Patterns of special consideration requests at a UK university: reasons given and associations with demographic factors

Abstract Students who are unable to complete an assessment due to circumstances beyond their control (e.g. illness) are often asked to submit a request for special consideration. However, few studies have looked at the reasons why these requests are made, or whether certain students are more likely to submit requests than others. The current study looked at 2126 such requests submitted by 461 students over one academic year and compared students who submitted requests with those who did not on several variables, including gender, full-time/part-time status and undergraduate/postgraduate courses. Distribution of these requests by type (e.g. physical health, bereavement) was examined, in addition to how many students submitted more than one request. The study found that around one-quarter of students submitted requests, with more than half of these related to mental or physical health issues. Full-time students were more likely to submit requests than part-time students although few other demographic differences emerged. The results suggest that some groups may be unevenly affected by special circumstances related to assessment although further work is needed to inform policies regarding special consideration.


Introduction
During an individual's academic studies, there may be times when circumstances beyond the student's control adversely affect their academic performance, such as attendance at an examination or other summative assessment. These occurrences (including, but not limited to, short-term illness, bereavement and participation in jury service; see de Lambert and Williams 2006, Table 1) are known as 'special considerations' , and institutions across both higher and further education have developed policies to help students in these situations. This practice is designed to ensure fairness in academic procedures (Collings, Garrill, and Johnston 2018; see also Office of the Independent Adjudicator for Higher Education (OIAHE) 2021) but is also open to abuse (Vidal Rodeiro 2010). Around 15% of university students submit an application for special consideration at some point in their studies (Zimmermann, Kamenetsky, and Pongracic 2015), although few reliable estimates of prevalence have been reported.
University processes for dealing with these circumstances, which are typically short-term (OIAHE 2021), differ from those within secondary institutions; requests are generally provided by the student (not the educational institution) and 'policies and procedures tend to be less flexible' (Zimmermann, Kamenetsky, and Pongracic 2015, p. 263). However, even with these constraints, criteria for special considerations vary. Relevant 'evidence' , such as a signed medical report or crime reference number is often required (Collings, Garrill, and Johnston 2018), and, if granted, one of several outcomes can be approved, such as extended deadlines, reducing penalties for late assignments or adjusting grades. This is different to systems used to support those with a 'permanent' disability, although there are similarities (Kimball et al. 2016).
In addition to its centrality in grading, assessment opportunities are important to students' learning cycles, and missed deadlines can help identify students who are at risk of failure or withdrawal (Linden et al. 2020). Therefore, it is important for universities to make provision for circumstances where students cannot submit assignments, although there has been little research into special considerations, a topic rarely covered in teaching texts (de Lambert and Williams 2006). Although this issue could be considered 'administrative' , the prevalence and consequences of this issue are acknowledged in recent UK guidance, including a recommendation that data are used to assess whether certain groups of students are over or under-represented in using such processes (OIAHE 2021). The growing proportion of 'non-traditional' students (often defined according to age; see Chung, Turnbull, and Chur-Hansen 2014) highlights the need for university policies to adequately serve this segment of the student population. Studies of students with particular needs (e.g. young adult carers) highlight the importance of flexibility around deadlines and consideration of their unique circumstances (Day 2019; see also Eifler 2009), and it has also been suggested that certain circumstances may impact individuals from some backgrounds more than others (Petro, Gonzales, and Calarco 2020).
Despite some devolution of pastoral care to specialist support services (marandet and Wainwright 2009), students often approach academic staff before disability services . Further, the impact of special circumstances on academics appears to be growing through, for example, greater involvement at varying levels of 'fitness to study' procedures (Caleb 2019), and some courses -often those involving professional registration -include teaching in this area. Greater understanding of reasons behind Special Circumstances Requests (SCRs) could inform policies to support students when they are experiencing difficult circumstances, as well as helping develop support for academic and non-academic staff.
Several reasons can be given to excuse a student's performance and a sample of 200 students from an Australian university (Croucher 1995) identified the following circumstances leading to a SCR prior to final examinations: student illness (31%); physical problem (24%); psychological problem (22%); and a problem with a family member or close friend (13%). The remaining 10% referred to miscellaneous issues. In addition to the range of factors influencing SCRs, it is not clear whether certain features of special circumstances, such as being beyond the student's control, are fully justifiable (de Lambert and Williams 2006), nor that clear evidence can always be presented (e.g. mental health issues or the impact of a bereavement). Illnesses such as influenza and the common cold can cause missed classes and impaired academic performance (nichol, Heilly, and Ehlinger 2005), and a student's experience of stress related to 'not coping academically' also represents a significant barrier to achievement (Vaez and Laflamme 2008, p. 192). It has been acknowledged that students' health concerns frequently impact academic outcomes, with the two closely linked (Duffy et al. 2019). Continued absences and student attrition also incur costs for universities and are likely to compromise learning and academic achievement for the student (newman-Ford et al. 2008). Prevalence of common mental health problems in students is estimated at around a third (Ibrahim et al. 2013) and there are suggestions that the severity of student mental health problems is increasing (Thorley 2017). Although individuals with mental health problems are more likely to drop out of university, little is known about how health conditions impact assessment directly, and how education providers should support students when they are negatively impacted Lipson and Eisenberg 2018).
Several factors have been found to be associated with mental health in university students, including demographic factors (e.g. gender; Tran et al. 2017, Wathelet et al. 2020, financial difficulties (Tran et al. 2017), and university-specific factors such as satisfaction with studies (Lipson and Eisenberg 2018) and year of study (Wathelet et al. 2020). A sizeable body of research suggests that academic issues are a 'major source of stress for students' (Pitt et al. 2018, p. 62) and recent increases in student numbers (Bolton 2020) are also likely to have contributed to a lack of personalised support (Stephen, O'Connell, and Hall 2008) and higher staff:student ratios, which might impact mental health (Tinklin, Riddell, and Wilson 2005). Other studies have found that students with mental health problems may experience feelings of guilt and failure related to SCRs specifically (martin 2010). Thus, it is important to evaluate how universities respond to, and accommodate, such challenges and to understand more about the impact of mental health problems on students' academic engagement and performance.
Special circumstances requests represent one way in which students with health problems (both physical and mental) can be supported in their studies. Given suggestions that greater support could increase retention rates of those considering leaving university (Rovai 2003;Tinklin, Riddell, and Wilson 2005;Linden et al. 2020), SCRs could form part of a suite of measures to assist students who may be at risk of lower achievement or drop-out. Although recommendations that better-coordinated support could 'assist in better student management of stressors at particular times' (Pitt et al. 2018, p. 72), there has been little exploration of the effectiveness of these interventions, and few estimates of the scale of the problem.
Special circumstances requests can be used to identify 'at-risk' students (Atif et al. 2020;OIAHE 2021), including those who are prone to academic disengagement. Although certain populations are less likely to complete their degree (Taniguchi and Kaufman 2005), policies are often developed with little consistency across institutions . Students defined as 'nontraditional' experience several procedural obstacles to academic success, as well as feelings of isolation and not fitting in (Goncalves and Trunk 2014), which can compound academic challenges (Wyatt 2011), and qualitative studies have suggested that such students can feel 'marginalized' by policies the believe favour traditional students (markle 2015, p. 277; see also Tinklin, Riddell, and Wilson 2005).
Data concerning students' age are of interest given recent population shifts and changes in university approaches to adult learning (Stephen, O'Connell, and Hall 2008; see also Humphrey and mcCarthy 1999). Despite unique challenges and differences in educational priorities, mature students (a commonly used cutoff is age over 25, in line with definitions of non-traditional students (Chung, Turnbull, and Chur-Hansen 2014), although this distinction is somewhat arbitrary and others (e.g. 21 years) have also been used; metzner and Bean 1987) appear to achieve no worse (and perhaps at a slightly higher level) than their younger counterparts (Hoskins, newstead, and Dennis 1997;Vaez and Laflamme 2008), possibly reflecting more adaptive strategies for learning (Richardson 1994) or more 'intrinsic' reasons for pursuing further education (murphy and Roopchand 2003). Although most mental health problems emerge by the age of 24 (Kessler et al. 2005), older (and other 'non-traditional') students may experience different stressors to their younger counterparts, including different financial situations (Shao et al. 2020), issues relating to families and children, and greater feelings of alienation at university (Wong 2018). It is not known how their different situations might impact older students in terms of SCRs, but this limited evidence suggests that family difficulties might feature more prominently than in younger cohorts.
Regarding other demographic factors, there is some evidence that men are more likely to be dissatisfied with their university experience compared to women (Lipson and Eisenberg 2018) and less likely to complete their studies (Ewart 2012). Whilst there is some variation across subject area (e.g. men have high rates of dropout in nursing programs; Herrera 2013), female students tend to hold more positive attitudes about help-seeking (nam et al. 2010) and may therefore be more likely to make SCRs. Students from ethnic minority backgrounds have cited financial difficulties and lack of specific support as contributing to proportionally higher attrition rates (Bond et al. 2008) and appear less likely to achieve the highest class of degree outcome (Wong 2018). These patterns align with theories that students who do not fit the traditional stereotype may feel less engaged with university (Rovai 2003;Wyatt 2011) although few studies have considered whether these profiles are associated with submission of SCRs.
It can be surmised, then, that little is known about patterns of SCRs within higher education, and whether previous trends (Zimmermann, Kamenetsky, and Pongracic 2015) are representative of other institutions using more recent data. Some studies have considered only undergraduate students or SCRs following a particularly devastating event (Collings, Garrill, and Johnston 2018), with less attention paid to postgraduate students. Understanding more about individuals who submit SCRs could help to standardise policies around student health and academic assessment, as well inform the support provided to those who struggle with the additional demands of university study at the earliest possible stage (Goncalves and Trunk 2014;Wong 2018).
With so little literature to draw on, the present study will address the following hypotheses: submission of SCRs will be more likely in female (as opposed to male) students, older (rather than younger) students, and part-time (as opposed to full-time) students (Zimmermann, Kamenetsky, andPongracic 2015, see also Croucher 1995). In addition, rates of SCRs will be assessed according to student status (undergraduate/postgraduate) and ethnicity, although no directional hypotheses are made given the limited data published in this field. Should evidence emerge for unequal representation of certain groups making SCRs, this could support calls for universities to review academic engagement and develop inclusive policies, particularly for non-traditional students (Taniguchi and Kaufman 2005;Goncalves and Trunk 2014). Identifying students who may be more likely to submit SCRs could highlight those at risk of struggling academically (Atif et al. 2020) and findings could help higher education institutions evaluate the impact of their policies and ensure that any changes are grounded, as far as possible, in empirical evidence.

Procedure
Data were obtained from the School of Psychology and Clinical Language Sciences (PCLS) of a large university in England and covered all SCRs submitted between 1 st October 2018 and 30 th September 2019. Data were collected retrospectively and complied with university policies regarding data protection. Conduct of the study was approved by the university's Research Ethics Committee and through consultation with the university's data management team.

Participants
The university hosts around 19,000 students, with campuses in the UK, South Africa and malaysia. Around 1 in 8 declare a disability -a number higher than previous estimates although one which may be increasing (Riddell and Weedon 2014). Students can submit SCRs throughout their studies, including separate requests for the same reason (e.g. if more than one assessment was affected). Therefore, data are also reported on students who submitted multiple requests on separate dates (indicating distinct periods of special circumstances).
The design was partly 'opportunistic' in that information was gathered from academic records, although the variables selected for investigation were based on previous research and represented issues of practical interest when looking at university processes around SCRs. Although the sample was obtained from one school, this is one of the largest in the university and offers a range of undergraduate and postgraduate degree programmes. Further, inclusion of students taking courses in psychology provides some correspondence with previous studies.

University policy on special circumstances
The university under study has a policy on special circumstances, which is periodically reviewed. In brief, it specifies what might constitute special circumstances, details those circumstances likely to be considered, and provides examples of 'evidence' which might be appropriate. much of this guidance is in-keeping with that of other institutions and recommendations of an independent body covering higher education policies and practices (OIAHE 2021). Students are reminded of this policy through several channels, including meetings with tutors, course inductions and dissemination via web-based university microsites. Academic staff are typically responsible for making decisions on SCR outcomes and are also involved in the appeals process.

Data extraction
Students completing an SCR are asked to indicate the nature of their concern through a drop-down menu (e.g. 'bereavement' , 'court attendance'). Where no reason was provided, this was recorded as 'other' . Further data (e.g. gender) were drawn from academic records. However, given shortcomings in the use of academic records for research purposes (Jove et al. 2020), additional checks were made to help ensure the integrity of the data.

Statistical analyses
Data were inspected for normality. Visual inspection and the Shapiro-Wilk test (p < .001) suggested that the distribution of age was positively skewed. Therefore, non-parametric mann-Whitney U tests were used to compare group means for analysis of age effects although it was not possible to compare the age of students submitting SCRs to those who did not as individual-level data was not available for the latter group. χ 2 tests (inherently a distribution-free test, reporting odds ratio (OR) and Cramer's V to estimate the strength of significant effects) were used to compare groups for categorical variables (gender, status, mode of study). In addition to comparing the profile of students who submitted SCRs to those who did not, those who submitted requests on more than one occasion were compared to those who did so only once. Only male and female students are included in analysis of gender; three did not state a gender and are excluded from this subgroup analysis. A cutoff of 25 was used to indicate older and younger students when looking at age as a categorical variable.

Power analysis
Given limited research in the area and methodological heterogeneity in these studies, a formal power analysis was not conducted. Results are therefore discussed in the light of small sample sizes in some groups.

Results
Of the 1859 students included in this study, 461 (24.8%) made at least one SCR, comprising a total of 2126 requests (4.6 per student). Participants were enrolled on 32 different degree programmes (e.g. BSc Psychology, mSc Psychology Conversion, mSci Speech and Language Therapy). Age of those who submitted SCRs ranged from 18 to 61 (M = 27.29, SD = 9.11), broadly in line with the wider school. Due to small sample sizes in minority ethnic groups, students who identified as a non-White ethnicity were collapsed into one group (n = 129; 27.3%; nine did not provide this information).
Following the method of Croucher (1995), the number of requests received per 100 students enrolled was calculated as 118.0 (1934 requests from 1639 students) for full-time students and 87.3 (192/220) for part-time students (114.4 for the total enrolment). Table 1 shows the breakdown of students by a range of demographic variables, according to those who submitted SCRs and those who did not.

Characteristics of students submitting SCRs
The 461 students submitting SCRs were compared to students who had not submitted SCRs on the hypothesised variables (gender, ethnicity, student status, mode of study); see Table 1. men and women were equally likely to submit SCRs (χ 2 (1) = 0.00001, p = 0.997) and there were no differences in the number of students submitting SCRs according to ethnicity (χ 2 (1) = 0.24, p = 0.63) or undergraduate/postgraduate status (χ 2 (1) = 2.69, p = 0.10). Full-time students were more likely to submit SCRs than part-time students (χ 2 (1) = 22.54, p < 0.001; OR = 2.70, V = 0.110) although small numbers of part-time students submitting SCRs (n = 26) precluded further analysis of this subgroup. Table 2 shows the nature of SCRs, according to individual requests (N = 2126) and students (N = 461). Issues related to mental or physical health accounted for over half of cases. miscellaneous reasons were provided in around one-third of requests, and this proportion was larger in the sample of postgraduate students compared to undergraduate students (150/220 (68.2%) versus 60/241 (24.9%); χ 2 (1) = 86.88, p < 0.001; OR = 6.46, V = 0.434). The distribution of specific reasons was affected in line (i.e. the proportions were generally lower in the postgraduate group). The distribution of reasons was similar across gender, ethnicity and student status, although there were small numbers in some groups (see Table 1). The reasons given for single versus multiple requests were similar, with mental health and physical health frequently cited as reasons across both groups (see Table 2).

Reasons given for SCRs
Looking at age more specifically, compared to older students, students aged under 25 were more likely to cite reasons related to mental health (26.1% versus 8.5%) or physical health (28.0% versus 12.0%). Family issues were evenly distributed (6.1% versus 6.0%) although younger students provided a response of 'other' less frequently (29.1% versus 67.0%).

Multiple SCRs
most students who submitted SCRs (n = 333; 72.2%) made multiple requests, and 223 (48.4% of total) made requests on different occasions, indicating that requests were often made for multiple assessments falling at a similar time (e.g. two deadlines in the same week). The characteristics of students who made requests on more than one occasion were compared to those who made SCRs on one occasion during the academic year.

Provision of evidence
Evidence (usually required to support an SCR) was provided by 290 students (62.9%), and 127 (27.5%) included a letter from a primary care doctor or other healthcare professional external to the university. 54 (13.3%) of the SCRs submitted by individual students were rejected, although this decision can be appealed. There was no significant difference between the rates of SCRs rejected when evidence was provided (11.5%) and when it was not (16.3%; χ 2 (1) = 1.417, p = 0.23), nor when this evidence was provided by a medical professional or from another source (8.55% versus 15.17%, χ 2 (1) = 2.499, p = 0.11).

Discussion
Only a handful of studies have looked at the characteristics of university students submitting requests for special consideration of an assessment. In the current study, around one-quarter of enrolled students submitted SCRs, with approximately half of these making requests on different occasions. This figure is higher than previous estimates and supports suggestions of *the university has a stipulation that students can 'backdate' for consideration work where they have been diagnosed with a disability (e.g. dyslexia) during the course of their studies upwards trends in the number of SCRs made in higher education (Croucher 1995;Zimmermann, Kamenetsky, and Pongracic 2015) and in UK secondary schools (Vidal Rodeiro 2010). Similarly, although reasons for this increase are not known, the high proportion of health-related requests in the current study, and particularly in students aged under 25, fits with existing theories about rises in health complaints for young adults (e.g. Potrebny et al. 2019). It has been suggested that universities are less likely to decline medical requests (Zimmermann, Kamenetsky, and Pongracic 2015) but there was no such evidence in the current study, although sample sizes were small. Although the number (and type) of assessments will influence the rates of SCRs recorded, the current study noted number of SCRs and number of students, including data from several undergraduate and postgraduate degree programmes and may be more representative than a focus on one course of study alone. The findings suggest that students who submitted SCRs differ little from those who did not according to demographic variables. This is in line with some previous work although two key studies (Croucher 1995;Zimmermann, Kamenetsky, and Pongracic 2015) both found that female students were more likely to submit SCRs than male students, which was not in evidence here. The reasons for these discrepant findings are unclear, although prior studies have rarely included postgraduate students, who show demographic differences as well as different approaches to studying and learning (Humphrey and mcCarthy 1999). It may be that postgraduate students have more diverse reasons for making SCRs, reflected by the greater proportion of miscellaneous requests seen in the current study. Overall, however, there was no evidence from the current study that 'non-traditional' students are more likely to submit SCRs than other students, although it is not possible to say whether awareness of the SCR process is a factor. Zimmermann, Kamenetsky, and Pongracic (2015) offers support for another finding of the current study; that women are no more likely than men to make multiple SCRs. These observations might reflect a growing awareness of support in the male population or a shift towards gender equality, and may reflect the 'social climate' of students in the current sample (Ferreira 2003). Given that trends relating to gender may vary according to disciplines (and countries), more research is needed to determine whether policies offer equality when issues such as gender and ethnicity are considered.
Another finding worthy of further research is that full-time students were nearly three times more likely to submit SCRs than part-time students, supporting findings from an Australian study conducted in 1993 (Croucher 1995). A later Canadian study, however, found that part-time students were more likely to submit multiple requests than full-time students (Zimmermann, Kamenetsky, and Pongracic 2015), whereas there was no difference in the current study. Given the limited research on this topic, conclusions regarding this are speculative. It could be that full-time students are more engaged with university and more aware of how to make requests, or that part-time students are 'sidelined' by higher education policies (Callender 2011). Given findings that part-time students are less likely than full-time students to graduate (Taniguchi and Kaufman 2005), the current findings suggest that greater support could be considered to enhance academic engagement and more research in this area could help determine how best to meet the needs of individuals studying through different modes (e.g. Tinklin, Riddell, and Wilson 2005;Goncalves and Trunk 2014).
There was an association with age and making multiple requests although this could be explained by the greater proportion of postgraduate students making multiple requests. Although previous studies (Zimmermann, Kamenetsky, and Pongracic 2015) have not found such associations, they have often recruited only undergraduate samples, consistent with the current findings when split by undergraduate or postgraduate status. The data could suggest that postgraduate students are more aware of the facility to submit SCRs (perhaps due to prior experience) or may represent a high-risk group for the experience of special circumstances.
Younger students (those aged under 25) were more likely to cite health reasons behind an SCR and equally likely to cite family issues, which was contrary to hypotheses (although numbers in the latter category were small). The data offer indirect support for epidemiological studies suggesting that young adults are experiencing a rise in health problems (Pitchforth et al. 2019;Potrebny et al. 2019) but more research on the relationship between health states and academic achievement is warranted (Eisenberg, Golberstein, and Hunt 2009). Given the association with student status, the relationship between student age and submission of SCRs would be interesting to explore further, particularly given that the age cutoff used in the current study was somewhat arbitrary and may obscure more subtle differences. These formative findings could be used to inform future studies which might consider (statistical) interactions between age and student status to look at associations with SCRs in more detail.
Findings can be seen in light of existing theories regarding university stress, such as the composite persistence model (Rovai 2003). This suggests that greater integration (e.g. with one's course or academic community) in addition to greater support might offset challenges faced throughout university (e.g. balancing employment and study). Given the possible impact of academic stress on mental health (Pitt et al. 2018; see also Tinklin, Riddell, and Wilson 2005), monitoring the frequency and nature of SCRs may identify students requiring additional support (OIAHE 2021). The proportion of SCRs submitted due to psychological reasons emphasises the association between poor mental health and academic submissions (martin 2010). Trends relating to multiple requests warrant further attention as this could indicate students with particular difficulties; in the current sample, around half made requests on different occasions during the course of one academic year, which can burden both students and university staff, as well as others outside of the university. For example, evidence of medical issues often requires healthcare professionals to provide a supporting letter and, in the current study, over half of requests providing evidence included a letter from a healthcare professional.
The issue of academic misconduct runs alongside that of SCRs. Although the current study cannot shed light on this as an explanation of the high rate of SCRs, other authors have raised the possibility of students submitting SCRs having feigned illness or falsified documents (Zimmermann, Kamenetsky, and Pongracic 2015). Those developing university policies should consider this possibility given the desirable incentives (e.g. extended deadlines) which may result from special consideration (Jasinski et al. 2011). Over one-third of SCRs in this sample had no documented 'evidence' and the integrity of evidence provided was not scrutinised. There was no association between provision of evidence and likelihood of the SCR being approved, and the requirement for evidence places further burden on the student and, in many cases, healthcare professionals. Therefore, universities should consider how to balance the need for evidence of impact in complex cases (alongside the associated burden) with the risks of duplicitous requests.
Alternative data analysis approaches were considered but rejected for several reasons. Individual-level data were not available on students not submitting SCRs, precluding detailed comparative analysis. For investigation of multiple requests, multivariate analyses (e.g. logistic regression) may have been strongly affected by high correlations between variables, as well as shortcomings in the application of regression models (e.g. forward stepwise; see Freedman, Pee, and midthune 1992). Thus, use of the chi-square test was appropriate given unequal group sizes and highly skewed data (mcHugh 2013). A further limitation is that the current study did not look in detail at the requests of students with 'declared' disabilities, an area on which future research might focus.
Given these caveats and taken alongside existing work, the findings might suggest that there is a degree of parity across demographics, with few groups more likely to submit SCRs. However, this interpretation is based on several assumptions, including that all those struggling will submit an SCR, and requires replication across different samples before any firm conclusions should be made. Findings regarding gender and mode of study warrant further attention and those regarding ethnicity should also be interpreted with caution as small numbers meant that ethnic minorities were analysed as a homogenous group. Although data from all eligible students were used, a wider sample would have improved validity and perhaps identified more subtle trends in the data (e.g. regarding 'miscellaneous' requests). Further, subsamples (e.g. those studying a particular degree) may not be representative of the wider student population and therefore findings should be interpreted with caution (Gainsbury, Russell, and Blaszczynski 2014).
Whilst the study is of one school, it offers a detailed insight into the reasons behind SCRs. Findings could be used to inform design of future studies in this area, helping researchers determine the strength of likely effects and thus appropriate sample sizes. Guidelines have recently been published by the UK Office of the Independent Adjudicator for Higher Education (OIAHE 2021) and could be taken alongside findings of the current study, recognising issues such as academic standards and the provision of evidence.

Conclusions
The current study is one of only a handful investigating SCRs in higher education, and the only one (to the author's knowledge) conducted in the UK. As such, results should be interpreted with caution but suggest that rates of SCRs may be increasing. Health issues were frequently cited as reasons for consideration and findings also suggest that postgraduate students may be particularly likely to submit requests at multiple time points. The needs of postgraduate students may warrant greater consideration, with dedicated support for both academic and personal issues (Lessing and Schulze 2002). Further research should aim to replicate the findings here, and might focus on the needs of previously under-represented groups, such as postgraduate students and those from ethnic minority groups.