents who had not completed Year
12 and those who had attained a qualification beyond school.
For parent involvement, adolescents were asked: ‘How much interest does your
mother/father show towards your learning and education (this could include helping
you with homework or otherwise encouraging your learning)?’ Responses included a
lot of interest, some interest, not much interest or no interest at all. Responses were
dichotomised to identify mothers and fathers showing not much or no interest at all,
versus a lot or some interest.
Family income was based on total household income from all sources. Families in
the lowest 25% of income were classified as low income.
Maternal psychological distress was assessed using the Kessler K6 scale (Kessler
et al., 2002). Mothers were asked how often in the past four weeks they had felt
nervous, hopeless, restless or fidgety, that everything was an effort, so sad that
nothing would cheer you up or worthless, and responded on a five-point scale
from 0 =all of the time to 4 =none of the time. Responses were reverse coded,
summed and adjusted to generate a total score ranging from 0 to 24, where higher
scores represented greater levels of non-specific psychological distress. Scores of
13 or above indicate a probable serious mental illness (Kessler et al., 2003) and a
lower cut-point of 6+identifies mild to moderate non-specific psychological dis-
tress (Furukawa et al., 2003; Prochaska et al., 2012). We used a slightly higher
cut-point of 8+to classify mothers as having elevated non-specific psychological
distress.
150 K. J. Hancock et al.
©2018 British Educational Research Association
School-related measures. Previous research has found that higher levels of school con-
nectedness are associated with higher attendance rates (Sanchez et al., 2005), aca-
demic self-efficacy (Ibanez et al., 2004) and better grades (Booker, 2007). To
determine the extent to which particular reasons for absence reflected broader prob-
lems at school and social/emotional wellbeing, a number of measures were used relat-
ing to school connectedness and academic self-concept.
A student’s degree of school connectedness was assessed using the Psychologi-
cal Sense of School Membership questionnaire (Goodenow, 1993). The full
scale is an 18-item scale of school connectedness, however the LSAC question-
naire only used 12 of the items, corresponding with the three subscales or fac-
tors identified in previous research (You et al., 2011). Adolescents could
respond to each item from 1 =not at all true to 5 =completely true. Subscale
scores were created for three factors underlying the item set comprising school
acceptance (five items, a=0.76, e.g. I can be myself at school), school caring
(four items, a=0.78, e.g. teachers are interested in me) and rejection (three
items, a=0.72, e.g. I feel like I don’t belong at this school). Responses were
summed for each subscale, with negative items reverse coded. As the scales
were highly skewed, we set cut-points to identify the lowest approximate 20%
on each subscale, categorising these adolescents as being low acceptance,low
school caring and high rejection.
Academic self-concept was based on three statements in the self-report question-
naire, including: (1) I learn things quickly in most school subjects; (2) I’m good at
most school subjects; and (3) I do well in most school subjects. Response options
ranged from 1 =strongly disagree to 5 =strongly agree. Responses were summed to
form a total score ranging from 1 to 4, where higher scores reflect higher academic
self-concept. Again, the lowest 20% of scores were categorised as having low academic
self-concept.
School liking was assessed by three items asking respondents whether they like: (1)
maths and number work at school; (2) reading and writing activities at school; and
(3) learning about science and science activities at school. Responses included 1 =
yes, 2 =sometimes and 3 =no. Responses were reverse scored and summed, so that
higher responses reflected higher school liking. The lowest 20% of scores were cate-
gorised as having low school liking.
School sector (government, Catholic or independent) was also included as a
covariate. All Australian schools are funded by state and federal governments on
a per-student basis, irrespective of school sector, where Catholic and independent
schools receive federal government funding but may also charge additional fees
for enrolment. Government schools are funded and managed by the states and
have minimal sources of additional funding through school fees. Catholic schools
are typically run by state Catholic education departments, and independent
schools can include those operated by other religious organisations or secular
organisations, and are typically associated with large school fees to enrol. In
2015, 59% of secondary students across Australia were enrolled in government
schools, 18% in independent schools and 23% in Catholic schools (Australian
Bureau of Statistics, 2016).
Does the reason matter? 151
©2018 British Educational Research Association
Analytic approach
Missing values on the absence behaviours, NAPLAN outcomes and all covariates
were imputed using chained regression multiple imputation techniques in SAS 9.4.
Of the original 4,983 children participating at Wave 1, 3,537 (71%) were still partici-
pating at Wave 6; 3,386 (68%) study children completed the Wave 6 adolescent self-
report; 3,109 (62%) had linked Year 9 NAPLAN numeracy scores; and 2,185 (44%)
had complete information across all analytic variables. All analytic variables were
used to impute missing values, along with auxiliary variables collected at Wave 1 that
were correlated with the likelihood of attrition from the study (Sipthorp & Misson,
2009). Auxiliary variables included if the child had a parent living elsewhere, the fam-
ily received rent assistance or a parenting payment, the family were in rented housing,
maternal education level and the global health rating of the study child. All analyses
were performed on the imputed datasets, and all reported results are the pooled esti-
mates of 30 imputed datasets. Standard errors were based on the combined within-
and between-imputation variance.
The frequencies and distributions of absence behaviours and the reasons provided
for absence are examined and compared across genders. Gender differences in pro-
portions (e.g. the difference in proportions of male and female students with 10 or
more absences) were assessed by differencing the proportions and testing if the differ-
ence score was significantly different from 0 using t-tests.
We then determined how the student, school and family characteristics vary
according to the reasons that adolescents provided for their most recent absence. For
dichotomous characteristics, we used binary logit models to determine if the likeli-
hood of the characteristic (e.g. if the adolescent has a SEWB problem) was signifi-
cantly different for students providing a given reason for absence compared with
students who reported no absences. These results are provided for both parent-per-
mitted and non-permitted absence types.
Linear regression models were then employed to estimate Year 9 numeracy, read-
ing and spelling achievement scores according to the reasons provided for absences
after adjusting for Year 7 achievement and other covariates. For each achievement
outcome, the results for two types of models are provided. The first are partially
adjusted models, which include both parent-permitted and non-permitted reasons
for absence and Year 7 achievement. The fully adjusted models then control for all
other factors and covariates, and are stratified by gender. The 3,109 students with
both self-report measures and linked NAPLAN scores attended 1,254 different
schools, with an average of 2.5 students per school. School-level clustering of errors
was therefore not appropriate for these data.
Results
Absence behaviours
Table 2 provides the frequency of absence behaviours within the previous six months.
Two-thirds of students reported they had been late at least once in the previous six
months, and 11% late 10 or more times. Only 7% said they had skipped class once or
152 K. J. Hancock et al.
©2018 British Educational Research Association
Table 2. Frequency of absence behaviours among 14–15 year olds, by gender
Boys
(N=2,536)
Girls
(N=2,447) p-Value
of difference
All students
(N=4,983)
%% %
Late for school
Never 34.7 32.9 0.277 33.8
1–2 times 34.6 34.2 0.806 34.4
3–6 times 17.8 15.5 0.056 16.6
7–9 times 4.3 3.9 0.609 4.1
10 or more times 10.9 11.2 0.808 11.1
Skipped class
Never 84.3 82.6 0.201 83.5
1–2 times 7.0 7.6 0.483 7.3
3–6 times 4.8 5.4 0.545 5.1
7–9 times 1.4 1.9 0.215 1.7
10 or more times 2.5 2.5 0.987 2.5
Absent without permission
Never 93.3 92.9 0.613 93.1
1–2 times 2.5 2.6 0.862 2.5
3–6 times 1.1 1.0 0.744 1.1
7–9 times 1.0 1.5 0.264 1.2
10 or more times 2.2 2.0 0.635 2.1
Absent with permission
Never 11.7 9.3 0.018 10.5
1–2 times 39.7 40.8 0.497 40.2
3–6 times 29.8 30.4 0.708 30.1
7–9 times 7.9 8.0 0.932 8.0
10 or more times 10.9 11.5 0.566 11.2
Trouble for not following school rules
Never 42.8 57.0 <0.001 49.7
1–2 times 32.3 24.8 <0.001 28.6
3–6 times 10.9 7.3 <0.001 9.2
7–9 times 4.2 2.7 0.019 3.5
10 or more times 9.8 8.1 0.110 9.0
Reasons for non-permitted absence (N=170) (N=175) (N=345)
School or social problems 46.7 45.3 0.836 46.0
Stress, anxiety or depression 11.5 24.3 0.010 18.0
Tiredness 11.7 15.5 0.399 13.6
Other illness or medical condition 9.2 3.7 0.048 6.4
Other reasons 20.9 11.2 0.023 15.9
Reasons for permitted absence (N=2,239) (N=2,219) (N=4,458)
Illness, medical condition
or appointment
56.9 52.7 0.023 54.9
Out-of-school activities 7.8 8.9 0.275 8.3
Family events 6.3 7.3 0.252 6.8
Tiredness 6.1 7.1 0.267 6.6
Stress, anxiety or depression 3.5 6.3 <0.001 4.9
Family health or caring reasons 3.5 3.7 0.741 3.6
School or social problems 5.3 6.2 0.400 5.8
Other reason 10.6 7.9 0.009 9.3
Does the reason matter? 153
©2018 British Educational Research Association
twice, and fewer than 9% had skipped class three or more times. Very few adolescents
reported having any absence without parental permission. Only 4% of adolescents
reported having one or two absences without parental permission, and 4% reported
having three or more absences without parental permission.
Of the 7% of students who reported having a non-permitted absence, the majority
(47% of boys and 45% of girls) indicated school or social reasons as the reason for
their most recent non-permitted absence (Table 2). Twice as many girls as boys
reported a non-permitted absence that was due to stress, anxiety or depression (24%
vs. 12%, p=0.010). In contrast, a higher proportion of boys than girls indicated the
most recent non-permitted absence was due to illness (9% vs. 4%, p=0.048) or
another reason (21% vs. 11%, p=0.023). A further 12% of boys and 16% of girls
indicated that the most recent absence without parental permission was due to
tiredness.
For parent-permitted absences, the majority of both boys and girls indicated that
their most recent absence was due to illness, a medical condition or appointment
(57% of boys and 53% of girls, Table 2). Again, a slightly higher proportion of girls
indicated that the most recent permitted absence was due to stress, anxiety or depres-
sion (6% vs. 4%, p<0.001). The remaining reasons for absence were similar for boys
and girls, including absences due to out-of-school activities (8%), tiredness (7%) or
family events (7%). Very few of the permitted absences were due to school or social
problems (6%), or family health or caring reasons (4%). Together, 45% of the most
recent parent-permitted absences were for reasons other than illness. Of these
absences, 41% could be attributed to parent withdrawal factors (out-of-school activi-
ties, family events or family health), 25% could be attributed to student withdrawal
because of student issues (stress, anxiety or depression, tiredness), 13% for student
withdrawal factors related to school and 21% for other reasons not defined.
Characteristics associated with different absences
Table 3 provides an overview of the student and family characteristics associated with
the reasons provided for parent-permitted absence. Each column corresponds to a
different reason, and for each reason the proportion of students with the given charac-
teristic is provided. For each reason for permitted absence, the difference in propor-
tions of students with the given family, child or school characteristic are compared
with the proportions for those with no absence. For example, among students who
had no parent-permitted absences, 12% had a mother who did not complete Year 12,
13% had a social or emotional wellbeing problem and 4% reported having a non-per-
mitted absence. These values were significantly higher among students who were
absent due to school or social problems, with values of 24% (p<0.05), 72% (p<
0.001) and 61% (p<0.001), respectively.
An important finding to note from Table 3 is that adolescents whose most recent
permitted absence was due to stress, anxiety or depression (64%), school or social
problems (58%) or other reasons (26%) also reported higher levels of absence (i.e.
more of them reported having seven or more permitted absences) compared with those
who were absent due to illness (14%). This would indicate that while only 6% of ado-
lescents said school-related problems were the reason behind their most recent absence,
154 K. J. Hancock et al.
©2018 British Educational Research Association
Table 3. Proportion of 14–15 year olds with given characteristics, by reason for most recent parent-permitted absence
No
absence(ref)
Reason for most recent parent-permitted absence
Illness
Out-of-
school
activities
Family
events Tiredness
Stress,
anxiety or
depression
Family
health/
caring
School or
social
problems Other
%%%%%%%%%
Family factors
Mother did not
complete Year
12
12.4 13.2 12.4 11.6 18.0 18.7 13.0 23.9* 14.8
Father did not
complete Year
12
11.9 13.8 14.0 12.7 20.0 30.6** 18.1 25.4 19.6
Mother low
interest in
education
6.4 3.1 3.1 3.2 9.8 36.4*** 13.7 43.8*** 11.0
Father low
interest in
education
13.3 10.2 8.2 12.6 18.8 62.9*** 27.2* 59.8*** 22.2*
Mother no
expectation of
SC post-school
study
18.1 12.6* 12.0 14.0 25.5 59.9*** 34.8 71.3*** 31.9*
Single-parent
family
11.8 14.5 15.5 13.9 27.4*** 27.7* 20.3 31.9** 25.4***
Child spoke
LOTE at home
(age
4–5)
14.1 13.3 10.4 10.8 11.6 8.5 10.8 14.9 10.5
Does the reason matter? 155
©2018 British Educational Research Association
Table 3. (Continued)
No
absence(ref)
Reason for most recent parent-permitted absence
Illness
Out-of-
school
activities
Family
events Tiredness
Stress,
anxiety or
depression
Family
health/
caring
School or
social
problems Other
%%%%%%%%%
Mother has
psychological
distress
10.6 11.5 9.7 12.0 19.3* 29.6*** 30.0*** 41.9*** 17.0*
Low-income
family
21.3 22.1 22.4 20.2 26.4 37.6* 27.4 45.4** 30.9*
Child factors
Male 56.5 52.1 47.0* 46.5* 46.4* 35.9*** 48.4 46.4* 57.5
Has SEWB
problem
13.3 11.4 8.9 15.7 26.2** 68.9*** 22.6 71.6*** 23.5**
Has disability 6.1 4.5 1.3* 2.8 5.7 35.5*** 11.5 35.5*** 9.5
Ever repeated a
grade
7.7 6.4 5.5 6.2 4.1 9.3 13.1 10.1 8.4
Is frequently late 8.2 8.2 6.0 6.1 15.0 45.9*** 19.5 78.4*** 20.7***
Has skipped
class
9.5 7.7 7.2 8.9 23.3*** 50.6*** 21.0 84.6*** 24.3***
Has had non-
permitted
absence
3.8 2.9 3.8 1.8 7.7* 6.1 2.3 61.2*** 10.0**
Is frequently in
trouble
14.3 13.1 14.1 14.9 26.5** 54.4*** 25.2 81.7*** 32.8***
Has 7+parent-
permitted
absences
NA 14.3 15.0 16.8 21.8* 64.2* 24.7 58.0* 25.6*
156 K. J. Hancock et al.
©2018 British Educational Research Association
Table 3. (Continued)
No
absence(ref)
Reason for most recent parent-permitted absence
Illness
Out-of-
school
activities
Family
events Tiredness
Stress,
anxiety or
depression
Family
health/
caring
School or
social
problems Other
% %%%% % % % %
Attends a
government
school
49.6 47.8 48.5 53.4 54.6 65.7* 61.4 73.7** 58.0*
Reports low
school
acceptance
21.5 18.8 13.7* 19.7 29.2* 57.6*** 34.1* 53.0*** 35.0***
Reports low
school caring
17.8 14.2 9.8** 15.9 20.2 38.2*** 22.9 43.5*** 27.2**
Reports high
rejection
15.9 14.0 14.9 12.3 27.9*** 73.5*** 24.6 52.8*** 23.6*
Has low
academic self-
concept
15.2 15.4 9.7 12.0 24.0* 47.0*** 19.8 41.3*** 26.4**
Has low school
liking
16.2 14.9 14.1 18.5 20.7 41.5*** 29.9* 41.1** 30.5***
Mean values
Grade 7
numeracy (SE)
0.23 (0.05) 0.22 (0.03) 0.32 (0.06) 0.27 (0.07) 0.11 (0.08) 0.23*** (0.10) 0.35*** (0.10) 0.31*** (0.12) 0.09*** (0.06)
Grade 7 reading
(SE)
0.21 (0.06) 0.26 (0.03) 0.36 (0.06) 0.26 (0.07) 0.19 (0.08) 0.06 (0.11) 0.43*** (0.12) 0.23** (0.12) 0.12** (0.07)
Grade 7 spelling
(SE)
0.10 (0.05) 0.13 (0.02) 0.20 (0.06) 0.17 (0.07) 0.10 (0.08) 0.11 (0.10) 0.38*** (0.11) 0.21* (0.12) 0.13** (0.06)
*p<0.05; **p<0.01; ***p<0.001.
Does the reason matter? 157
©2018 British Educational Research Association
these do not necessarily reflect the proportion of all absences that can be attributed to
each reason. As such, the proportion of absences that can be attributed to reasons like
stress, tiredness or school problems is likely underestimated in Table 2.
Table 3 shows that the characteristics of students varied widely according to the
reason for their most recent absence. There were few differences in the characteristics
of students whose absence was due to illness, out-of-school activities, family events or
family health or caring responsibilities as compared with students who had no permit-
ted absences. Compared with students with no absences, students who were absent
for out-of-school activities were less likely to report low school acceptance (14% vs.
22%, p<0.05), low school caring (10% vs. 18%, p<0.01) or a disability (1% vs. 6%,
p<0.05). Students absent for family health or caring responsibilities had higher rates
of low father interest in education (27% vs. 13%, p<0.05), maternal psychological
distress (30% vs. 11%, p<0.001), low school acceptance (34% vs. 32%, p<0.05)
and low school liking (30% vs. 16%, p<0.05). Students whose most recent absence
was due to tiredness were more likely than students with no absences to be in a single-
parent family (27% vs. 12%, p<0.001), have SEWB problem (26% vs. 13%, p<
0.01), have ever skipped class (23% vs. 10%, p<0.001), report low school acceptance
(29% vs. 22%, p<0. 05), high rejection (28% vs. 16%, p<0.001) and low academic
self-concept (24% vs. 15%, p<0.05). Students whose most recent permitted absence
was due to stress, anxiety or depression, school or social problems or other reasons
had substantially and statistically significantly higher levels of almost all of the charac-
teristics examined than students who had no absences.
Table 3 also shows that across all learning domains, the scores of adolescents who
were absent due to illness, out-of-school activities, family events or tiredness were not
statistically different from those who were never absent. Students who were absent
due to stress, anxiety or depression had significantly lower prior numeracy scores in
Year 7 than non-absent students (but not reading or spelling), with the difference
equivalent to nearly half a standard deviation (0.23 vs. 0.23, p<0.001). Adolescents
who were absent for family health or caring responsibilities, school or social problems
or other reasons scored significantly lower on all learning domains than non-absent
students.
Table 4 provides a similar overview, but for absences that were not permitted by
parents. Compared with adolescents who reported no non-permitted absences, those
who had non-permitted absences because of tiredness, stress, anxiety or depression,
or school or social problems had substantially higher rates of other absence beha-
viours like skipping class, had more parent-permitted absences, were in low-income
families and were substantially more likely to be enrolled in a government school.
These adolescents also had significantly higher rates of low school connectedness,
low school caring, low academic self-concept and low school liking.
Reasons for absence and achievement outcomes
Table 5 provides the results of the multivariate regression analysis estimating Year 9
numeracy achievement. The partially adjusted analyses controlling only for prior
numeracy achievement show that compared with boys with no permitted absences,
boys who had an absence due to family health or caring reasons had statistically
158 K. J. Hancock et al.
©2018 British Educational Research Association
Table 4. Proportion of 14–15 year olds with given characteristics, by reason for most recent non-permitted absence
No absence(ref)
Reason for most recent non-permitted absence
Illness Tiredness
Stress, anxiety or
depression
School or
social problems Other
%%%% %%
Family factors
Mother did not
complete Year 12
14.0 15.5 19.1 10.0 22.2 8.1
Father did not complete
Year 12
15.4 14.5 32.0 19.5 24.6 17.2
Mother low interest in
education
6.8 4.5 28.2* 29.3 41.8** 13.6
Father low interest in
education
15.6 9.2 36.3* 44.2* 52.2* 32.4*
Mother no expectation
of SC post-school
study
18.6 37.4* 44.4* 52.7** 77.1*** 30.3*
Single-parent family 16.7 27.9 29.7 21.1 31.2 33.0**
Child spoke LOTE at
home (age 4–5)
12.5 5.3 18.1 11.5 12.9 9.3
Mother has
psychological distress
13.6 28.2 20.4 36.0 43.7*** 26.6*
Low-income family 24.0 21.9 32.8 49.0** 42.0* 27.9
Child factors
Male 51.0 70.6 42.5 30.3* 50.1 64.5*
Has SEWB problem 16.0 18.5 50.4*** 84.3*** 79.5*** 37.9***
Has disability 6.6 4.5 15.8 23.1 65.7*** 9.7
Ever repeated a grade 6.9 9.8 2.6 8.3 11.4 4.1
Is frequently late 11.3 23.1 62.6*** 73.5*** 85.5*** 20.9*
Has skipped class 11.4 43.2*** 80.2*** 86.0*** 97.4*** 64.0***
Is frequently in trouble 23.0 47.0*** 67.0*** 73.7*** 91.1*** 52.7***
Does the reason matter? 159
©2018 British Educational Research Association
Table 4. (Continued)
No absence(ref)
Reason for most recent non-permitted absence
Illness Tiredness
Stress, anxiety or
depression
School or
social problems Other
%%%% %%
Has 7+parent-
permitted absences
17.3 18.2 39.7* 39.2* 59.7*** 24.7
Attends a government
school
50.2 75.4* 63.7 80.3*** 84.6*** 75.2***
Reports low school
acceptance
22.7 37.1 47.2** 54.2*** 54.6*** 48.7***
Reports low school
caring
16.7 31.8 44.4*** 45.3*** 49.5*** 38.8***
Reports high rejection 18.8 18.1 41.6*** 63.7*** 49.3*** 38.5***
Has low academic self-
concept
17.2 23.4 42.3** 43.3** 46.7*** 43.5***
Has low school liking 18.5 23.7 36.4* 24.8 44.8** 39.1***
Mean values
Grade 7 numeracy (SE) 0.17 0.41* (0.23) 0.07 (0.18) 0.25 (0.19) 0.25*** (0.20) 0.26** (0.15)
Grade 7 reading (SE) 0.20 0.34* (0.23) 0.09 (0.19) 0.24* (0.19) 0.41*** (0.14) 0.20** (0.15)
Grade 7 spelling (SE) 0.09 0.49** (0.22) 0.06 (0.18) 0.12 (0.20) 0.36** (0.14) 0.25* (0.15)
*p<0.05; **p<0.01; ***p<0.001.
160 K. J. Hancock et al.
©2018 British Educational Research Association
significantly lower numeracy scores in Year 9 (0.26, p=0.004), as did those absent
for school or social problems (0.23, p=0.018), stress, anxiety or depression
(0.23, p=0.021) or another reason (0.14, p=0.009). Boys who had a non-per-
mitted absence for illness also scored 0.32 standard deviations lower on numeracy
than those who did not have any non-permitted absences (p=0.026). After adjusting
for family and school factors, the only absence reasons that remained statistically sig-
nificant in their associations with poor numeracy were those for family health or car-
ing reasons (0.22, p=0.013) or another reason (0.10, p=0.047), with school or
social problems marginally significant (0.20, p=0.057). Among girls, the fully
adjusted analyses indicated that girls with permitted absences scored lower on numer-
acy if the reason was illness (0.10, p=0.026) or another reason (0.13, p=0.040),
or if they had a non-permitted absence due to stress, anxiety or depression (0.31,
p=0.009).
The results for reading achievement (Table 6) showed that for boys, the only par-
ent-permitted reason for absence that was associated with lower reading achievement
was stress, anxiety or depression (0.30, p=0.014), though this association was not
significant after adjusting for family and school factors. Among girls, the partially
adjusted analyses showed lower reading scores for those absent for family health or
caring reasons (0.28, p=0.020), school or social problems (0.33, p=0.004),
stress, anxiety or depression (0.23, p=0.044) and tiredness (0.17, p=0.030).
After adjusting for family and school factors, the only statistically signification associ-
ation with reading achievement was absences due to family health or caring reasons
(0.23, p=0.032). Few of the family and school-related factors were associated with
lower reading achievement in Year 9 for girls, with the exception of low maternal
expectations for post-school study (0.17, p=0.003) and low school liking (0.10,
p=0.014). More of these factors were relevant for boys, including if their father had
not completed Year 12 (0.10, p=0.043), no maternal expectation for post-school
study (0.26, p<0.001), the adolescent was frequently in trouble (0.09, p=
0.041), had low school liking (0.13, p=0.010) or if they attended an independent
school relative to a government school (0.08, p=0.043).
Finally, Table 7 shows that after full adjustment, boys who were absent for illness,
a medical condition or appointment had significantly lower spelling scores than those
never absent (0.08, p=0.042), as did girls who were absent for family health or car-
ing reasons (0.19, p=0.033). Similar to the results for reading achievement, no sig-
nificant differences in spelling were observed for the different reasons for non-
permitted absence. Among the covariates, adolescents whose mother had no expecta-
tion for the adolescent to undertake any post-school study also had significantly lower
spelling scores (boys: 0.14, p<0.001; girls: 0.13, p<0.001), a similar result to
those observed for numeracy and reading achievement scores.
Discussion
To date, research examining the different reasons that students have for missing
school has been limited. This is problematic, because the success of interventions
aimed at addressing absenteeism may depend on how well they address the underly-
ing reasons that students have for missing school. This study set out to provide an
Does the reason matter? 161
©2018 British Educational Research Association
Table 5. Models estimating numeracy achievement according to reasons for most recent absence
Boys Girls
Partially adjusted Fully adjusted Partially adjusted Fully adjusted
Est. p-Value Est. p-Value Est. p-Value Est. p-Value
Intercept 0.06 0.073 0.12 0.004 0.10 0.017 0.14 0.002
Reason for most recent permitted absence
Never absent Ref Ref Ref Ref
Family events 0.09 0.117 0.08 0.200 0.06 0.345 0.05 0.380
Family health or caring reasons 0.26 0.004 0.22 0.013 0.20 0.016 0.15 0.065
Illness, medical condition or appointment 0.05 0.193 0.05 0.178 0.09 0.029 0.10 0.026
Outside of school activities 0.01 0.866 0.02 0.671 0.08 0.129 0.08 0.110
School or social problems 0.23 0.018 0.20 0.057 0.17 0.062 0.08 0.350
Stress, anxiety or depression 0.23 0.021 0.08 0.369 0.25 0.001 0.13 0.066
Tiredness 0.10 0.154 0.05 0.418 0.13 0.044 0.10 0.120
Other reason 0.14 0.009 0.10 0.047 0.17 0.004 0.13 0.040
Reason for most recent non-permitted absence
Never absent Ref Ref Ref Ref
Illness 0.32 0.026 0.27 0.067 0.13 0.617 0.11 0.644
Tiredness 0.10 0.547 0.08 0.617 0.12 0.430 0.06 0.669
Stress, anxiety or depression 0.10 0.574 0.08 0.708 0.28 0.023 0.31 0.009
School or social problems 0.05 0.675 0.00 0.987 0.14 0.174 0.10 0.428
Other 0.04 0.720 0.01 0.896 0.10 0.493 0.04 0.784
Year 7 numeracy z-score 0.86 <0.001 0.81 <0.001 0.87 <0.001 0.82 <0.001
Age (years) 0.00 0.916 0.01 0.780
Mother did not complete Year 12 0.08 0.035 0.08 0.033
Father did not complete Year 12 0.08 0.056 0.02 0.582
Mother low interest in education 0.06 0.406 0.03 0.691
Father low interest in education 0.09 0.059 0.05 0.209
Mother no expectation of
SC post-school study
0.11 0.004 0.11 0.011
162 K. J. Hancock et al.
©2018 British Educational Research Association
Table 5. (Continued)
Boys Girls
Partially adjusted Fully adjusted Partially adjusted Fully adjusted
Est. p-Value Est. p-Value Est. p-Value Est. p-Value
Single-parent family 0.00 0.987 0.00 0.894
LOTE spoken in home 0.06 0.106 0.14 <0.001
Mother has psychological distress 0.01 0.713 0.06 0.133
Low-income family 0.03 0.405 0.05 0.082
Has SEWB problem 0.09 0.031 0.05 0.170
Has disability 0.04 0.477 0.03 0.678
Ever repeated a grade 0.06 0.243 0.07 0.257
Is frequently late 0.02 0.640 0.03 0.565
Has skipped class 0.05 0.283 0.03 0.511
Is frequently in trouble 0.05 0.107 0.02 0.706
Attends a Catholic school
(vs. government)
0.03 0.263 0.01 0.821
Attends an independent
school (vs. government)
0.11 <0.001 0.08 0.008
Reports low school acceptance 0.01 0.857 0.02 0.491
Reports low school caring 0.00 0.904 0.01 0.796
Reports high rejection 0.08 0.054 0.07 0.044
Has low academic self-concept 0.09 0.014 0.03 0.390
Has low school liking 0.09 0.019 0.11 0.004
Estimates with a p-value less than 0.05 are highlighted in bold.
Does the reason matter? 163
©2018 British Educational Research Association
Table 6. Models estimating reading achievement according to reasons for most recent absence
Boys Girls
Partially adjusted Fully adjusted Partially adjusted Fully adjusted
Est. p-Value Est. p-Value Est. p-Value Est. p-Value
Intercept 0.03 0.572 0.09 0.091 0.12 0.022 0.23 <0.001
Reason for most recent permitted absence
Never absent Ref Ref Ref Ref
Family events 0.08 0.333 0.08 0.332 0.08 0.264 0.08 0.246
Family health or caring reasons 0.24 0.099 0.16 0.243 0.28 0.020 0.23 0.032
Illness, medical condition or appointment 0.01 0.885 0.00 0.936 0.03 0.639 0.04 0.439
Outside of school activities 0.02 0.786 0.03 0.634 0.02 0.827 0.03 0.684
School or social problems 0.11 0.388 0.09 0.526 0.33 0.004 0.19 0.118
Stress, anxiety or depression 0.30 0.014 0.01 0.965 0.23 0.044 0.03 0.748
Tiredness 0.09 0.274 0.01 0.916 0.17 0.030 0.14 0.065
Other reason 0.03 0.726 0.05 0.475 0.05 0.511 0.03 0.682
Reason for most recent non-permitted absence
Never absent Ref Ref Ref Ref
Illness 0.32 0.098 0.15 0.406 0.02 0.950 0.02 0.956
Tiredness 0.09 0.631 0.15 0.444 0.06 0.705 0.09 0.614
Stress, anxiety or depression 0.05 0.847 0.26 0.340 0.07 0.778 0.02 0.884
School or social problems 0.14 0.338 0.02 0.914 0.02 0.894 0.03 0.849
Other 0.00 0.997 0.13 0.310 0.04 0.805 0.06 0.725
Year 7 reading z-score 0.83 <0.001 0.78 <0.001 0.81 <0.001 0.76 <0.001
Age (years) 0.04 0.239 0.02 0.577
Mother did not complete Year 12 0.06 0.257 0.09 0.069
Father did not complete Year 12 0.10 0.043 0.09 0.057
Mother low interest in education 0.01 0.904 0.06 0.401
Father low interest in education 0.01 0.831 0.09 0.057
Mother no expectation for post-school study 0.26 <0.001 0.17 0.003
Single-parent family 0.01 0.901 0.02 0.704
164 K. J. Hancock et al.
©2018 British Educational Research Association
Table 6. (Continued)
Boys Girls
Partially adjusted Fully adjusted Partially adjusted Fully adjusted
Est. p-Value Est. p-Value Est. p-Value Est. p-Value
LOTE spoken in home 0.08 0.083 0.08 0.114
Mother has psychological distress 0.01 0.875 0.02 0.664
Low-income family 0.05 0.258 0.06 0.105
Has SEWB problem 0.08 0.165 0.01 0.744
Has disability 0.02 0.820 0.04 0.583
Ever repeated a grade 0.04 0.597 0.07 0.350
Is frequently late 0.02 0.792 0.03 0.559
Has skipped class 0.12 0.061 0.05 0.387
Is frequently in trouble 0.09 0.041 0.06 0.254
Attends a Catholic school (vs. government) 0.04 0.336 0.02 0.495
Attends an independent school (vs. government) 0.08 0.043 0.05 0.170
Reports low school acceptance 0.05 0.307 0.04 0.373
Reports low school caring 0.02 0.688 0.06 0.227
Reports high rejection 0.08 0.107 0.06 0.131
Has low academic self-concept 0.05 0.237 0.02 0.715
Has low school liking 0.13 0.010 0.10 0.014
Estimates with a p-value less than 0.05 are highlighted in bold.
Does the reason matter? 165
©2018 British Educational Research Association
Table 7. Models estimating spelling achievement according to reasons for most recent absence
Boys Girls
Partially adjusted Fully adjusted Partially adjusted Fully adjusted
Est. p-Value Est. p-Value Est. p-Value Est. p-Value
Intercept 0.03 0.316 0.10 0.019 0.08 0.033 0.14 <0.001
Reason for most recent permitted absence
Never absent Ref Ref Ref Ref
Family events 0.04 0.473 0.05 0.371 0.01 0.835 0.01 0.882
Family health or caring reasons 0.20 0.039 0.18 0.056 0.23 0.014 0.19 0.033
Illness, medical condition or appointment 0.07 0.054 0.08 0.042 0.05 0.186 0.01 0.176
Outside of school activities 0.04 0.509 0.05 0.329 0.03 0.605 0.03 0.611
School or social problems 0.19 0.069 0.12 0.294 0.13 0.095 0.04 0.606
Stress, anxiety or depression 0.13 0.111 0.01 0.884 0.14 0.061 0.04 0.526
Tiredness 0.05 0.450 0.03 0.694 0.09 0.122 0.05 0.382
Other reason 0.09 0.091 0.07 0.207 0.04 0.486 0.00 0.994
Reason for most recent non-permitted absence
Never absent Ref Ref Ref Ref
Illness 0.14 0.328 0.07 0.605 0.05 0.835 0.02 0.935
Tiredness 0.01 0.971 0.02 0.903 0.10 0.420 0.02 0.856
Stress, anxiety or depression 0.02 0.910 0.06 0.781 0.18 0.111 0.18 0.117
School or social problems 0.10 0.407 0.24 0.067 0.03 0.818 0.11 0.359
Other 0.06 0.547 0.11 0.287 0.13 0.360 0.23 0.112
Year 7 spelling z-score 0.89 <0.001 0.86 <0.001 0.87 <0.001 0.84 <0.001
Age (years) 0.05 0.092 0.06 0.042
Mother did not complete Year 12 0.00 0.911 0.04 0.248
Father did not complete Year 12 0.06 0.155 0.01 0.736
Mother low interest in education 0.09 0.174 0.06 0.309
Father low interest in education 0.01 0.806 0.01 0.761
Mother no expectation for post-school study 0.14 <0.001 0.13 0.001
Single-parent family 0.04 0.331 0.02 0.566
166 K. J. Hancock et al.
©2018 British Educational Research Association
Table 7. (Continued)
Boys Girls
Partially adjusted Fully adjusted Partially adjusted Fully adjusted
Est. p-Value Est. p-Value Est. p-Value Est. p-Value
LOTE spoken in home 0.07 0.067 0.11 0.015
Mother has psychological distress 0.01 0.872 0.04 0.316
Low-income family 0.05 0.145 0.02 0.622
Has SEWB problem 0.00 0.943 0.02 0.585
Has disability 0.07 0.332 0.01 0.906
Ever repeated a grade 0.02 0.711 0.01 0.901
Is frequently late 0.05 0.319 0.00 0.961
Has skipped class 0.08 0.067 0.07 0.147
Is frequently in trouble 0.02 0.489 0.01 0.770
Attends a Catholic school (vs. government) 0.02 0.564 0.04 0.153
Attends an Independent school (vs. government) 0.00 0.883 0.04 0.224
Reports low school acceptance 0.04 0.294 0.00 0.965
Reports low school caring 0.02 0.554 0.03 0.408
Reports high rejection 0.04 0.323 0.00 0.993
Has low academic self-concept 0.11 0.002 0.05 0.132
Has low school liking 0.05 0.123 0.10 0.002
Estimates with a p-value less than 0.05 are highlighted in bold.
Does the reason matter? 167
©2018 British Educational Research Association
assessment of the drivers, contexts and potential consequences of different types of
absence. We examined the broader range of self-reported reasons students have for
missing school, the extent to which parents are aware of these absences, the student,
family and school contexts associated with the different reasons, and how these rea-
sons are differentially associated with achievement outcomes at school.
There are three features of this study that merit comment. First, according to the
self-report of adolescents, most parents are aware of and sanction the absences of
their 14–15 year old children. Only 7% of the study adolescents reported an absence
in the previous six months that their parent had not permitted. While this figure
undoubtedly underestimates the total proportion of absences that are sanctioned by
parents, this result suggests that parental involvement is imperative for any interven-
tion aimed at improving school attendance.
The second key finding is that while schools bear the responsibility for monitoring
and responding to absenteeism, many absences among 14–15 year olds may result
from factors that schools cannot realistically address without assistance from families
and support services. Over half of adolescents indicated that their most recent
absence was for illness, which is consistent with other research (Rothman, 2002). Of
the remaining absences, 41% of students attributed their most recent absence to par-
ent-related factors like family events, other out-of-school activities or family caring
responsibilities. A quarter of the remaining students said that their absence was due
to personal issues like tiredness, stress or mental health concerns, and 13% indicated
that their most recent absence was related to a school-based factor like problems with
teachers, bullying or peer problems. These findings suggest that while school may not
be the root cause of absenteeism for many students, where schools may be able to
make a difference is by supporting students and their families to deal with problems
when they arise. A programme piloted in New York City, for example, successfully
reduced chronic absenteeism in trial schools by launching a collaborative effort across
government agencies to provide cross-sectoral support to chronically absent students
(Balfanz & Byrnes, 2013). The programme was developed in recognition of the multi-
ple and diverse drivers of absence that schools cannot address alone. While not
focused specifically on absenteeism, the Extended Schools programme in the UK is
another example of schools providing multiple supports to disadvantaged students
and communities. The programme funds schools to facilitate partnerships across
multiple organisations that then provide support services to assist student learning
and development outcomes, including homework clubs, study support, sports and
parenting programmes.
Third, our findings support the results of previous studies suggesting that not all
absences are equal when it comes to student outcomes. For boys, absences for family
health or caring reasons were associated with lower numeracy and spelling achieve-
ment, absences due to illness were associated with lower spelling scores, and ‘other’
absences associated with lower numeracy scores. For girls, absences for illness or
stress, anxiety or depression were associated with lower numeracy scores, absences
for family health and caring reasons were associated with lower reading and spelling
scores. In contrast, adolescents who reported permitted absences due to family events
or out-of-school activities achieved similar levels to students who were never absent.
This finding might signal to schools that they need not be too concerned about family
168 K. J. Hancock et al.
©2018 British Educational Research Association
events like holidays. However, the lack of an association between these family-related
absences and achievement outcomes (or student and family characteristics) could
also reflect a practice among families to ensure that any such absences are made up
for. Some families may also only opt to take holidays during school time if they think
the student is doing well enough, or is otherwise capable of making up missed work.
In this case, rather than suggesting that schools turn a blind eye to students who are
absent for family events, they should continue to engage with families to ensure that
students can handle the absence.
Similarly, although absences for family events may not be cause for great concern,
our results also showed that girls whose most recent permitted absence was for illness
also had lower numeracy scores, and similarly for boys’ spelling achievement. This
finding highlights that even for a typical excused absence like illness, there is still a
negative association with learning outcomes. This could reflect a lack of response
from schools and families, who view illness as ‘normal’ and something that affects the
majority of students at some point. The acceptance of these absences as normal may
mean that schools do not consider it necessary to help students catch up on school
work when they have been ill.
Non-permitted absences (except tiredness) were not associated with Year 9
achievement. This may be because these absences were strongly associated with fam-
ily background factors, which would be accounted for by controlling for prior
achievement. It may also be a result of small cell sizes for each non-permitted reason,
given the very small proportion of adolescents that had any non-permitted absences.
One notable exception was girls with non-permitted absences for stress, anxiety or
depression, who achieved 0.3 of a standard deviation lower on numeracy, indepen-
dently of prior numeracy achievement or other covariates. That girls would be having
absences for stress, anxiety and depression without parental permission may indicate
that parents are not aware of any psychological distress experienced by their daugh-
ters (and to a lesser extent their sons), or that psychological distress is keeping some
girls out of school—to the detriment of their learning. This pattern is consistent with
other research showing that self-reports of anxiety and depression among adolescent
girls identify a much higher prevalence of these problems than when the prevalence is
based on parent report alone (Lawrence et al., 2015).
A clear implication of this study is that addressing absenteeism requires a dual
approach of preventing avoidable absence and mitigation strategies for when either
avoidable or unavoidable absences occur. In terms of prevention, schools certainly
have a role to play in ensuring that schools are safe and engaging places to be, how-
ever, there needs to be better recognition that many absences result from circum-
stances unrelated to the school. For any improvement to occur, schools need to have
better information about why their students are absent from school. This may be
facilitated by improved parent engagement and positive, supportive relationships with
both parents and students. Schools may also need to adopt more advanced methods
of recording and monitoring absences and make this information readily accessible to
schools, teachers and support staff, so they can track absences when they first occur
and intervene earlier. To this end, improved systems that provide more detailed infor-
mation about the reasons why students are absent—rather than just excused or unex-
cused—may assist schools to identify students who need greater support.
Does the reason matter? 169
©2018 British Educational Research Association
There also needs to be better recognition that many absences are unavoidable, and
in such cases some effort is needed to reduce the potential impacts of an absence (i.e.
mitigation). This might be achieved by communicating to parents the importance of
catching up so that students do not fall behind, and parents can then support the stu-
dent to do so. In schools where parent engagement or capability is low, additional
school-based learning support is warranted. This may include improving awareness
among teachers that students may not be ‘at fault’ for missing school. In highly disad-
vantaged schools where chronic absenteeism is highly disruptive, additional educa-
tional support like teacher aides, education support officers or after-school tutoring
may help to provide catch-up services to students. Of course, for these services to be
effective, students also need to be present at school. Thus, both prevention and miti-
gation approaches are needed. Ideally, schools should be linked to a range of support
services that help students and their families address the factors preventing school
attendance.
Limitations
There are several limitations to the study. One concern is whether adolescents from
highly disadvantaged backgrounds are adequately captured in the survey. While gen-
erally considered broadly representative, over time the LSAC has become more
biased to include more advantaged families, meaning that the absence behaviours
reported in this study may be underestimated. We attempted to account for this using
multiple imputation to ensure complete case data and avoid attrition and response
bias.
Students were only asked about the number of permitted and non-permitted
absences they had in the last six months, with responses provided in categories. We
found that approximately 11% of students had been absent 10 or more times, with
permission, which would correspond to an absence rate of approximately 10% for a
semester. Other Australian data suggests that 40% of Year 9 students have absence
rates greater than 10% (Hancock et al., 2013). While some degree of recall bias and
underestimation would be expected when asking young people to self-report how
many days they had been absent in the previous six months, these data nonetheless
suggest that the frequency of absence in this study was substantially underestimated.
Moreover, students only report the main reason for their most recent absence, and
not the frequency of absences due to each reason. As such, we could not determine
the total number of absences due to a particular reason.
While this study provides insights as to the different types of factors contributing to
student absence, there is still a substantial gap in the literature that could be ade-
quately addressed by the availability of population-level administrative data recording
more specific reasons for absence than whether or not absences were excused or
unexcused. More detailed data would allow schools to be more aware of the absence
problems their students are encountering, and look to specific solutions that meet
particular student needs. Additionally, these data could then be linked to achieve-
ment and enrolment records to develop a better understanding of absence problems,
the risks that particular types of absences pose to students and which problems may
be of greater relevance to schools, assisting educators to better understand attendance
170 K. J. Hancock et al.
©2018 British Educational Research Association
problems and the resources that schools need to deal with problem absence
behaviour.
Avenues for future research include expanding this type of analysis to younger stu-
dents to identify the prevalence of different reasons (and therefore different
approaches) for missing school. As attendance ‘careers’ commence upon entry into
school (Hancock et al., 2013), early intervention strategies are necessary to ensure
that young children and their families develop good attendance patterns at the outset,
which continue on throughout the school years. Early intervention may prevent
chronic absenteeism and other drivers of absence, such as mental health problems or
disengagement, from developing further in adolescence.
Conclusion
This study found that most parents are aware of and sanction the absences of their
14–15 year old children. The majority of absences relate to family and student factors,
suggesting that even though schools typically bear the responsibility for monitoring
and responding to absenteeism, the drivers of absences among 14–15 year olds may
not be factors that schools can realistically address alone. This study also showed that
only some reasons for absence (e.g. family caring responsibilities, illness, stress, anxi-
ety or depression) were problematic for achievement outcomes. Given that these
absences will be both avoidable and unavoidable, addressing absenteeism requires a
dual approach of preventing avoidable absence and mitigation strategies for when
either avoidable or unavoidable absences occur. Schools require assistance from fami-
lies, communities and support organisations to enable both prevention and mitigation
strategies.
Acknowledgements
This study was supported by the Australian Research Council Centre of Excellence
for Children and Families over the Life Course (CE140100027). The Centre is
administered by the University of Queensland, with nodes at the University of
Western Australia, The University of Melbourne and The University of Sydney.
This study used data from the Longitudinal Study of Australian Children
(LSAC). The LSAC was conducted in partnership between the Department of
Social Services (DSS), the Australian Institute of Family Studies (AIFS) and the
Australian Bureau of Statistics (ABS). The datasets used in the study are available
to researchers upon application to the DSS, and are subject to licensing conditions
and user agreements. The authors are grateful for the provision of a travel grant
awarded by the Life Course Centre that allowed the establishment of the collabora-
tion between the authors.
References
Alexander, K. L., Entwisle, D. R. & Horsey, C. S. (1997) From first grade forward: Early founda-
tions of high school dropout, Sociology of Education, 70(2), 87–107.
Does the reason matter? 171
©2018 British Educational Research Association
Aucejo, E. M. & Romano, T. F. (2014) Assessing the effect of school days and absences on test score per-
formance (London, London School of Economics and Political Science).
Australian Bureau of Statistics (2016) Schools, Australia, 2015. Available online at: www.abs.gov.au/
ausstats/abs@.nsf/mf/4221.0 (accessed 28 December 2017).
Australian Institute of Health and Welfare (2005) Selected chronic diseases among Australia’s children.
Available online at: http://webarchive.nla.gov.au/gov/20170816101132/ http://www.aihw.gov.
au/publication-detail/?id=6442467756 (accessed 1 August 2017)
Baker, J. A. (2006) Contributions of teacher–child relationships to positive school adjustment dur-
ing elementary school, Journal of School Psychology, 44, 211–229.
Balfanz, R. & Byrnes, V. (2013) Meeting the challenge of combating chronic absenteeism: Impact of the
NYC mayor’s interagency task force on chronic absenteeism and school attendance and its implications
for other cities (Baltimore, MD, Johns Hopkins School of Education).
Booker, K. C. (2007) Likeness, comfort and tolerance: Examining African American adolescents’
sense of school belonging, The Urban Review, 39, 301–317.
Chang, H. N. & Romero, M. (2008) Present, engaged and accounted for: The critical importance
of addressing chronic absence in the early grades (New York, National Center for Children in
Poverty).
Darr, C. W. (2012) Measuring student engagement: The development of a scale for formative use,
in: D. Christensen, A. L. Reschly & C. Wylie (Eds) Handbook of research on student engagement
(London, Springer), 707–724.
deJung, J. E. & Duckworth, K. (1986) Measuring student absences in the high schools, paper pre-
sented at the AERA Annual Meeting, San Francisco, CA, 16–20 April. Available at https://files.
eric.ed.gov/fulltext/ED271889.pdf (accessed 28 December 2017).
Department for Education (2011) A profile of pupil absence in England (London, Department for
Education).
Department of Education, Employment and Workplace Relations (2014) Improving school enrolment
and attendance through welfare reform measure (SEAM) trial (2009–2012) (Canberra, Deparment
of Education, Employment and Workplace Relations).
Finkelhor, D., Turner, H. A., Shattuck, A. & Hamby, S. K. (2015) Prevalence of childhood expo-
sure to violence, crime and abuse: Results from the National Survey of Children’s Exposure to
Violence, JAMA Pediatrics, 169(8), 746–754.
Furukawa, T. A., Kessler, R. C., Slade, T. & Andrews, G. (2003) The performance of the K6 and
K10 screening scales for psychological distress in the Australian National Survey of Mental
Health and Wellbeing, Psychological Medicine, 33, 357–362.
Gershenson, S., Jacknowitz, A. & Brannegan, A. (2017) Are student absences worth the worry in
U.S., primary schools? Education Finance and Policy, 12, 137–165.
Goodenow, C. (1993) The psychological sense of school membership among adolescents: Scale
development and educational correlates, Psychology in the Schools, 30, 79–90.
Goodman, R. (2001) Psychometric properties of the Strengths and Difficulties Questionnaire
(SDQ), Journal of the American Academy of Child and Adolescent Psychiatry, 40, 1337–1345.
Goodman, R., Renfrew, D. & Mullick, M. (2000) Predicting type of psychiatric disorder from
Strengths and Difficulties Questionnaire (SDQ) scores in child mental health clinics in London
and Dhaka, European Child and Adolescent Psychiatry, 9, 129–134.
Gottfried, M. A. (2009) Excused versus unexcused: How student absences in elementary school
affect academic achievement, Educational Evaluation and Policy, 31, 392–415.
Hancock, K. J., Shepherd, C. C. J., Lawrence, D. & Zubrick, S. R. (2013) Student attendance and
educational outcomes: Every day counts. Available online at: www.telethonkids.org.au/globalasse
ts/media/documents/research-topics/student-attendance-and-educational-outcomes-2015.pdf
(accessed 12 November 2017).
Hancock, K. J., Lawrence, D., Shepherd, C. C. J., Mitrou, F. & Zubrick, S. R. (2017) Associations
between school absence and academic achievement: Do socioeconomics matter?, British Edu-
cational Research Journal, 43, 415–440.
Houng, B. & Justman, M. (2014) NAPLAN scores as predictors of access to higher education in Victoria.
Melbourne Institute Working Paper Series, 22/14. Available online at: melbourneinstitute.
172 K. J. Hancock et al.
©2018 British Educational Research Association
unimelb.edu.au/downloads/working_paper_series/wp2014n22.pdf (accessed 12 November
2017).
Hughes, J. N., Zhang, D. & Hill, C. R. (2006) Peer assessments of normative and individual
teacher–student support predict social acceptance and engagement among low-achieving chil-
dren, Journal of School Psychology, 43, 447–463.
Ibanez, G. E., Kuperminc, G. P., Jurkovic, G. & Perilla, J. (2004) Cultural attributes and adapta-
tions linked to achievement and motivation among Latino adolescents, Journal of Youth and
Adolescence, 21, 1–26.
Kearney, C. A. (2008) School absenteeism and school refusal behavior in youth: A contemporary
review, Clinical Psychology Review, 28(3), 451–471.
Kessler, R. C., Andrews, G., Colpe, L. J., Hiripi, E., Mrocrek, D. K. & Normand, S. L. (2002)
Short screening scales to monitor population prevalence and trends in non-specific psychologi-
cal distress, Psychological Medicine, 32, 959–976.
Kessler, R. C., Barker, P. R., Colpe, L. J., Epstein, J. F., Gfroerer, B. A., Hiripi, E. & Zaslavsky, A.
M. (2003) Screening for serious mental illness in the general population, Archives of General
Psychiatry, 60, 184–189.
Kessler, R. C., Avenevoli, S., Costello, J., Georgiadis, K., Greif Green, J., Gruber, M. J., ...
Merikangas K. R. (2012) Prevalence, persistence, and sociodemographic correlates of DSM-
IV disorders in the National Comorbidity Survey Replication Adolescent Supplement, Archives
of General Psychiatry, 69(4), 372–380.
Lawrence, D., Johnson, S. E., Hafekost, J., de Boterhoven Haan, K., Sawyer, M., Ainley, J. &
Zubrick, S. R. (2015) The mental health of children and adolescents. Report on the second Aus-
tralian Child and Adolescent Survey of Mental Health and Wellbeing (Canberra, Department
of Health).
Malcolm, H., Wilson, V., Davidson, J. & Kirk, S. (2003) Absence from school: A study of its causes and
effects in seven LEAs. Research Report RR424.
Messias, E., Kindrick, K. & Castro, J. (2014) School bullying, cyberbullying, or both: Correlates of
teen suicidality in the 2011 CDC youth risk behavior survey, Comprehensive Psychiatry, 55(5),
1063–1068.
Neild, R. C. & Balfanz, R. (2006) An extreme degree of difficulty: The educational demo-
graphics of urban neighborhood schools, Journal of Education for Students Placed at Risk,
11, 123–141.
Office of the California Attorney General (2015) In school + on track: Attorney General’s 2015 report
on California’s elementary school truancy & absenteeism crisis. Available online at: oag.ca.gov/sites/
all/files/agweb/pdfs/tr/truancy_2015.pdf (accessed 12 November 2017).
Prochaska, J. J., Sung, H. Y., Max, W., Shi, Y. & Ong, M. (2012) Validity study of the K6 scale as a
measure of moderate mental distress based on mental health treatment need and utilization,
International Journal of Methods in Psychiatric Research, 21, 88–97.
Rothman, S. (2002) Student absence in South Australian schools, The Australian Educational
Researcher, 29(1), 69–91.
Rumberger, R. W. (1995) Dropping out of middle school: A multilevel analysis of students and
schools, American Educational Research Journal, 32, 583–625.
Sanchez, B., Colon, Y. & Esparza, P. (2005) The role of sense of school belonging and gender in
the academic adjustment of Latino adolescents, Journal of Youth and Adolescence, 34(6), 619–
628.
Shaw, S. R. & McCabe, P. C. (2007) Hospital-to-school transition for children with chronic illness:
Meeting the new challenges of an evolving health care system, Psychology in the Schools, 45(1),
74–87.
Silverman, R. M. (2012) Making waves or treading water? An analysis of charter schools in New
York State, Urban Education, 48, 257–288.
Sipthorp, M. & Misson, S. (2009) Wave 3 weighting and non-response (Melbourne, Australian
Institute of Family Studies).
Soloff, C., Lawrence, D. & Johnstone, R. (2005) Sample design. LSAC Technical Paper No. 1
(Melbourne, Australian Institute of Family Studies).
Does the reason matter? 173
©2018 British Educational Research Association
Soloff, C., Lawrence, D., Misson, S., Johnstone, R. & Slater, J. (2006) Wave 1 weighting and non-
response (Melbourne, Australian Institute of Family Studies).
Tobin, K. J. (2014) Homeless students and academic achievement: Evidence from a large urban
area, Urban Education, 51, 197–220.
Wylie, C. & Hodgen, E. (2012) Trajectories and patterns of student engagement: Evidence from a
longitudinal study, in: S. L. Christenson, A. L. Reschly & C. Wylie (Eds) Handbook of research
on student engagement (London, Springer), 585–600.
You, S., Ritchey, K. M., Furlong, M. J., Shochet, I. M. & Boman, P. (2011) Examination of the
latent structure of the Psychological Sense of School Membership Scale, Journal of Psychoedu-
cational Assessment, 29, 225–237.
Yusuf, H. & Bowcott, O. (2017, October 31) Almost 20,000 people prosecuted for pupil absences
in 2015, The Guardian. Available online at: www.theguardian.com/education/2017/apr/05/al
most-20000-people-prosecuted-for-pupil-absences-in-2015.
Zahn-Waxler, C., Shirtcliff, E. A. & Marceau, K. (2008) Disorders of childhood and adolescence:
Gender and psychopathology, Annual Review of Clinical Psychology, 4, 275–303.
©2018 British Educational Research Association
174 K. J. Hancock et al.
