This chapter examines the relationship between experiences of economic deprivation such as food insecurity and missing school for economic reasons, and the use of sustainable learning strategies and attitudes towards learning. The chapter also looks at how working for pay before or after school relates to socio-economic background and the use of learning strategies.
PISA 2022 Results (Volume V)
Learning Strategies and Attitudes for Life
PISA
Abstract
For Australia*, Canada*, Denmark*, Hong Kong (China)*, Ireland*, Jamaica*, Latvia*, the Netherlands*, New Zealand*, Panama*, the United Kingdom* and the United States*, caution is advised when interpreting estimates because one or more PISA sampling standards were not met (see Reader’s Guide, Annexes A2 and A4).
Economic deprivation – the inability of households to meet their basic material and financial needs – limits students’ capacity to learn and develop learning strategies (Schenck-Fontaine and Ryan, 2022[1]). While the effects of economic deprivation on learning outcomes have been the subject of many studies, little is known about its effect on the development of strategies for sustained learning. This chapter examines how economic deprivation, and, particularly, food insecurity, relates to students’ use of sustainable learning strategies. The chapter also looks at indirect proxies for deprivation such as working part-time to earn money and the inability to attend school for financial reasons: the latter is described in Box V.7.2.
PISA finds that, unsurprisingly, 15-year-olds who experience food insecurity or work part-time during the school week tend to come from disadvantaged socio-economic status. In many PISA-participating countries and economies, they are more likely to be boys or come from an immigrant background.
PISA 2022 findings show that economic deprivation relates negatively to students’ ability to develop and use sustainable learning strategies. In particular, students are less likely to exhibit proactive learning strategies and to show control over their own learning.
There is a somewhat negative relationship between deprivation and young people being able to understand different points of view and ways of looking at the world. These are all hurdles to young people’s lifelong learning, keeping in mind, as well, that this is a student population at high risk of dropping out from formal education. Interestingly, students suffering from food insecurity or working for pay show similar levels of cognitive agility as other students when accounting for socio-economic status and mathematics performance.
On a bright note, students suffering from economic deprivation show as much as or even more interest in learning and curiosity about the world around them than their advantaged peers. This is true especially of students who hold part-time jobs1. They show marked creativity and persistence, and are positive about learning, particularly mathematics. These positive attitudes are a springboard to lifelong learning strategies that education systems can build on.
1. Experiences of economic deprivation are inter-related as students facing food insecurity are more likely to work for pay or take long leaves of absence from school to work or care for family members (Schenck-Fontaine and Ryan, 2022[1]). While accounting for socio-economic status helps isolate the effects of economic deprivation on students’ development of learning strategies, the relationship between experiences of economic deprivation and the development of learning strategies is not always straightforward. For example, students’ part-time employment is often taken on to mitigate their economic deprivation but it can also result in costly trade-offs that impede their learning as they have less time and energy for schoolwork.
A key dimension of material deprivation, food insecurity is defined as the inability to meet the nutritional needs of an individual (Aurino, Fledderjohann and Vellakkal, 2019[2]). Students participating in PISA were asked to report how many times a week they are not able to eat due to lack of money1. While not encompassing all dimensions of food security such as the nutrition level of meals, access to food is a good proxy for examining food insecurity among 15-years-olds (Aurino, Fledderjohann and Vellakkal, 2019[2]).
Food insecurity is widespread in all PISA-participating countries/economies
Food insecurity is observed in all PISA-participating countries and economies with available data. An average of 8% of students in OECD countries reported that there was at least one day a week in the previous 30 days they had not eaten because there was not enough money to buy food. Some 3% of students reported that this happens every day or almost every day (see Figure V.7.1).
In middle-income countries and economies, about one student in five, on average, reported that they did not have enough money to buy food at least once a week in the 30 days prior to the PISA test and 8% reported that they skip meals daily or almost daily due to lack of means. In high-income countries, 10% of students reported not eating at least once a week, on average (see Figure V.7.1).2
Percentage of students who reported not eating once a week or more because there was not enough money to buy food in the past 30 days
Note: Only countries and economies with available data are shown.
Countries and economies are ranked in descending order of the percentage of students who, at least once a week in the past 30 days, did not eat because there was not enough money to buy food.
Source: OECD, PISA 2022 Database, Table V.B1.7.2. See Table V.7.1 for StatLink at the end of this chapter.
See Table V.7.1 for StatLink at the end of this chapter.
Food insecurity is most widespread among immigrant students and boys
Immigrant students suffer more from food insecurity than non-immigrant students in many PISA-participating countries and economies (4.9 percentage-point difference, OECD average). This may be explained by immigrant students’ more precarious socio-economic status and overall exposure to economic deprivation in many PISA-participating countries/economies. In addition, boys reported slightly higher levels of food insecurity than girls (one percentage-point difference that is statistically significant) on average across OECD countries. While a gender difference in food insecurity is small among OECD countries (within two percentage-point difference), food insecurity is, nonetheless, more widespread among boys than girls in most PISA-participating countries, and exceeds 10 percentage points in difference in Jordan, Qatar, the Palestinian Authority and the Philippines (see Table V.B1.7.3).
Recurring food insecurity is related to lower learning outcomes
Students suffering from food insecurity tend to have lower learning outcomes in all countries and economies with available data. Students who reported not eating at least one day a week in the 30 days leading up to the test because of lack of money scored, on average, lower in mathematics (42 points difference after accounting for socio-economic profile across OECD countries) than their peers who never experience food insecurity (see Table V.B1.7.4). Hunger and poor nutrition lead to fatigue, limited attention-span and slower working memory. Students who suffer from food insecurity are more likely to work within or outside the household to help the family respond to food insecurity. Additionally, families facing food insecurity likely invest less time and financial resources in school education as the priority is to secure food (Aurino, Fledderjohann and Vellakkal, 2019[2]; Gallegos et al., 2021[3]).
Given the strong relationship between food insecurity and learning, many PISA-participating countries and economies are addressing students’ nutritional needs through school-level policy interventions such as free meals, food stamps or food vouchers to socio-economically disadvantaged students. For example, New Zealand’s school lunch programme, “Ka Ora; Ka Ako” has shown promising results in addressing food insecurity (see Box V.7.1).
Food insecurity is a challenge for many young people in New Zealand. Around 1 in 5 children live in households that face difficulties accessing good quality food.
New Zealand’s school-based healthy lunch programme, Ka Ora, Ka Ako, was first introduced in 2019 and aims to reduce food insecurity by providing lunches to students everyday. A study in 2022 showed that the programme had a positive impact on food security and student well-being, particularly among the most disadvantaged students who experience the greatest levels of food insecurity. The programme also improved school attendance among these students but not the overall population.
Ka Ora, Ka Ako provides free lunches to all students everyday in schools and kura (schools where teaching is based on Māori culture and values) with the highest number of students experiencing socio-economic barriers. Instead of targeting students on the basis of individual need, this reduces the stigma associated with receiving free meals and ensures that students receive the lunches they need. In May 2024, the programme served over 236 000 students in 1 013 schools and kura.
Schools and kura are selected based on the Ministry of Education’s Equity Index (EQI), which measures the extent to which students at school may face socio-economic barriers that could impact their academic achievements. Currently, participating schools can choose to make their own lunches internally or provide lunches through an approved external supplier. There is currently no set lunch menu for the programme, and schools and suppliers can decide what to include in their lunches. The cost of lunches depends on the model of school lunch delivery and age of the learners.
Part-time employment is the experience of holding a job and receiving pay for work outside of schooling hours (Holford, 2020[8]). Students participating in PISA were asked to report how many days per week in a typical school week they had worked before or after school3. Part-time employment is not unusual among adolescents in many countries/economies in PISA. Some 30% of 15-year-old students work at least once a week before or after school and 10% reported working at least 5 times per week before or after school, on average across OECD countries. In some countries, more than half of students hold part-time employment. This is the case, for example, in the Netherlands* where 70% of 15-year-old students reported working part-time at least once a week. The share of students working for pay is higher among middle-income countries and economies participating in PISA (33%) than high-income countries (26%). Similarly, the share of intensive part-time employment (at least five times a week) is higher in middle-income countries and economies in PISA (18%) than high-income ones (10%) (see Figure V.7.2).
Percentage of students working for pay before or after school during a typical school week
Note: Only countries and economies with available data are shown.
Countries and economies are ranked in descending order of the percentage of students working for pay before or after school at least once a week.
Source: OECD, PISA 2022 Database, Table V.B1.7.12. See Table V.7.1 for StatLink at the end of this chapter.
While not a measure of economic deprivation, part-time employment tends to be most prominent among students facing high levels of poverty or economic deprivation in most PISA-participating countries and economies. Socio-economically disadvantaged students generally hold part-time jobs more than advantaged students (4.5 percentage-point difference, on average across OECD countries). However, in 20 participating countries and economies, there is no difference in the share of disadvantaged and advantaged students working for pay at least once a week and in 8 countries and economies advantaged students report more instances of part-time employment than disadvantaged students: Albania, Canada*, Germany, Kosovo, New Zealand*, Norway, Sweden and the United Kingdom*. This is particularly the case in some OECD countries where a large share of the 15-year-old student population holds a part-time job, such as Australia*, Denmark*, the Dominican Republic and the Netherlands* (see Table V.B1.7.12 and Table V.B1.7.14).
Socio-economically disadvantaged students tend to also work more intensively during the school week (+0.42 points more on the work-for-pay index, on average across OECD countries) (see Table V.B1.7.15). While among OECD and high-income countries and economies, slightly more advantaged students than disadvantaged work once or twice a week (+1.1 percentage-point difference among high-income countries and economies, and +1.9 percentage-point difference among OECD countries), more disadvantaged students than advantaged work for pay five times a week or more (-4.6 percentage-point difference among high-income countries and -4.8 percentage points among OECD countries) (see Figure V.7.3).
Percentage-point difference between advantaged and disavantaged students reporting to work for pay before or after school
Notes: Only countries and economies with available data are shown.
A socio-economically disadvantaged (advantaged) student is a student in the bottom (top) quarter of the PISA index of economic, social and cultural status (ESCS) in his or her own country/economy.
Statistically significant differences are shown in a darker tone (see Annex A3).
Countries and economies are ranked in descending order of the difference of students who never work for pay related to socio-economic status (advantaged minus disadvantaged students).
Source: OECD, PISA 2022 Database, Table V.B1.7.13. See Table V.7.1 for StatLink at the end of this chapter.
Part-time employment is most prominent among boys
Boys reported working for pay outside of school hours at least once a week more than girls in almost all PISA-participating countries and economies. Among OECD countries, 33% of boys on average reported working for pay against 27% of girls (see Table V.B1.7.14). This gender gap is particularly pronounced among middle-income countries and economies participating in PISA (19.4 percentage-point difference on average among middle-income countries/economies compared to 6.6 on average among OECD and high-income countries and economies). Additionally, boys have a higher intensity of work during the school week than girls (+0.5 point increase in the work‑for‑pay index on average across OECD countries) (see Table V.B1.7.15). This may be explained by the gendered socio-cultural division of labour in which girls are more likely to work in the household or take care of family members in many PISA-participating countries and economies (Hayford and Halliday Hardie, 2020[9]; UNICEF, 2022[10]).
Part-time employment is related to lower mathematics performance
Students who spend more time on part-time employment tend to perform worse in mathematics in PISA even after accounting for socio-economic profile (see Table V.B1.7.18). This finding corroborates national studies from the United States and the United Kingdom that find a negative relationship between part-time employment among adolescents and their learning outcomes, in particular, for intensive work (Staff et al., 2020[11]; Holford, 2020[8]). Intensive part-time employment also increases the risk of drop-out and fading out from education as students invest more time in work and are not able to keep up with schoolwork demands (Staff et al., 2020[11]). This also heightens the risks of students not developing needed learning strategies to engage with lifelong learning.
Experiences of food insecurity and working for pay before or after school are related to somewhat lower levels of development and use of strategies for sustained learning even when accounting for students’ and schools’ socio-economic profile, and students’ mathematics performance. Students with these life experiences are less likely to be in control of their own learning or demonstrate critical-thinking skills in their decision making. Poorer learning strategies among an already vulnerable student population exacerbates the risk of drop-out from formal education and may limit lifelong learning in the future.
Students suffering from economic deprivation are somewhat less likely to be proactive and in control of their own learning
Students suffering from food insecurity or who hold part-time jobs are somewhat less likely to report using some self-regulated learning strategies such as carefully checking their homework, on average across OECD countries, even after considering students’ and schools’ socio-economic profile and students’ mathematics performance. In general, these students also show lower levels of proactive study, on average across OECD countries. This includes making sure there are no mistakes or connecting new material to what was learnt previously – although they are as likely to connect new material to what they learned previously once students’ and schools’ socio-economic profile and students’ mathematics performance were accounted for (see Figure V.7.4). Economic deprivation seems to distract students from schoolwork. They are less likely to have the time or energy for proactive and self-regulated learning, making them passive learners (Gallegos et al., 2021[3]).
Likelihood of reporting using learning strategies associated with control of one's own learning and proactive learning when students suffer from food insecurity or when they work for pay; OECD average
1. The socio-economic profile is measured by the PISA index of economic, social and cultural status (ESCS).
Note: Odds ratio coefficients that are statistically significant are shown in a darker tone (see Annex A3).
Items are ranked (for each graph) in ascending order of the odds ratio, after accounting for students' and schools' socio-economic profile and performance in mathematics.
Source: OECD, PISA 2022 Database, Tables V.B1.7.6 and V.B1.7.20. See Table V.7.1 for StatLink at the end of this chapter.
Economic deprivation relates somewhat negatively to critical thinking but not to cognitive agility
Economic deprivation relates to somewhat less development and use of critical-thinking strategies for sustained learning. Students suffering from food insecurity are 7% less likely to consider everybody's perspective before taking a position, after accounting for socio-economic profile and mathematics performance, on average across OECD countries. They also exhibit slightly less cognitive agility and are more likely to agree that there is only one correct position in a disagreement before accounting for students’ and schools’ socio-economic profile and mathematics performance. It is worth noting that no significant difference is observed in cognitive agility once socio-economic profile and mathematics performance are accounted for (see Figure V.7.5).
Students with part-time jobs also show less developed critical-thinking strategies. Students working for pay at least once a week are 20% less likely than those who do not to consider others’ perspectives before making a decision. They are also more likely to think there is only one correct position in a disagreement (30% more likely, on average across OECD countries) (see Figure V.7.5).
Likelihood of reporting using learning strategies associated with critical thinking when students suffer from food insecurity or when they work for pay; OECD average
1. The socio-economic profile is measured by the PISA index of economic, social and cultural status (ESCS).
Note: Odds ratio coefficients that are statistically significant are shown in a darker tone (see Annex A3).
Items are ranked (for each graph) in ascending order of the odds ratio, after accounting for students' and schools' socio-economic profile and performance in mathematics.
Source: OECD, PISA 2022 Database, Tables V.B1.7.8 and V.B1.7.22. See Table V.7.1 for StatLink at the end of this chapter.
Students facing economic deprivation or holding part-time employment are motivated to learn and are curious about the world. While economic deprivation in the form of food insecurity is related to negative self-beliefs about learning, students holding part-time jobs, on the other hand, tend to feel more positive about learning – in particular, in mathematics. And economic deprivation does not impede students’ motivation to learn or their interest in learning. Students with part-time jobs may feel even more motivated to learn. These findings provide policy makers with valuable insights. Flexible learning systems that give economically deprived students opportunities for combining their study with work or re-entry points into schooling can keep students facing economic deprivation from fading out from formal education.
Economic deprivation is related to heightened mathematics anxiety but part-time student employment is associated with slightly lower mathematics anxiety
Students in OECD countries who experience food insecurity reported higher levels of mathematics anxiety than their peers, on average (0.24 points higher at the OECD average), even when accounting for socio-economic profile and mathematics performance (0.08 at the OECD average) (see Figure V.7.6). It is probable that this anxiety prevents these students from developing sustainable learning strategies (see Chapter 4). Students who work for pay have somewhat similar levels of mathematics anxiety than those who do not work before accounting for socio-economic profile and mathematics performance (see Figure V.7.6). The sense of a lack of self-efficacy and agency over one’s own learning may also heighten mathematics anxiety among economically deprived students. However, once socio-economic profile and mathematics performance is accounted for, having a job and earning money correlates with a slight reduction in mathematics anxiety, on average across OECD countries (see Figure V.7.6 and Table V.7.25). This may point to a positive effect of using mathematical skills and knowledge in an applied setting (workplace) on mathematics anxiety. This finding opens up new possibilities for students at risk of dropping out to re-engage with mathematics and develop learning strategies. Education systems might look at applied learning opportunities and connecting learning more closely to students’ lived experiences outside of the classroom.
Change in mathematics anxiety when students reported suffering from food insecurity or working for pay
1. The socio-economic profile is measured by the PISA index of economic, social and cultural status (ESCS).
Notes: Only countries and economies with available data are shown.
Change in the index of mathematics anxiety that are statistically significant are shown in a darker tone (see Annex A3).
Countries and economies are ranked, within each chart, in descending order of the change of the index of mathematics anxiety, after accounting for students' and schools' socio-economic profile and mathematics performance.
Source: OECD, PISA 2022 Database, Tables V.B1.7.11 and V.B1.7.25. See Table V.7.1 for StatLink at the end of this chapter.
Students experiencing economic deprivation are motivated to learn and interested in learning
Although lacking sustainable learning strategies, students suffering from economic deprivation are as motivated to learn as their peers. In some cases, they are even more motivated. When accounting for both socio-economic profile and mathematical performance, students who experience food insecurity are just as curious and open-minded about the world as their peers who do not have food insecurity, on average across OECD countries (see Table V.B1.7.10 and Table V.B1.7.11).
Student employment tends to be associated with higher levels of curiosity, creativity and persistence in learning after accounting for socio-economic profile and mathematic performance. Students who work part-time are more likely to report being persistent and enjoying challenging schoolwork than those who do not, on average across OECD countries. They are also more likely to demonstrate creativity by thinking up new ways to solve problems (see Figure V.7.7). Finally, they also demonstrate slightly higher levels of curiosity about the world than others, on average across OECD countries (see Table V.B1.7.25). This positive disposition towards learning in both formal and informal settings may come from students’ exposure to work situations and interaction requiring the application of skills and knowledge. It may also result from a self-selection process where students most motivated to learn are also more likely to work (Mortimer, 2010[12]).
Likelihood of reporting using strategies associated with persistence and creative thinking when students work for pay; OECD average
1. The socio-economic profile is measured by the PISA index of economic, social and cultural status (ESCS).
Note: All odds ratio coefficients are statistically significant (see Annex A3).
Items are ranked (for each graph) in ascending order of the odds ratio, after accounting for students' and schools' socio-economic profile and performance in mathematics.
Source: OECD, PISA 2022 Database, Table V.B1.7.24. See Table V.7.1 for StatLink at the end of this chapter.
PISA asked students to report if they had missed school for three months or more in primary or secondary education and the reasons for this long-term absenteeism. Some of these reasons relate directly to economic deprivation such as taking care of a family member; helping with work at home, the family business or on the family land; getting work to bring money home; or not being able to pay for school fees. Only a small minority of students have experienced long-term absenteeism for these economic reasons (1.5% on average across OECD countries; 1.8% in high-income countries and economies and reaching 4.7% among middle-income countries and economies). While only a minority of students are concerned, the consequences for students’ learning are dire. Long disruptions to schooling for economic reasons are related to lower learning outcomes (58 score points lower in mathematics after accounting for students’ and schools’ socio-economic profile among OECD countries, and 65 and 49 score points lower among high-income and middle-income countries, respectively) (see Table V.B1.7.27).
Missing school for an extensive amount of time is not only related to lower learning outcomes for 15-year-old students but lower use of sustained learning strategies and, in particular, the ability to control one’s own learning. Missing school for an extended period for economic reasons is associated with less control of one’s own learning strategies. For example, students who experience long-term absenteeism for economic reasons are less likely to report making sure there are no mistakes in their work or checking their homework before turning it in (31% and 27% less likely, respectively, among OECD countries after accounting for socio-economic profile and mathematics performance). The lack of control of their learning may further exacerbate and widen the gap between students who missed school for an extended period of time for economic reasons and those who did not (see Figure V.7.8)
Likelihood of reporting using learning strategies associated with control of one's own learning and proactive learning when students missed school for more than three months in a row for economic reasons; OECD average
1. The socio-economic profile is measured by the PISA index of economic, social and cultural status (ESCS).
Note: Odds ratio coefficients that are statistically significant are shown in a darker tone (see Annex A3).
Items are ranked in ascending order of the odds ratio, after accounting for students' and schools' socio-economic profile and performance in mathematics.
Source: OECD, PISA 2022 Database, Table V.B1.7.28. See Table V.7.1 for StatLink at the end of this chapter.
References
[2] Aurino, E., J. Fledderjohann and S. Vellakkal (2019), “Inequalities in adolescent learning: Does the timing and persistence of food insecurity at home matter?”, Economics of Education Review, Vol. 70, pp. 94-108, https://doi.org/10.1016/j.econedurev.2019.03.003.
[3] Gallegos, D. et al. (2021), “Food Insecurity and Child Development: A State-of-the-Art Review”, Int J Environ Res Public Health, Vol. 18(17)/Aug 26, https://doi.org/10.3390/ijerph18178990.
[9] Hayford, S. and J. Halliday Hardie (2020), “Gender Differences in Adolescents’ Work and Family Orientations in the United States”, The Sociological Quarterly, Vol. 62/3, pp. 488-509, https://doi.org/10.1080/00380253.2020.1775529.
[8] Holford, A. (2020), “Youth employment, academic performance and labour market outcomes: Production functions and policy effects,”, Labour Economics, Vol. 63, https://doi.org/10.1016/j.labeco.2020.101806.
[4] Ministry of Education (2024), Different types of primary and intermediate schools, https://parents.education.govt.nz/primary-school/schooling-in-nz/different-types-of-primary-and-intermediate-schools/ (accessed on 24 May 2024).
[5] Ministry of Education (2024), Ka Ora, Ka Ako | Healthy School Lunches Programme, https://www.education.govt.nz/our-work/overall-strategies-and-policies/wellbeing-in-education/free-and-healthy-school-lunches/ (accessed on 24 May 2024).
[6] Ministry of Education (2024), Ka Ora, Ka Ako | New Zealand Healthy School Lunches Programme Supplementary Attendance Analysis for Most Underdeserved Ākonga, https://assets.education.govt.nz/public/Documents/our-work/strategies-and-policies/Ka-Ora-Ka-Ako-attendance-18-03-24.pdf (accessed on 24 May 2024).
[12] Mortimer, J. (2010), The Benefits and Risks of Adolescent Employment.
[1] Schenck-Fontaine, A. and R. Ryan (2022), “Poverty, Material Hardship, and Children’s Outcomes: A Nuanced Understanding of Material Hardship in Childhood”, Children, Vol. 9/7, p. 981, https://doi.org/10.3390/children9070981.
[11] Staff, J. et al. (2020), “Is Adolescent Employment Still a Risk Factor for High School Dropout?”, J Res Adolesc, pp. 406-222, https://doi.org/10.1111/jora.12533.
[10] UNICEF (2022), “Time spent on household chores”, Multiple Indicator Cluster Surveys, Demographic and Health Surveys, 2005–2020, https://data.unicef.org/adp/snapshots/transition-to-work/#chores (accessed on 5 June 2024).
[7] Vermillion Peirce, P. et al. (2022), Ka Ora, Ka Ako | New Zealand Healthy School Lunches Programme Impact Evaluation, Ministry of Education, https://assets.education.govt.nz/public/Documents/Ka-Ora-Ka-Ako/Ka-Ora-Ka-Ako-Evaluation_Impact-Report-Final-v2.pdf (accessed on 24 May 2024).
← 1. Students participating in PISA 2022 were asked “In the past 30 days, how often did you not eat because there was not enough money to buy food?” with five response options (“Never or almost never”, “Almost once a week”, “2 to 3 times a week”, “4 to 5 times a week” and “Every day or almost every day”).
← 2. For the classification of countries by economic development levels, please refer to Table V.B1.7.1. High-income countries and economies are defined as countries or economies with GNI per capita superior to USD 13 845 in year 2022 (source: World Bank Analytical Classification, presented in the World Development Indicators, https://datahelpdesk.worldbank.org/knowledgebase/articles/378834-how-does-the-world-bank-classify-countries). Middle-income countries refer to low and upper middle-income countries and economies defined as countries or economies with a GNI per capita between USD 1 136 and USD 13 205 in year 2022 (source: World Bank Analytical Classification, presented in the World Development Indicators, https://datahelpdesk.worldbank.org/knowledgebase/articles/378834-how-does-the-world-bank-classify-countries).
← 3. The index of work for pay is a simple derived variable from questionnaire items in the PISA 2022 student questionnaire. Students’ answers on how many days during a typical school week they worked for pay before going to school and/or after leaving school were scaled into the index of “Work for pay before or after school”. Each item included six response options (“0 days”, “1 day”, “2 days”, “3 days”, “4 days”, “5 or more days”). Values on this index range from 0 (no work for pay) to 10 (10 or more times of working for pay per week).