Introduction

University students worldwide are facing increasing financial pressure. In the UK, higher education has become substantially more expensive in recent decades, with students leaving university with significantly higher debt than at the turn of the century [1–3]. Notable changes over this period have included increasing tuition fees for UK students [4, 5], higher inflation and increased cost of living [6, 7]. Growing concerns around student living costs, tuition fees and loan burdens have also been documented in Australia and Asia [8, 9]. In the US, it has been argued that rising debt levels among graduates, who typically take student loans, may both discourage potential applicants and motivate graduates to seek career paths based on speedier debt repayment [10]. For most students, starting university presents a cluster of major life transitions: typically, full-time students live for the first time away from home, and must adapt to newfound financial independence alongside new social and learning environments [11, 12]. Autonomously managing day-to-day spending requires students to negotiate multiple and potentially competing expenses, including rent and living costs, groceries, transport, tuition fees, social activities and other everyday expenses [13, 14]. University students often lack knowledge and experience in managing their money [15] and tend to make relatively poor financial decisions [16, 17]. Given also that students typically have little disposable income, money-management is often stressful for students, and can adversely impact their mental health and wellbeing [5, 18, 19]. Financial pressure can also lead to poorer academic performance [20]. While upstream structural changes may be needed to tackle the root causes of financial stress among students, behaviour change interventions that encourage students to manage their money optimally–by, for example, budgeting, or minimising unnecessary expenditure–may also be beneficial.

Developing effective money-management interventions for students requires an understanding of students’ money-management behaviours and their determinants [21]. Interventions should ideally be designed systematically, whereby knowledge surrounding the psychological determinants of that behaviour is used to select techniques to change behaviour via those determinants [22]. Psychological theory should be used to identify which determinants to focus on, and to organise these into a coherent framework and thereby generate hypotheses around how best to change behaviour [23]. Yet, much of the evidence regarding students’ money-management has focused on financial knowledge or literacy as outcomes of interest, rather than on specific financial behaviours [24–26]. Where studies have assessed student money-management behaviour, operationalisations of behaviour appear to have varied. For example, some research on ‘money-management’ or ‘financial behaviour’ among students has focused on specific behaviours such as budgeting [27], whereas other studies have focused on a broader range of actions [e.g., 14, 28]. Discrepancies among conceptualisations of ‘money-management behaviour’ makes it difficult to draw intervention design conclusions. Synthesising evidence surrounding the determinants of behaviour is an important preliminary step in intervention design [e.g., 21, 23]. Where evidence in a domain is heterogeneous, evidence synthesis should focus on summarising how a concept has been operationalised [29].

Scoping reviews aid intervention development by identifying and organising evidence surrounding how a target behaviour has been studied to date [29]. Whereas a ‘traditional’ systematic review would typically aim to establish the strength of relationships between a behaviour and its determinants, a scoping review aims to document what evidence exists in the area [29, 30]. By providing an assessment of the breadth and depth of a particular research field, scoping reviews can identify knowledge gaps that must be filled prior to intervention development [31].

The present study reports a scoping review designed to establish how money-management behaviour and its psychological determinants have been assessed in observational studies of university students. We reviewed both quantitative evidence, which has the potential to reveal statistical relationships between behaviour and its determinants, and qualitative evidence, which can reveal important experiences and meanings surrounding money-management among students. While intervention development requires understanding the psychological antecedents of behaviour, we reviewed evidence relating to any potential psychological correlate of behaviour to allow for evidence to be extracted from cross-sectional or retrospective studies.

The review aimed to address three research questions, one of which was operationalised differently for quantitative and qualitative studies:

1. 1. Which money-management behaviours have been studied among university students?
2. 2a. Which psychological correlates of students’ money-management behaviour have been assessed in quantitative studies?
3. 2b. Which experiences surrounding money-management behaviour have been explored in qualitative studies?
4. 3. Which psychological theories have been used to understand students’ money-management behaviours, and correlates and experiences surrounding those behaviours?

Materials and methods

This review is structured using applicable components of the Joanna Briggs Institute’s (JBI) framework [31], and was conducted using appropriate items from the PRISMA Extension for Scoping Reviews [32]. We did not publish a protocol for this study prior to execution of the review.

Study selection

Screening was undertaken by author TC. Of 1,005 papers identified in the database search, 226 were duplicates, and 748 were removed based on title and abstract screening. Of the remaining 41 papers, 29 were removed after full-text review, leaving 12 papers. In two instances, two papers each reported analyses of the same dataset, so were treated as single studies for review purposes [14, 25, 33, 34]. Thus, the 12 retained papers reported a total of 10 studies (see Fig 1).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. PRISMA flow chart.

https://doi.org/10.1371/journal.pone.0307137.g001

Results

Study characteristics

Setting and sample characteristics.

Of the ten studies, four were undertaken in the USA, two in Indonesia, and one in each of Australia, Hong Kong, Malaysia, and Sri Lanka. Participant age ranged from 18–40 years. Nine studies focused on undergraduate students only, and one focused on both undergraduates and postgraduates, but only undergraduate data are reported here [36].

Study designs.

Six studies used quantitative data only and were cross-sectional in nature; two studies used qualitative data only, with one using focus group data only and the other combining focus group and semi-structured interview data; and two studies used cross-sectional mixed-methods sequential designs, whereby quantitative data were collected (via questionnaires and surveys) prior to qualitative data collection. In one mixed-methods study, qualitative data were collected via open-ended, free-text responses to questionnaires, and in the other study, focus group. The 10 studies thus comprised a corpus of eight quantitative datasets and four qualitative datasets for review. All data in all 10 studies were self-reported.

Mean sample size was 392 in the quantitative-only studies, and 40 in the qualitative-only studies. In the mixed-methods studies, mean sample size was 474 in the quantitative component, and sample size for the qualitative component was only reported in one study [N = 28; 37]. All six quantitative-only studies aimed to understand the correlates of students’ behaviours.

Which money-management behaviours have been studied among university students?

Across the 10 studies, we identified 15 discrete subtypes of ‘money-management behaviour’, which we organised into five categories (Table 1). The ‘budgeting’ category included five behaviours that described some element of planning or reflecting on expenditure. At least one ‘budgeting’ behaviour was assessed in three quantitative-only, one qualitative-only and one mixed-methods study. ‘Borrowing’ encompassed three distinct behaviours, which were assessed across two quantitative-only and one mixed-methods studies. ‘Settling debts’ incorporated three behaviours, which were assessed across four quantitative-only and one qualitative-only study. ‘Spending’ included two behaviours, which were assessed in two quantitative-only, one qualitative-only and one mixed-methods study. ‘Saving’, which we defined in relation to the accumulation of funds (rather than minimising expenditure), incorporated two behaviours, assessed in two quantitative-only and one qualitative-only study. Six of the eight quantitative datasets used behaviour measures that integrated multiple behaviours (e.g., budgeting and settling debts).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Behaviours and categories of behaviours identified in all reviewed studies, and correlates modelled in quantitative studies.

https://doi.org/10.1371/journal.pone.0307137.t001

Which psychological correlates of students’ money-management behaviour have been assessed in quantitative studies?

Across the eight quantitative datasets, we identified 22 distinctive psychological correlates, which we clustered into six categories: Personality Characteristics, Financial Beliefs and Knowledge, Attitudes, Affective Responses, Self-Efficacy and Control, and Social Influences (Table 2).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Psychological correlates observed in eight quantitative datasets.

https://doi.org/10.1371/journal.pone.0307137.t002

Personality characteristics were examined in one of the eight quantitative datasets (13%), which found a small correlation between impulsivity and ‘planning finance’ (r = -.17).

Financial beliefs and knowledge variables were assessed in three of the eight studies (38%). Two out of the three showed that financial literacy had a positive relationship with ‘using credit cards’ and ‘making credit card payments in full’. One study found that ‘healthy financial behaviour’, as indexed by a combined measure of budgeting, settling debts, and saving behaviours (i.e., tracking expenses, maintaining a budget, making credit card payments in full, saving money, and investing) had a medium sized positive relationship with financial knowledge (r = .48) and a small-to-medium positive correlation with parental direct teaching (r = .43). Another study found that anticipated future income had no relationship with ‘using credit cards’.

Attitude variables were explored in three of the eight studies (38%). Two studies found a positive attitude-behaviour relationship, such that people with more financially prudent attitudes engaged more in ’healthy financial behaviour’ (i.e., combined measure of budgeting, settling debts, and saving behaviours). This relationship was moderate in one study (r = .56), and no coefficient was reported in the other study (p < .05). In another study, attitude towards budgeting had a medium-to-large positive correlation with ‘creating or maintaining a budget’ (r = .66).

Affective responses were assessed in three studies (38%). Two studies exploring anxiety respectively showed that higher anxiety was associated with higher ‘overspending’ (p < .02) and ‘maximising credit cards’ (p < .04; coefficients not reported). A combined measure of budgeting, settling debts, and saving behaviours showed a null relationship with anxiety (r = -.09), and a small positive correlation with financial well-being (r = .14). One study found that negative emotional responses to managing personal finances had a small negative relationship with ‘maintaining a budget’ (r = -.25).

Self-Efficacy and Control variables were assessed in three of the eight studies (38%). Two studies focused on perceived financial ability. One found perceived ability to have a small positive correlation with ‘budgeting’ (r = .26), and another showed a small-to-medium positive relationship with ’healthy financial behaviour’ (i.e. combined measure of regularly maintaining budget, saving, tracking expenses, use of credit card; r = .44). One study showed financial efficacy and financial controllability had a negative relationship with a combined measure of ‘using credit cards’ and ‘maximising credit cards’ (p’s < .01).

Social Influences were assessed in four of the eight studies (50%). One study showed that exposure to finance social media users and supportive peers was positively associated with a combined measure of budgeting, settling debts, and saving behaviours (p < .05). Another study found that parental norms were positively related with a combined measure of budgeting, settling debts, and saving behaviours (p < .001). One study of parental influences found that a combined measure of budgeting, settling debts, and saving behaviours showed a small positive correlation with parental subjective norms (r = .18) and small-to-medium positive correlations with parental descriptive norms (r = .34) and parental relationships (r = .26). Another study found that communication with parents and peers about finances had a positive relationship with ‘creating and maintaining a budget’ and ‘accruing money’, though no coefficients were reported.

Discussion

Financial pressure among students has the potential to adversely impact mental health and wellbeing and academic performance. The development of effective interventions to support students to manage their money effectively will be aided by understanding specific money-management behaviours and their psychological determinants. This scoping review was undertaken to identify how money-management behaviour has been studied to date, and document psychological correlates and students’ experiences surrounding money-management. We identified ten observational studies of student money-management behaviour, across which 15 discrete behaviours, within five distinct categories, were identified. We also observed six categories of psychological variables surrounding money-management. Variables pertaining to financial beliefs and knowledge, attitudes, self-efficacy and control, and social support from parents and peers were generally associated with ‘better’ money management practices. However, heterogeneity in the behaviours and correlates studied precludes definitive conclusions regarding which specific variables are consistently predictive of which specific behaviours. We documented underuse of psychological theory to organise and understand the determinants of money-management behaviours. Our findings suggest that terms such as ‘money-management behaviour’ or ‘financial behaviour’ are too broad. Future studies should seek to identify which specific forms of ‘money-management behaviour’ have most impact on mental health, wellbeing, academic and other outcomes, and which psychological determinants are most closely related with which money-related behaviours. Existing theoretical frameworks should be drawn on to identify a more comprehensive range of psychological determinants of money-management behaviours, and test hypotheses regarding how these may interact to determine students’ specific money-management behaviours.

The term ‘money-management behaviour’ is often implicitly used to denote a single, unidimensional behaviour [25, 35, 39, 40, 46]. However, we identified 15 distinctive ’money-management behaviours’ that have been studied in observational studies to date, and these could be sorted into five categories: budgeting, saving (building funds), spending borrowing, and settling debts. ‘Money-management’ thus encompasses a comprehensive range of practices [33], but few studies have explicitly conceived of these as distinct behaviours with potentially distinct determinants. Indeed, the behaviour measures used in most of the reviewed studies combined multiple actions, such as budgeting and building funds. Using such measures as outcomes precludes conclusions regarding which variables predict which specific behaviours. It is important that distinct behaviours are conceived of and studied distinctively, for several reasons. First, different money-management behaviours may differ in their impact on students’ health and wellbeing, academic and other outcomes [47]. For example, the ability to accumulate funds is closely linked to long-term financial security and has a protective effect against future stress, while settling debts on time can manage more immediate stressors and so protect wellbeing in the shorter term [48–50]. Conversely, excessive borrowing and mismanagement of credit cards has been found to prompt anxiety and worsen academic performance [39, 51]. Second, and relatedly, the impact of behavioural money-management interventions for students on outcomes such as health and wellbeing will vary according to which behaviour they target. Further research is needed regarding which specific money-related behaviours have greatest impact on students’ health and wellbeing, academic performance, and other outcomes.

We identified 22 discrete psychological variables examined as correlates of a money-management behaviour across eight quantitative datasets, and additional potential determinants and sequelae of money-management from four qualitative datasets. We clustered these into six categories, spanning personality characteristics, financial beliefs and knowledge, attitudes, affective responses, self-efficacy and control, and social influences. We found statistical relationships between variables within each of these categories and at least one form of money-management behaviour within the reviewed quantitative data. For example, impulsivity was associated with overspending [39], which resonates with broader psychological literature suggesting that personality is a core predictor of decision-making [52]. Similarly, affective responses, including anxiety and positive emotions, have been linked to overspending and budgeting [24, 42]. This aligns with psychological theory showing that emotional states can significantly impact decision-making processes [53]. All quantitative studies included in the review were cross-sectional in nature, which precludes conclusions around the causal direction of relationships between variables. However, insights from the qualitative data suggested that the impact of affective states on some money-management behaviours is bidirectional; specifically, negative affect prompted spending behaviour, which in turn worsened affective states. Together, these findings point both to the limitations of modelling statistical relationships based solely on cross-sectional, observational data, and the advantages of used qualitative evidence to explore and explain statistical findings. Given the dearth of evidence surrounding the determinants of students’ money-management behaviours, we recommend that mixed-methods designs are used to optimise insights generated by future studies in this area.

Syntheses of quantitative data showed that variables relating to financial beliefs and knowledge, attitudes, self-efficacy and control, and parental and peer social influence were predominantly related with ‘healthier’ money management practices, so should be harnessed as mechanisms for interventions to support students to better manage their money. Relationships between these variables and money-management practices appeared inconsistent, however. This may in part be because quantitative methodologies can oversimplify potentially complex phenomena. For example, parental norms and communication with parents and peers were each found to be positively associated with ‘better’ money management among students in quantitative studies, yet qualitative studies provided examples of students deliberately adopting positive money-management practices in response to their perceptions that their parents had managed their money badly [42]. Similarly, peer influence can have an adverse influence on money management when students engage in spending behaviours aimed at matching their peers, to maintain social standing [37]. The discrepancy in findings across studies underscores the complexity of social influences on money management behaviour [37, 54], and suggests that further research is needed to explore what determines whether others will have a positive or negative impact on money management. Other observed relationships may have lacked consistency across studies owing to heterogeneity in the specific behaviours and correlates studied. Such inconsistency is perhaps unsurprising given the considerable heterogeneity we observed in methods used to study money-management behaviour and its correlates. We showed that a diverse range of behaviours have been studied, a variety of factors have been modelled as correlates or studied as experiences, and multiple different measures have been used to document these factors. Robust estimation of relationships between two variables requires a consistent operationalisation of those variables, and standardised methods for capturing those variables, such as validated questionnaires.

We observed a lack of coherence in the selection of potential psychological correlates of students’ money-management behaviours in the literature to date. Notably, fewer than half of the reviewed studies drew on psychological theory. Using theory allows researchers to build on a comprehensive account of potential determinants and interrelationships between these determinants, and to identify potential psychological pathways through which behaviour may be changed. Of the four studies that used theory, only one tested the theory in its entirety, by modelling the impact of all theorised psychological determinants of risky spending and borrowing behaviours [38]. The remaining three studies sought to model some but not all hypothesised predictors of behaviour. Our findings thus suggest that theory has been underused in attempts to understand students’ money-management behaviours. We recommend that psychological theory is used more consistently and comprehensively in this domain, to better capitalise on existing knowledge regarding the determinants of behaviour.

While theory should ideally be used to inform the selection of variables to study as potential determinants of behaviour, it can also be used retrospectively, to organise determinants that have been studied into a coherent framework [23]. For example, the TPB proposes that behaviour is most closely influenced by intentions to perform the behaviour, which summates all sources of conscious motivation [44]. Intention is determined by three variables–i.e., attitudes towards the behaviour, social norms and perceptions of control over the behaviour–which in turn are shaped by personality characteristics [44]. Variables relating to personality, attitudes and beliefs, social environments, and control perceptions were observed across all studies but, except one notable study that tested the TPB in full [38], these areas have been under-examined in this domain. For example, the qualitative studies we reviewed offered rich coverage of social influences, highlighting the impact of parents, peers, social media, religion and culture, yet quantitative studies have used only a narrow conceptualisation of social pressures. The lack of research attention to the full breadth of motivational influences means that the relative strength of these determinants on specific money-management behaviours remains unclear. The apparent strategy of simply cataloguing potential determinants of behaviour fails to capture complex interplay among these variables; for example, control perceptions can influence attitudinal beliefs, and vice versa [44]. Additionally, only one of the reviewed studies included a measure of intentions, such that it was unclear in the nine other studies how strongly motivated participants were to engage in specific money-management behaviours. Future observational studies should use theory to ensure a comprehensive range of psychological influences on behaviour are captured.

None of the reviewed studies included any assessment of non-reflective processes. While the TPB focuses on conscious and reflective determinants of behaviour, such as evaluations of the overall favourability of action, dual-process models suggest that behaviour is shaped by a combination of reflective and non-reflective (i.e., automatic) processes. Much human behaviour is driven by automatic processes that can either support or undermine intentional actions [55]. For example, spending habits–i.e., actions cued automatically due to learned cue-behaviour associations [56]–can prompt unplanned purchases, especially among those with low momentary self-control [57, 58]. People may also develop habits that support well-being-conducive money-management behaviours, such as saving [57]. Automatic processes can also overpower conscious motivation, such that some students may momentarily engage in money-management behaviours that conflict with their attitudes, beliefs or values [59]. Future research should focus on the role of non-reflective factors in determining everyday money-related behaviours among students.

Limitations must be acknowledged. We reviewed only observational studies of correlates of, or experiences surrounding, money-management behaviour among undergraduate university students. This excluded research that may offer indirect insights into students’ money-management, such as intervention studies [e.g., 60], studies undertaken to predict proximal determinants of behaviour such as intentions [e.g., 61, 62], and studies undertaken among both student and non-student samples from which data relating to undergraduate students could not be isolated [e.g., 57, 63]. However, the development of effective behaviour change interventions requires an understanding of a behaviour change problem as it is experienced by members of the target population and other stakeholders [64].

Another limitation of our research is that we mostly summarised quantitative and qualitative data separately, rather than incorporating these into a mixed-methods synthesis. However, this limitation arises mostly from the lack of consistency in the literature to date. Mixed-methods analyses require a degree of homogeneity within the qualitative and quantitative datasets regarding the core behaviour of interest that was not available in our dataset [65]. Indeed, the limitations of our study call attention to important avenues for further research to better understand students’ money-management behaviour, including more consistency regarding behaviours of interest.

Conclusion

Developing effective interventions to support students’ money-management behaviours requires understanding which behaviour or behaviours to target, and which psychological variables determine these behaviours. Financial knowledge, attitudes, control perceptions, and social support variables were observed as correlates of money-management behaviours. Yet, considerable heterogeneity in the behaviours and potential determinants studied in the literature to date precludes clear recommendations for designing interventions to modify specific money-management actions among students. We recommend that researchers identify which specific money-management behaviours to target and why, and use comprehensive psychological theories, such as dual process models, to generate and test hypotheses regarding the determinants of money-management in students.

Supporting information