Journal of Further and Higher Education
Vol. 28, No. 1, February 2004
Student Debt and its Relation to
Student Mental Health
RichardCookePsychological Therapies Research CentreUniversity of LeedsUKr.firstname.lastname@example.org
RICHARD COOKE,MICHAEL BARKHAM,KERRY AUDIN,
Psychological Therapies Research Centre, School of Psychology, University of Leeds,
17 Blenheim Terrace, Leeds LS2 9JT, UK. Email: email@example.com
School of Computing, University of Leeds, UK
ABSTRACT The present paper provides an analysis of the relationship between attitudes
toward debt and mental health among university undergraduates. Data were collected from
the same cohort of students across their three years of university, with responses from 2146,
1360 and 1391 ﬁrst, second and third year students, respectively. Mental health was
measured using the General Population version of the Clinical Outcomes in Routine
Evaluation (GP-CORE). Attitudes toward debt were measured using items that tapped
current ﬁnancial concerns and worry about debt on leaving university. Results showed that
students become more concerned about their ﬁnances as they progress through university,
that there was no relationship between anticipated debt and mental health and that
attitudes toward debt were related to mental health levels. Students who were identiﬁed as
having high ﬁnancial concerns possessed signiﬁcantly worse CORE-GP scores than students
with low ﬁnancial concern in all three years of university. In all three years students with
high ﬁnancial concerns felt more ‘tense, anxious or nervous’, more ‘criticised by other
people’ and found it more ‘difﬁcult getting to sleep or staying asleep’ than students with low
ﬁnancial concerns. There was also evidence that students with high worry about their debt
anticipated leaving university with higher amounts of debt than low debt worry students.
These ﬁndings are discussed in relation to the pattern of increased student debt in UK higher
Student Debt and its Impact on Student Mental Health
In the last decade students attending university have had to take on higher levels of
debt to pay for their time at university (Hesketh, 1999). This trend has coincided
with the removal of the maintenance grant which was abolished in 1998. Post-1998
a system of student loans was introduced to provide ﬁnancial assistance, however
students also had to pay £1100 in tuition fees each year. So the system of student
ﬁnance shifted from the state paying for the costs of higher education to a situation
ISSN 0309-877X print; ISSN 1469-9486 online/04/010053-14 2004 NATFHE
54 R. Cooke et al.
where most students are reliant on borrowing money from their parents or from the
state. The recent White Paper The Future of Higher Education (HMSO, 2003)
proposes another change to the system, namely that students from low-income
backgrounds will receive a grant of £1100 to help with their costs. However, this will
not cover yearly expenditure (estimated at around £6000 per year for students living
outside London—NUS, 2002) meaning that students will still be in debt. In
addition, there is the possibility that some universities will charge higher tuition fees
(up to £3000) from 2006 onwards. It appears that student debt is here to stay.
Research has examined some of the consequences of increased student debt. For
example, Taylor et al. (1999) found that many students undertake paid work to pay
for their time at university and they argue that students ﬁnd it difﬁcult to juggle work
and study commitments, with work usually impacting on their social life. Moreover,
Humphrey (in press) reports that students who work tend to have worse grades and
join fewer university societies compared to students who don’t work. Looking at the
impact of debt on applying for university, Callender (2003) reported that there were
differences between students who intended to go to university and those that did
not. Students from lower social classes, older students, women and ethnic minorities
were all put off applying by their worries about debt accrual while at university.
Indeed, attitudes to debt were a signiﬁcant predictor of decision to go to university.
The current paper focuses on attitudes toward debt as predictors of student
mental health. Work by Humphrey and McCarthy (1998) found no relationship
between debt (in expenditure terms) and stress levels among students (although see
Roberts et al., 2000). They argued that debt might now be considered an accepted
part of university life. However, other research suggests that while debt itself may
not be directly related to mental health, attitudes toward debt may be linked to
mental health. Stradling (2001) has shown that students’ perceptions of their debt
impact on mental health: students who perceived their anticipated graduate debt as
‘excessive’ were more likely to be anxious or depressed than students who viewed
their anticipated debt as ‘manageable’. Therefore, it may be that attitude toward
debt is more important than actual amount of debt. Research by Lea et al. (2001)
shows that students become more debt-tolerant as they progress through university.
Modifying attitudes to debt may be one coping strategy students use, and this may
reduce feelings of cognitive dissonance, i.e. possessing anti-debt attitudes but having
to borrow money to survive at university (Festinger, 1957).
The evidence outlined suggests that attitudes to debt are important in deciding to
attend university or not, and that these attitudes impact on mental health when
students are at university. This paper builds on earlier ﬁndings in a number of ways.
First, the measures of mental health employed in earlier studies have not been
validated with student populations. In contrast, the measure employed here (GP-
CORE: Sinclair et al., submitted) has. Second, data were collected from students in
all three years of their degree course, to allow for the assessment of trends in the
data. Third, the sample sizes used (greater than 1000 at all times) are unlike any
collected before in terms of scale. In sum, we hope to provide the most comprehen-
sive analysis of the impact of attitudes to debt on mental health to date. The present
study aims to address two issues. First, to provide a description of the impact of debt
Student Debt and Mental Health 55
on student mental health for students in their ﬁnal year of university. This will
provide a baseline to judge other results by. Second, to examine the impact of
attitudinal measures on student mental health, over their time at university.
The ﬁndings presented here are part of the data yielded by the UNIversity Quality
of Life and Learning (UNIQoLL) project currently being conducted at a Red Brick
university in the UK (Audin et al., 2003). UNIQoLL is a longitudinal study of
student perceptions of university life that covers a variety of issues including mental
health, ﬁnancial concerns, and levels of paid work. The project started in 2000 and
data have been collected for three years, with student mental health measured at
seven time-points. In this study we focus only on the measure of mental health
completed at the end of Semester 1 each year because issues relating to ﬁnance were
also collected at that time-point. In short, these data allow us to both hone in on a
particular time-point for thorough analysis (e.g. the third year) as well as examine
trends over time.
Students were asked to report their gender, age, nationality, whether they worked
during term, and in the third year they were asked to report what their anticipated
debt would be at the end of their course. Student mental health was measured using
the GP-CORE (Sinclair et al., submitted) a shortened form of the CORE-OM
(Evans et al., 2002). The GP-CORE has been validated on a student population and
yields high reliability. Students answer 14 items which are drawn from three
subscales of the CORE-OM (functioning, problems and subjective well-being). All
items are scored from 0 (not at all) to 4 (all the time). Lower scores indicate better
mental health. Sinclair et al. demonstrated the discriminant validity of the GP-
CORE for student samples by showing that students’ score signiﬁcantly lower on the
scale than patients undergoing therapy. In the present study, internal consistency
was high with alpha values of 0.85 for the GP-CORE in each year.
Two attitude items were employed. In all three years, students were asked ‘Are
ﬁnancial concerns a current issue?’ (1 ⫽ not at all to 5 ⫽ a lot). In the third year
questionnaire, students were also asked ‘To what extent does your debt worry you?’
(1 ⫽ not at all to 5 ⫽ a lot, 0 ⫽ not applicable).
The participants in this study were all students who entered university in 2000.
Table I provides demographic data for each year. Across the three years, there were
more female students (around 50%, compared to about 40% male students, 10%
did not list their gender), more non-mature students (around 80% were aged under
21 on entry to university), and more UK students (about 80% as an average). In the
ﬁrst year 23% of students were employed in paid work during term-time, and this
56 R. Cooke et al.
TABLE I. Demographic information for the three student groups
Demographics First Years Second Years Third Years
Male 841(39%) 475(35%) 522(38%)
Female 1099(51%) 622(46%) 707(51%)
Not Speciﬁed 206(10%) 263(19%) 162(11%)
Mature 113(5%) 51(4%) 79(6%)
Non-mature 1827(85%) 1046(77%) 1150(83%)
Not speciﬁed 206(10%) 263(19%) 162(11%)
UK 1793(84%) 1034(76%) 1152(84%)
International 145(6%) 63(5%) 76(6%)
Not speciﬁed 208(10%) 263(19%) 148(10%)
Work in term 492(23%) 434(32%) 442(32%)
Don’t work in term 1438(67%) 818(60%) 778(56%)
Not speciﬁed 216(10%) 108(8%) 171(12%)
rose to about 32% in the second and third years. So a majority of students were not
employed in paid work during term time.
Students completed the questionnaires toward the end of the ﬁrst semester in their
ﬁrst, second, or third year. Response rates were 38% (N ⫽ 2146), 23% (N ⫽ 1360)
and 26% (N ⫽ 1391) of the total student population at the university, for ﬁrst,
second and third years, respectively.
In the present paper, effect size differences were used to compare mean scores.
Rather than use ANOVA, which is likely to produce signiﬁcant effects due to the
large samples tested, effect size differences were chosen, as they allow for differences
in the sample size. We set a level of ⫾ 0.4 standard deviation units (SDUs) as being
an effect sufﬁciently large to have utility and therefore be worthy of comment. In
choosing this level we followed previous work (Cooke et al., submitted). In the same
way that a Z-score of 1.96 is associated with the top 5% of the values in a
distribution, a Z-score of 0.4 is associated with 34% of the values in a distribution.
We have chosen this Z-score as indicative of meaningful differences because it is
known that one standard deviation above a standardized mean is 34% of the
population of values in the distribution. Therefore a mean difference of at least 0.4
of a Z-score is about one standard deviation from the mean in the distribution of all
values and is worthy of comment. In other words, of all the possible values a variable
can take (from 0–4 in the present case) a difference between two values (e.g. the
Student Debt and Mental Health 57
average score for those who worry ‘a lot’ versus ‘not a lot’) of 0.4 is sufﬁciently large
to suggest that these two groups are responding differently .
To provide a start point for the analyses, we calculated the mental health of the
student groups measured via the GP-CORE. The measure is scored so that lower
scores are indicative of better mental health. The mean score in year 1 (M ⫽ 1.30)
was lower than the values for year 2 (M ⫽ 1.47) and year 3 (M ⫽ 1.43), which
indicates a trend for worsening mental health as students progress through univer-
sity. There were two waves of analyses. First, we examined the impact of demo-
graphic variables, debt and paid work on mental health, to compare the present data
with previous results. Second, we investigated the relationship between attitudinal
variables and mental health.
Students supplied us with information about their gender, age and nationality. To
control for the possibility that different groups respond differently, and that this may
affect their attitude scores, we ﬁrst compared the scores on the GP-CORE items for
these three variables. A number of differences emerged from the analyses. In all
three years, women scored lower (i.e. better) on the measure ‘I have felt warmth and
affection for someone’ than men. In year 1 and year 2 women scored lower on the
measure ‘I have felt I have someone to turn to for support when needed’. There were
no other differences. Also, in all three years international students scored higher (i.e.
worse) on the two measures mentioned above, compared to UK students. Finally,
mature students (classed as those who were 21 or older on entering university) had
worse scores in year 1 and year 2 for the measure ‘I have felt I have someone to turn
to for support when needed’, than younger students. Thus, although there were
some differences due to demographic variables, the majority of items were scored
similarly regardless of differences in gender, age or nationality.
Debt and Paid Work
We used effect size differences to ascertain if there were any differences in mental
health levels between students who worked in term and those who did not. Although
there was a trend for worsened mental health among students who worked, in each
year, none of these differences were signiﬁcant. Therefore, paid work does not result
in signiﬁcantly worse mental health.
Third-year students were asked to report how much debt they anticipated they
would have by the time they completed university. The average amount of antici-
pated debt on leaving university was £9082.78, with a standard deviation of
£5081.61. Because there was very little correlation between mental health and debt
(r ⫽ 0.03, ns), further analyses were not conducted.
58 R. Cooke et al.
The correlation between mental health and debt worry was 0.28 (p ⬍ .001) among
third-year students, indicating as GP-CORE scores decreased (i.e. mental health
improved) debt worry lessened. Effect sizes were computed between those students
who answered ‘a lot’ (high worry) versus those who answered ‘not a lot’ (low worry)
and students who answered ‘a lot’ versus ‘not applicable’. It was assumed that
students who answered ‘not applicable’ had no debt, and examining the data on
anticipated debt this was found to be the case. Therefore, this group was assumed
to have no debt worry and were classed as the ‘no worry’ group. The effect sizes are
displayed in Table II.
Table II shows that students categorised as high worry (N ⫽ 292) had signiﬁcantly
higher GP-CORE scores (M ⫽ 1.64) compared to low debt worry students
(N ⫽ 130, M ⫽ 1.19). To provide further insight into this difference we decided to
compare the mean scores for each group on the 14 items that make up the
GP-CORE. High debt worry students felt more tense, anxious, or nervous, more
criticised by other people, more unhappy, and more irritable with other people. In
addition, high debt worry students also felt less OK about themselves, less able to
cope when things go wrong, less happy with things they have done, less able to do
things they have needed to, and less optimistic about the future, than students low
on debt worry. This analysis shows that differences in how students perceive their
debt are reﬂected in their mental health levels. In addition, there is a signiﬁcant
difference between the high worry and low worry groups in their anticipated amount
of debt on leaving university. High worry students expect to leave university with
more debt (M ⫽ £11667, SD ⫽ 4034) than low worry students (M ⫽ £7631,
SD ⫽ 4812).
A second analysis, comparing students with high debt worry with students who
had no debt (N ⫽ 89) worry was conducted. This was done to see if there were
differences in GP-CORE scores between students who were very worried about their
debt and students who had no debt to worry them. GP-CORE scores were
substantially higher for those with high debt worry (M ⫽ 1.64) compared to students
with no debt worry (M ⫽ 1.28). Moreover, high debt worry students felt more tense,
anxious or nervous, felt less able to cope when things go wrong and felt that they had
not achieved the things they wanted to. However, there were fewer substantial
differences between students classed as high worry and those with no worry than
between high worry and low worry students.
There was only one substantial difference in mental health measures between
students with low debt worry and those with no debt worry: Students with low debt
worry had less feeling that they had achieved things they wanted to compared to
students with no debt.
Financial Concerns a Current Issue
This question was asked in all three years, so it was possible to plot how students’
ﬁnancial concerns differ over time. Figure 1 illustrates the results. There is a trend
Student Debt and Mental Health 59
TABLE II. Means, standard deviations and effect size differences for students with high worry, low worry, or no worry
High Worry (N ⫽ 292) Low Worry (N ⫽ 130) No Worry (N ⫽ 89)
Variable Mean Sd Mean Sd ES Mean SD ES
I have felt tense, anxious or nervous 2.56 1.03 1.58 1.14 0.92* 1.99 1.12 0.54*
I have felt that I have someone to turn 1.06 1.03 0.95 1.15 0.10 1.10 1.16 ⫺ 0.04
to for support
I have felt OK about myself 1.43 0.99 0.93 0.95 0.51* 1.08 0.89 0.36
I have felt able to cope when things go wrong 1.53 1.02 0.91 0.91 0.63* 1.08 0.82 0.46*
I have been troubled by aches, pains or other 1.62 1.34 1.3 1.26 0.24 1.23 1.22 0.30
I have been happy with the things I have done 1.63 0.94 1.1 0.84 0.58* 1.34 0.97 0.31
I have had difﬁculty getting to sleep or staying 1.97 1.38 1.5 1.34 0.34 1.53 1.35 0.32
I have felt warmth and affection for someone 1.01 1.03 1.29 1.33 ⫺ 0.25 1.03 1.20 ⫺ 0.02
I have been able to do most things I needed to 1.61 0.96 1.2 0.87 0.44* 1.24 0.99 0.38
I have felt criticised by other people 1.37 0.99 0.88 0.98 0.50* 1.05 0.90 0.33
I have felt unhappy 1.66 1.01 1.02 0.9 0.66* 1.31 0.89 0.36
I have been irritable when with other people 1.70 1 1.18 0.94 0.53* 1.31 1.03 0.39
I have felt optimistic about the future 1.92 1.03 1.41 1 0.50* 1.53 1.05 0.38
I have achieved the things I wanted to 1.93 0.97 2.18 0.87 ⫺ 0.27 1.49 0.88 0.46*
GP-CORE 1.64 0.6 1.19 0.6 0.75* 1.31 0.58 0.56*
*Effect size difference of 0.4 or above
60 R. Cooke et al.
FIG.1. Financial concerns over time at university.
for students to be more concerned about their ﬁnances as they progress through
university, with more students indicating that concerns are ‘a lot’ of a problem in the
third year than in the ﬁrst or second year.
We calculated the correlation between ﬁnancial concerns and mental health in
year 1 (r ⫽ .19, p ⬍ .001), year 2 (r ⫽ .22, p ⬍ .001) and year 3 (r ⫽ .24, p ⬍ .001).
These correlations suggest that there is some relationship between ﬁnancial concern
and mental health. We decided to compare the mental health responses of the
participants who answered that ﬁnancial concerns where ‘not a lot’ of a current issue
(low concern group) with those who indicated ‘a lot’ of concern (high concern
group). In all three years, GP-CORE scores were greater for the high concern group
than for the low concern group (see Table III). Therefore, the individual items were
examined in each year as well. The results show that the high concern group scored
signiﬁcantly higher on more items as they progressed through university.
For year 1 students, high concern students felt more tense, anxious or nervous,
troubled by physical problems and aches and pains, criticised, irritable with people
and had greater difﬁculty getting to sleep than low concern students. In year 2,
students who were more concerned reported feeling more tense, anxious or nervous,
more criticised, less OK, less able to do things they needed to, less optimistic, less
able to achieve things they wanted to, and more difﬁculty getting to sleep compared
to those who were little troubled by ﬁnancial concerns. However, the biggest
differences occur for year 3 students, with all but four of the 14 GP-CORE items
showing signiﬁcant differences (see Table III).
It is interesting to examine which variables were consistently rated differently by
high and low concern groups. Three variables were rated signiﬁcantly worse by high
concern students in all three years—feeling tense, anxious or nervous, feeling
criticised by other people and having difﬁculty getting to sleep or staying asleep.
Also, two of the 14 GP-CORE items did not signiﬁcantly differ between the two
groups in any year—‘felt I have someone to turn to for support’ and ‘felt warmth and
Student Debt and Mental Health 61
TABLE III. Means, standard deviations and effect sizes for comparisons on ﬁnancial concerns as a current issue.
Year 1 Year 2 Year 3
High concern Low concern High concern Low concern High concern Low concern
N ⫽ 395 N ⫽ 175 N ⫽ 334 N ⫽ 113 N ⫽ 369 N ⫽ 105
Variable M SD M SD ES M SD M SD ES M SD M SD ES
I have felt tense, anxious 1.84 1.14 1.20 1.11 0.57* 2.4 1.08 1.85 1.24 0.49* 2.55 1.05 1.65 1.14 0.85*
I have felt that I have 1.39 1.17 1.16 1.12 0.20 1.30 1.08 1.00 1.12 0.28 1.11 1.06 1.05 1.15 0.05
someone to turn to for
I have felt OK about myself 1.30 1.03 1.09 0.94 0.21 1.55 1.05 1.10 1.02 0.43* 1.41 0.97 0.97 0.92 0.46*
I have felt able to cope 1.29 1.03 1.03 0.88 0.26 1.53 1.07 1.14 0.99 0.37 1.53 1.03 1.02 0.87 0.52*
when things go wrong
I have been troubled by 1.52 1.33 0.93 1.09 0.47* 1.69 1.29 1.27 1.30 0.32 1.59 1.34 1.22 1.31 0.28
aches, pains or other
I have been happy with the 1.53 0.97 1.24 0.96 0.30 1.72 0.96 1.35 0.96 0.39 1.65 0.94 1.21 0.95 0.46*
things I have done
I have had difﬁculty getting 1.94 1.37 1.19 1.19 0.57* 1.99 1.39 1.31 1.27 0.50* 1.92 1.40 1.29 1.27 0.46*
to sleep or staying asleep
I have felt warmth and 1.12 1.11 1.26 1.17 ⫺ 0.12 1.06 1.12 1.13 1.15 ⫺ 0.06 0.99 1.05 1.21 1.33 ⫺ 0.20
affection for someone
I have been able to do most 1.39 0.97 1.04 0.90 0.37 1.76 1.05 1.24 0.92 0.51* 1.63 0.95 1.23 0.93 0.43*
things I needed to
I have felt criticised by other 1.34 1.06 0.87 0.94 0.46* 1.52 1.06 1.04 0.97 0.46* 1.34 1.07 0.82 0.91 0.50*
I have felt unhappy 1.45 1.11 1.05 1.03 0.37 1.71 1.03 1.41 1.12 0.28 1.67 1.00 1.16 0.92 0.52*
I have been irritable when 1.48 1.01 1.07 0.94 0.41* 1.77 1.03 1.35 1.08 0.40* 1.66 1.04 1.14 0.88 0.51*
with other people
I have felt optimistic about 1.69 1.11 1.35 1.03 0.31 1.94 1.09 1.46 1.17 0.43* 1.90 1.01 1.53 1.11 0.35
I have achieved the things 1.71 1.00 1.35 1.01 0.36 1.93 0.97 1.48 1.01 0.46* 1.93 0.93 1.50 0.98 0.47*
I wanted to
GP-CORE 1.5 0.65 1.13 0.59 0.59* 1.7 0.61 1.31 0.65 0.63* 1.63 0.61 1.21 0.63 0.69*
*Effect size difference of 0.4 or above
62 R. Cooke et al.
affection for someone’. The implications of these ﬁndings will be covered in detail
in the discussion.
One ﬁnal point to make is the relationship between the two attitude items. The
correlation between debt worry and ﬁnancial concerns in the third year was high
(r ⫽ .66, p ⬍ .001) suggesting that current concerns are linked to worries about ﬁnal
debt, though this may reﬂect the fact that students are in their ﬁnal year. In addition,
comparing the effect sizes between the high worry versus low worry groups and
between the high concern and low concern groups there were similar ﬁndings, e.g.
variables that were different in one analysis being different in the other analysis.
However, there were three variables that differed. Difﬁculty sleeping was
signiﬁcantly different in the high versus low concern comparison, but not in the high
versus low worry comparison as was achieved the things I wanted to. In contrast,
there was a signiﬁcant difference between high and low worry students for the item
‘I have felt optimistic about the future’, while there was no difference between the
high concern and low concern groups. These ﬁndings will be further discussed
In summary, demographic and economic measures do not appear to predict differ-
ences in mental health among students. In contrast, students’ perceptions of their
ﬁnances and debt differentiated mental health scores. Students’ concerns about their
ﬁnances appear to be associated with better and worse mental health. All students
face debt and ﬁnancial concerns but how they perceive these difﬁculties seems to
impact on their mental health.
Analysis of demographic data suggests that younger (under 21 on entry to
university), female and UK students develop better social support networks than
other student groups. Further work is needed to explore the reasons for this.
Nevertheless, the results also suggest that any differences in mental health levels
are not primarily due to demographic information, as the majority of GP-CORE
items remained the same regardless of which group the participant belonged
The fact that there was little relationship between GP-CORE scores and antici-
pated debt was probably because there was limited variation in the amount of debt
students anticipate facing. The amount students expect to owe is higher than the
ﬁgure reported in the most recent UNITE poll (M ⫽ £9082 compared to
M ⫽ £8816, UNITE, 2003). It is unclear whether this reﬂects an effect of averaging
across universities in the MORI survey, or university life is more expensive at certain
universities. What the current ﬁgure shows is that the cost of student life is
continuing to rise.
The increase in the cost of university appears to have impacted on students’ debt
tolerance. Work by Lea et al. (2001) indicates that students become more debt-tol-
erant as they progress through higher education. In contrast, the current data show
that students become more concerned with ﬁnances as they progress through
university. Apart from the fact that the two studies used different measures, another
Student Debt and Mental Health 63
reason for the difference is the situation students were in during the period in which
the studies were completed. The Lea et al. study was based on data from 1995, when
there was less necessity in taking loans because grants still existed. This is obviously
a different situation to that facing the students in the present study, with most
students having to take loans to pay for higher education, and these loans being
much larger than in 1995. This may be the main reason for the differences in the
Looking at the psychological data, ﬁnancial concerns become more prevalent over
time, affecting more mental health variables in the third year than in either the
second or ﬁrst years. Since the mental health averages are almost equivalent for the
second and third years, it seems that ﬁnancial concerns are heightened as students
complete their degrees. This is unsurprising as students will have accumulated more
debt by their third year and may have very real concerns about paying these debts
off when they complete their degree.
It was found that students who had high ﬁnancial concern possessed higher scores
on measures of feeling tense, anxious or nervous, difﬁculty getting to sleep and
feeling criticised by other people. Humphrey and McCarthy (1998) argue that
students have higher stress levels than individuals who do not attend university and
Stradling (2001) reported high levels of anxiety among students in his research. This
ﬁnding suggests that while most students ﬁnd university stressful, those who are
worried about ﬁnancial issues are particularly anxious.
There is evidence that shows that greater anxiety is associated with sleep problems
such as insomnia (e.g., Moul et al., 2003; Morin et al., 2003; Viens et al., 2003), so
it may be that high concern/high worry students have more trouble sleeping because
of their anxiety. The other item that appeared to be different for high and low groups
was ‘I have felt criticised by other people’. It is difﬁcult to explain this ﬁnding
without further research. It is necessary to ﬁnd out who these people are. For
example, do students with more debt worry get more criticised by lecturers, friends,
family, etc? It could also be the case that high worry students may be sensitive to
criticism that less worried students would ignore.
There were no differences between high and low concern/worry groups on the
following two items: ‘I have felt I have someone to turn to for support’ and ‘I have
felt warmth and affection for someone’. These two items are concerned with
friendships and social support, and the lack of difference suggests that the high
worry students may have in place coping strategies for dealing with their worries that
are as effective as the low worry students. This seems to rule out the suggestion that
low worry students are people who simply cope better with university than high
worry students, but it also begs the question of why the high worry students have
worse GP-CORE scores. Research examining risks and buffers to mental health in
students is needed to try and untangle the various effects of university life on student
mental health (Cooke et al., submitted).
There were a number of limitations in the present research. One weakness is that
it fails to account for the possibility that students in the high concern/worry groups
are just generally more anxious. Debt is clearly a contributing factor but not
necessarily the sole factor accounting for their distress. Although this explanation is
64 R. Cooke et al.
possible, we argue that debt and ﬁnancial concerns are an extra psychological
burden on top of the other pressures of university life.
Examining the pattern of results for the comparisons between the high, low and
no debt worry groups suggests that the high worry group are not necessarily more
worried about things in general. First, there were fewer differences between the high
and no worry groups, than between the high and low worry groups, suggesting that
ﬁnances do impact on mental health measures as the both high and low worry
groups have debt. Second, there was only one difference between the low and no
worry groups which suggests that these two groups give similar responses. Put
together, these ﬁndings suggest that the high worry group are similar to the no worry
group, who in turn are similar to the low worry group, and this argues against the
likelihood that the high worry group are just generally more worried about a number
of issues. However, until research has been conducted to compare these explana-
tions it is not possible to be certain at present.
The one difference between students with low and no debt worry was that
students with low debt worry had less feeling that they had achieved things they
wanted to compared to students with no debt worry. This interesting ﬁnding
suggests that although the low debt worry students are not suffering worsened
mental health due to their debt, they may be prevented from doing desired activities
by their debt. This ﬁnding is consistent with Humphrey’s (submitted) study that
demonstrated that students who work join fewer university societies. These results
show that debt can have a detrimental impact on students’ experience of university.
Another concern is the use of two attitude measures which correlate highly, as it
could be suggested that the two measures were tapping similar responses. However
the differences in the comparisons noted above suggest that the ﬁnancial concern
item taps present concerns whereas the worry item taps future concerns. For example,
comparing someone with high current ﬁnancial concerns and someone with low
current ﬁnancial concerns on the item ‘I feel optimistic about the future’ is unlikely
to produce a difference, because the question is about the future, whereas thinking
about expected ﬁnal debt may differentiate responses to feeling optimistic about the
Further research examining the impact of paid work on mental health, such as
investigating the number of hours a student works, is needed, as it seems likely that
student employment is becoming an unavoidable aspect of university life for many
students. Also, research examining students’ experiences after graduation is re-
quired. Callender (2003) notes that certain students are put off university by the
debt associated with attending, while the present paper indicates that debt only
appears to impact on the mental health of certain students (i.e. those with high debt
worry). It is unknown how graduates experience full-time employment and pay for
their university debts. It would be interesting to see if graduates earn enough money
to pay off their ever-increasing debts, and see if they view university as a good
investment. Other areas for future research include examining why females develop
better social support networks than males and seeing if better networks can protect
students from the effects of debt, to further examine the causes of higher GP-CORE
scores and to conduct similar research at other institutions. This latter research is
Student Debt and Mental Health 65
important because the majority of students attending red-brick universities tend to
come from wealthy backgrounds, so the effects reported here may be even stronger
in newer universities.
The results presented have implications for professional action. First, it is import-
ant to make professionals aware of the actual ﬁnancial situation facing students, and
to let students know that they are aware of their ﬁnancial burden. This should foster
better understanding between staff and students. Second, departments could assign
a member of staff to oversee the impact of debt on studying. It would be this
person’s responsibility to encourage students to approach them for help when they
need it and to suggest ways to reduce conﬂict between students’ need to study and
the need to work to support themselves. For example, submitting work via email
may be a way to allow students to work when they have the time available, rather
than limiting the submission of work to ofﬁce hours which may be a problem for
certain students. Finally, universities could run workshops on coping with ﬁnancial
hardship after graduating. If students are reminded that their current ﬁnancial
situation is temporary and that they will be in a better position after graduation, this
may reduce some of the concerns students have and, potentially, lead to better
In conclusion, the results presented highlight the utility of attitude items in
describing issues related to student debt and mental health. Attitude measures have
allowed us to discover subtle differences in students’ mental health that would have
been missed by other measures. It is important to include such measures in future
studies as they have been shown to usefully add to our understanding of the impact
of debt on the stresses and strains students face at university.
UNIQoLL is funded by the University of Leeds. We thank members of the Steering
Group, Departmental Representatives and successive Union Executive members for
their encouragement and constructive criticism. Authors afﬁliated to PTRC were
also funded byR&Dlevy from Leeds Community and Mental Health Teaching
 It is worth noting that analyses were run using ANOVA to check the ﬁndings. Post-hoc
analyses using Tukey’s HSD reinforce the results from the effect size analyses. These data
are available from the ﬁrst author.
AUDIN, K., DAVY,J.&BARKHAM,M. (2003) University Quality of Life and Learning
(UNIQoLL): an approach to student well-being, satisfaction and institutional change,
Journal of Further and Higher Education, 27(4), pp. 365–382.
ALLENDER,C.(2003) Attitudes to Debt: school leavers and further education students’ attitudes to debt
and their impact on participation in higher education (London, Universities UK).
66 R. Cooke et al.
COOKE, R., BARKHAM, M., AUDIN, K., BRADLEY, M., DAVY,J.&SLACK,J.(submitted) Psycholog-
ical well-being of ﬁrst year university students: risks and buffers across their ﬁrst year at
VANS, C., CONNELL, J., BARKHAM, M., MARGISON, F., MCGRATH, G., MELLOR-CLARK,J. &
UDIN,K.(2002) Towards a standardised brief outcome measure: psychometric properties
and utility of the CORE-OM, British Journal of Psychiatry, 180, pp. 51–68.
ESTINGER,L.(1957) A Theory of Cognitive Dissonance (Stanford, CA, Stanford University Press).
ESKETH,A.J. (1999) Towards an economic sociology of the student ﬁnancial experience of
higher education, Journal of Educational Policy, 14, 385–410.
HMSO (2003) The Future of Higher Education, Cmnd 5735 (London, HMSO).
UMPHREY,R. (submitted) Drawing structured inequality into higher education: the impact of
part-time working on British university students, Higher Education Quarterly.
UMPHREY,R. & MCCARTHY,P. (1998) Stress and the contemporary student, Higher Education
Quarterly, 52, pp. 221–242.
EA,S.E.G., WEBLEY,P.&BELLAMY,G. (2001) Student debt: expecting it, spending it and
regretting it, in: A. J. S
COTT,A.LEWIS &S.E.G.LEA (Eds) Student Debt: the causes and
consequences of undergraduate borrowing in the UK (Bristol, The Policy Press).
ORIN,C.M., RODRIGUE,S.&IVERS,H. (2003) Role of stress, arousal and coping skills in
primary insomnia, Psychosomatic Medicine, 65, pp. 259–267.
OUL,D.E., NOFZINGER,E.A., PILLKONIS,P.A., HOUCK,P.R., MIEWALD,J.M.&BUYSSE,D.
J. (2003) Symptom reports in chronic insomnia, Sleep, 25, pp. 548–558.
NUS (2002) NUS Press Pack 2002–2003: higher education student ﬁnance (London, NUS).
OBERTS, R., GOLDING, J., TOWELL, T., REID, S., WOODFORD, S., VETERE,A. & WEINREB,I.
(2000) Mental and physical health in students: the role of economic circumstances, British
Journal of Health Psychology, 5, pp. 289–297.
INCLAIR, A., BARKHAM, M., EVANS, C., CONNELL,J.&AUDIN,K. (in preparation) A core
measure of psychological well-being for the general population: development, psychometric
status and student norms for the GP-CORE.
AYLOR, P., SMITH,N.&COOPER,C. (1999) Hardship debt and stress: the experience of the
student worker. Student Wellbeing in Higher Education Conference, Glasgow.
UNITE (2003) Student living report 2003 (London, UNITE).
IENS, M., KONINCK,J.D., MERCIER, P., ST-ONGE,M.&LORRAIN,D. (2003) Trait anxiety and
sleep-onset insomnia: evaluation of treatment using anxiety management training, Journal of
Psychosomatic Research, 54, pp. 31–37.