The Scottish Government

« Previous | Contents | Next »

Listen

Chapter Four Predicting which pupils are likely to have a part-time job

Introduction

4.1 The descriptive statistics presented in chapter 3 show which pupils have part-time jobs and which do not. In compiling these descriptive statistics, we looked at each factor on its own. This is useful in describing part-time employment among school pupils in Scotland but it does not show us the inter-relationship between the various factors. Descriptive statistics do not let us see which factors are more important and which are less important in predicting whether or not pupils will have a part-time job. Statistical modelling, on the other hand, lets us see which factors are more influential than others.

4.2 In this chapter we shall see how in some cases, the results for certain factors change once we take the others into account in the statistical modelling. An example of this is pupils' ethnic background. We noted in chapter 3 that whether pupils had a part-time job or not varied according to their ethnic background but, as we shall see in this chapter, once we consider other factors alongside their ethnic background, we find that their ethnicity is much less important than other factors in predicting their likelihood of having a part-time job. We would not have known this if we had only relied on descriptive statistics.

4.3 In other cases, a factor remains significant, even after the other factors are controlled for. This means we can confidently state that this factor is influential in whether pupils do or do not have a part-time job. For example, the descriptive statistics showed that whether pupils had a part-time job or not was related to their stage of schooling. Using the statistical modelling approach confirmed that stage in schooling is important, even when we looked at factors such as social class and attainment. We can therefore be much more confident in saying that stage of schooling is important than if we had relied solely on the descriptive statistics.

The analysis

4.4 We examined the likelihood of pupils currently having a paid part-time job compared with never having had a part-time job. The analysis was built up in a series of steps or models to enable us to examine the effect of introducing each new set of factors. All the factors that proved significant in the prior model were carried forward to the next step while the non-significant factors were generally omitted from the next stage of the process.

4.5 The factors included in the analysis were:

Model 1: Background factors

• School stage
• Gender
• Attainment or study at Credit SG (study for S3s and S4s; attainment for S5s and S6s)
• Disability
• Ethnicity
• Father's social class; mother's social class
• Father's current activity; mother's current activity
• Living arrangements
• Location
• Unpaid work

Model 2: Attitudinal factors added

• Levels of truancy (self-reported)
• Attitudes to school - pupils' response to the following statements:
• school is helping to give me confidence to make decisions
• school is a waste of time
• school is doing very little to prepare me for life after school
• school work is worth doing
• my friends take school seriously
• The extent to which pupils view themselves as enterprising in their attitudes. This variable was constructed from the answers pupils gave to a series of questions designed to assess their level of enterprise.

Model 3: Enterprise in education inputs factors added

The factors added relate to the enterprise in education inputs that pupils had experienced. We have sub-divided these into: work experience; work related learning; enterprise education and career education.

• work experience
• work-related learning inputs (ie discussed how school subjects linked with industry and business; visited an employer or business; had talks from an employer/someone from industry or business)
• enterprise education inputs (ie discussed what's involved in running a business; taken part in an enterprise project/challenge day etc.)
• career inputs (ie visited college/university; had mock/practice job interview; prepared application/ CV; used career library/website; careers fair/convention; talks/visits from college/university staff; discussed career ideas with CA/someone from Careers Scotland)

Model 4: Factors relating to other time commitments added

• Unpaid work
• Frequency of pupils' leisure activities, including:
• Involvement in sports activity
• Involvement in groups or societies
• Disco, concert, cinema, sporting event
• Spend time with friends outside of school
• Watch TV
• Use computer (excluding school work)
• Voluntary work
• Extent of housework
• Extent of care duties

Model 5: Career related factors added

• Career focus
• Post school plans

4.6 We focus on the final model that contains all the factors that had proved significant in the earlier models with the addition of the final set of explanatory factors (Table 4.1). In Table 4.1 we do not present the full results but simply indicate whether a factor had a positive, negative or non-significant effect on predicting part-time employment. The full results are given in Table 1, Appendix 2. We consider each group of factors in turn and where appropriate, comment on relevant points arising from the earlier stages of the analysis.

Table 4.1: part-time job (logistic regression)

4.7 In Table 4.1 we indicate whether a factor had a positive, negative or non-significant effect on predicting whether pupils would have a current part-time job. The full results are given in Table 1, Appendix 2.

The effect of background factors

School stage

4.8 It was evident from the descriptive statistics that pupils' involvement in part-time work varied substantially across the year groups. The final results of the modelling confirms that pupils' stage of schooling remains a key factor in predicting their likelihood of current part-time employment even after controlling for a wide range of other variables. Compared with S3 pupils, those in the other year groups were more likely to be working part-time and the probability of doing so increases over the school years, especially between S5 and S6.

Gender

4.9 Another factor that remains significant in predicting current part-time work is gender: taking all other factors into account, girls were more likely than boys to currently have a paid part-time job than to have never worked. The exception to this is girls from a Pakistani background.

Ethnicity

4.10 The descriptive statistics showed that the likelihood of part-time work varied by pupils' ethnic background but the statistical modelling reveals that it is less important when other factors are taken into account. When ethnicity was considered in Model 1 alongside attainment, class and other background factors, most of its effect disappeared. The only difference in respect of pupils' ethnic background that remained concerned pupils from Pakistani backgrounds and there is a strong gender dimension to this. While boys from Pakistani backgrounds were more likely to have a part-time job than Scottish boys, Pakistani girls were less likely to have a job than Scottish girls. The involvement of pupils from ethnic minority backgrounds in unpaid work does not explain the results in respect of ethnicity; we examined involvement in unpaid work in earlier models and it made no difference to the effect of ethnicity.

Attainment

4.11 We saw in the chapter 3 that there was some limited variation in pupils' part-time work status according to their attainment level at Standard Grade ( SG). ⁶SG attainment level did prove significant in the early stage of the statistical modelling where attainment was considered along with other background factors ie school stage; gender; disability; ethnicity; parents' social class; parents' current activity; pupils' living arrangements; location and any unpaid work At this early stage of the analysis, the main contrast in pupils' probability of having a current part-time job was between those with 8+ Credit SGs - less likely to have a current job - and those in the other attainment groups. But once the various attitudinal factors were controlled for in the next stage of the analysis (Model 2), pupils' SG attainment level was no longer a significant predictor of their chances of having a current part-time job.

Family background

4.12 In respect of social class, Table 4.1 confirms the results of the descriptive findings: that it is pupils whose father is in the 'missing' category who were less likely to work part-time. Consistent with this finding, those who responded 'don't know' when asked to state their father's current activity, were also less likely to be working part-time than pupils whose father was in full time employment.

4.13 But it is notable that, after taking all the other factors into account, we found no difference in the likelihood of pupils from working class, intermediate and managerial and professional backgrounds having a current part-time job.

4.14 We would also note the impact of enterprise in education inputs on the effect of social class background and whether or not pupils are likely to have a current part-time job. In the earlier stages of the modelling (Models 1 and 2), having a father in the 'unclassified' category was associated with a reduced chance of being in current part-time work. (This is similar to the results of the descriptive analyses.) It was the addition of the enterprise in education inputs in Model 3 which changed the effect of this and having a father in the 'unclassified' category became non-significant at this point. It would appear that for pupils from this background, some of the enterprise in education inputs may be helping to compensate in some way for not having a father who is active in the labour market, perhaps by helping pupils develop the knowledge and skills to apply for a part-time job and/or by providing some direct contact with employers.

4.15 Table 4.1 shows that pupils who were living with a parent and step-parent were more likely to have a current part-time job compared with pupils living with their mother and father. This is the only situation relating to pupils' living arrangements that remained significant once other factors were controlled for.

4.16 The descriptive statistics had shown that pupils living in a lone parent household, who were with foster parents or in a children's home or at boarding school were all less likely to have a job. But these factors proved not to be significant when the other background factors were controlled for. Taking academic attainment into account made the key difference in respect of pupils who were staying with foster parents or being in a children's home. Similarly, living with one parent was not significant when parental social class was controlled for. These findings provide another illustration of the complexity of the factors that influence young people's involvement in part-time work and which is pertinent to concerns about equity issues.

Location

4.17 We saw in chapter 3 that a bigger proportion of pupils living in a rural area had a part-time job. We wondered if this would remain an important factor once we controlled for other factors in the statistical modelling. In fact location proved to be a consistently significant factor throughout the modelling process as other variables were taken into account. It is therefore a strong finding that, compared with pupils who live in large urban areas, those who live elsewhere, especially in rural locations are more likely to have a current part-time job.

The effect of attitudinal factors

Attitudes to school

4.18 In the modelling we included several factors - truancy, pupils' opinion of school, and the amount of time spent on homework - that might be seen as indicative of pupils' commitment and engagement with school. The conclusions from the modelling when we analysed these factors are similar to those from the descriptive statistics - there is little to suggest that pupils who are disengaged with school are more likely to be currently working.

4.19 The final model shows that pupils who have ever truanted are more likely to have a part-time job than those who have never truanted (Table 4.1). However, there is not a clear relationship between frequency of truanting and the likelihood of being involved in part-time work. The differences according to frequency of truanting are neither large nor straightforward, for example, pupils who reported truanting for a 'lesson here and there' had a higher probability of part-time work than those who truanted for 'days at a time'. The results therefore do not offer support to the view that there is a straightforward link between serious truancy and engagement in part-time work. It might also be noted that in the survey, when asked specifically if they had engaged in part-time work when they were absent, 86% of pupils who had truanted responded that they had not done so.

4.20 Pupils' attitudes to school do not appear to have much of an effect on their likelihood of currently having a part-time job once other factors are taken into account. Most of the items relating to pupils' attitude to school, for example that 'school is a waste of time', 'that school work is worth doing' became non significant as other factors were controlled for in the modelling. Disenchantment with school does not appear to be related to an increased probability of having a current part-time job. The only attitudinal factor which remained significant in the final model is the extent to which pupils thought that school is preparing them for life after school. Pupils who felt strongly that school is doing little in this respect were more likely to have a part-time job. This might suggest that some pupils actively seek out part-time work to gain practical experience. However, we need to be cautious in interpreting this since this 'attitude' may have emerged as a result of the employment experience. Longitudinal data would be needed to address this question.

4.21 There is little to suggest that a low level of time on homework is particularly related to having part-time work. Compared with those undertaking the greatest amount of homework (13+ hours per week), pupils spending less time than this on homework were more likely to be working part-time. However, Model 5 shows that those who did 9-12 hours of homework per week were almost as likely to have a current part-time job as those spending less time on homework (0-4 hours and 5-8 hours).

Enterprising attitudes

4.22 When we examined the extent to which pupils are enterprising in their attitudes and behaviours and their part-time work status in chapter 3, we saw that those with a low enterprise score were less likely to have a part-time job. It is notable this finding holds even after taking account of the wide range of other factors included in the modelling. Table 4.1 shows that pupils' level of enterprising attitudes has an independent effect on their probability of currently being in a part-time job. Compared with pupils who rated themselves to be the most enterprising, those who assessed themselves as less enterprising were less likely to be in a current part-time job; the probability of part-time employment decreased in line with the decrease in the level of enterprise.

The effect of enterprise in education factors

4.23 In the models we took account of the extent to which pupils had experienced certain enterprise in education inputs to assess if they had any effect on pupils' probability working. Work experience was not a significant factor in predicting involvement in part-time employment. However, two of the elements of enterprise in education inputs did increase the likelihood of current part-time work. Compared with those who had highest levels of work related learning inputs, other pupils were less likely to have a current part-time job and the chances of them doing so declined in line with the extent of the inputs they had received. Similarly, the extent of career related enterprise in education inputs had an independent effect on pupils' chances of having a part-time job all other factors being equal. In comparison with pupils who had the highest level of career related inputs, the others were less likely to have a part-time job. Enterprise education inputs were not significant.

4.24 A possible explanation for the lack of effect of enterprise education activities may be that for the cohorts we studied, the overall levels of enterprise related activities had been limited so that even those who had relatively high levels compared with other pupils still had not experienced extensive enterprise related inputs. The link between career and work related learning and part-time employment may be explained by the contribution exposure to this type of material makes to pupils' knowledge and self confidence. It is also possible that pupils may gain specific knowledge and skills which employers recognise.

The effect of pupils' other time commitments

4.25 In chapter 3 we examined pupils' activities and responsibilities outside of school and considered if they were associated with pupils' part-time work status. Here we assess if any of them were significant factors in predicting current part-time work once other factors are taken into account in the modelling process.

Involvement in sports

4.26 The level of pupils' participation in sport had some effect on their likelihood of having a current part-time job. There appears to be two aspects to this. Firstly, pupils with the highest level of participation (6 or more times in the past 2 weeks) were less likely than those who had had not taken part in any sports in that time. Secondly, pupils who had participated in sport 1-2 times were more likely to have a part-time job than those with no sporting activity. In the first case it seems that pupils with a high level of involvement in sport may rule themselves out of work because of the time commitment to their sport. In the second case, a moderate level of participation in sport may reflect a more general level of activity and engagement which is also reflected in having a part-time job. The latter suggestion is given more weight by the effect of the other factors related to how pupils spend their time out of school.

An active social life

4.27 A more active social life in terms of going out to discos, concerts, cinema or sporting events is a predictor of part-time employment. Compared with those who had not been to such events in the past two weeks, those who had done so 6 or more times were more likely to have a part-time job. Two possible explanations can be offered. First, previous research has shown that having a part-time job is a means of funding one's social life and these results may simply be an example of this. An alternative explanation is that it is the more 'sociable' 'outgoing' and 'active' pupils who are likely to be currently working, either because employers select them from the pool of labour or because they are more skilled at securing a job. It is worth noting in the chapter 3 we showed that a number of the never worked pupils had applied unsuccessfully for work. Model 5 also shows that compared with pupils who did not spend time with friends outside of school over the previous 2 weeks, those who did so were more likely to be in part-time employment.

4.28 Watching TV was a significant factor in predicting current part-time employment, the likelihood of having a current part-time job decreased in line with the frequency of watching TV. Previous research would suggest that as pupils gain employment other pastimes may be affected. However, pupils will not rate all pastimes equally and it may be that TV viewing diminishes in order to provide the time to maintain other social activities.

4.29 The results in respect of TV watching also indicates that it cannot be assumed that if someone is not working part-time then they will spend their time doing homework. This is supported by the findings on homework reported above where there is little difference in the likelihood of part-time work between those doing 0-4 hours per week and those doing 9-12 hours of homework

Household chores

4.30 When we consider the effect of helping with household chores, again it appears that a moderate level of activity is a predictor of part-time work. Compared with those who did not help with jobs around the house, those who reported a moderate level of housework (1- 2 and 3-5 times in the past 2 weeks) were more likely to have a part-time job. Those who had helped out more than 6 times over this period were less likely to have a current part-time job. This may be related to a lack of time given their involvement in household chores or possibly to the reduced need to earn money - we found that that those who had helped out 6+ times were slightly more likely to get money for doing so than pupils who helped out less frequently.

Caring responsibilities

4.31 In contrast, any level of involvement in caring for brothers or sisters, someone sick or disabled at home or a relative not living at home, increased the likelihood of having a part-time job compared with no involvement. The extent of caring duties does not make a big difference to the probability of having a part-time job but it does remain significant even after we have controlled for all the other factors. In these circumstances having a part-time job may act as a means of escaping such roles for at least sometime and establishing one's independence. However, it is clear that more information is needed before a definitive explanation of this result could be offered.

The effect of career related factors

4.32 The final set of factors that we examined was whether pupils' career ideas had any effect on the likelihood of currently working part-time.

4.33 It is sometimes suggested, for example, that pupils who are intending to enter the labour market directly from school are more likely to have part-time employment. Taking all other factors into account, we found that pupils who intended to find a job or training place after leaving school had a greater probability of being in a current part-time job than those who planned to go to HE. Pupils who were planning to take a time out ('year out/volunteer work/travel etc') were also more likely to be in part-time employment than those aiming for HE.

4.34 In the survey pupils were asked how far they had clear ideas about their career, we refer to this as 'career focus'. The extent to which pupils had a definite career focus increased the likelihood of working part-time. (This is not to make any assumptions about whether a definite career focus is good or bad.) Pupils who have no idea about what they want to do; whose career ideas keep changing; and who have some ideas but have not made up their minds yet were all less likely to have a part-time job than those who have had a definite career idea for a long time. The probability of not having a part-time job increased in line with career uncertainty. It may be that relationship between rates of part-time work and career focus is related to pupils' general outlook on life. A pupil's uncertainty and lack of focus about their career ideas may reflect a more general lack of focus in their life; this is consistent with some of the other findings in respect of pupils' activities and enterprising attitudes. Alternatively, it is possible that part-time work has resulted in pupils being able to perceive themselves in work roles and envisage themselves in the workplace. Both explanations have some value but the present research design does not allow us to unravel the issue of causality.

Overview

4.35 The approach taken in this chapter has enabled us to consider how a range of factors interact to influence pupils' involvement in current part-time employment. The results are important in deepening our understanding of differences in pupils' involvement in part-time work and in informing questions about equity in access to part-time work.

4.36 When all other factors are taken into account, school stage emerges as a key predictor of part-time work: it is clear that pupils are more likely to have a part-time job as they move up the school years. Pupils' location was another predictor of part-time work independent of any other factor; in particular, pupils in rural areas were more likely to have a part-time job than their counterparts in large urban areas.

4.37 Irrespective of other factors, young women have a greater likelihood of working part-time. We found that once other factors such as attainment and class were controlled for, the only difference that remained in respect of variation according to ethnicity concerned pupils from a Pakistani background. Girls from a Pakistani background were less likely to have a part-time job while their male counterparts were more likely to do so.

4.38 After we took various attitudinal factors into account, pupils' level of Standard Grade attainment was no longer a predictor of current part-time employment. Previous research indicates, however, that there is a complex relationship between part-time work and attainment and in later chapters we consider whether attainment makes a difference to the working hours and the quality of the job of pupils who do work part-time.

4.39 There is little to suggest that disenchantment with school (ie attitudes to school, time devoted to homework and truancy levels) means that pupils are more likely to engage in part-time work. On the specific question of any link between truancy and having a part-time work, no clear relationship is evident between the likelihood of currently having a part-time job and serious truancy.

4.40 The findings in respect of the influence of family background illustrate the complexity of the factors that influence young people's involvement in part-time work and which is pertinent to concern about equity issues. They also indicate that part-time work may function at several different levels for pupils, for example, pupils who lived with a parent and step-parent; these pupils were more likely to have a current part-time job than those who stayed with their mother and father. This may indicate that changing family dynamics impacts on a pupil's desire, or need, to have a part-time job. Previous research has shown that employment may provide an 'arena of comfort' for pupils whose families are undergoing change (Call, 1996).

4.41 We found that having higher levels of career or work related enterprise in education inputs increased pupils' chances of having a part-time job, independent of other factors. It may be that they increased pupils' knowledge and self confidence or helped them develop specific knowledge and skills which employers recognise. It is notable that for pupils whose father was not active in the labour market, some of the enterprise in education inputs appear to be compensating for this, perhaps by helping them develop the job seeking skills and/or by providing some direct contact with employers.

4.42 Pupils who judged themselves to have a more enterprising approach, were more likely to be in a part- time job even after controlling for other factors. This result is in keeping with the effect of other factors related to how pupils spend their time outside of school hours. All of this suggests that pupils who have relatively high level of activities in their life outside of school are more likely to have a part-time job irrespective of other factors. We have offered possible explanations for this in the chapter but note that more longitudinal data is needed to unravel this area. Other factors that, in themselves, made a difference to the likelihood of pupils having a part-time job are if pupils' intend to enter the labour market directly from school and also the extent to which they had definite career ideas. Both factors are linked with a higher probability of working but without longitudinal data it is not possible to explain the direction of the relationship.

« Previous | Contents | Next »