SOCIOECONOMIC FACTORS AND STUDENT ACHIEVEMENT IN QUEENSLAND

INTRODUCTION

Developing literacy and numeracy skills in the early years of schooling continues to be a key focus of recommendations around the Australian education system.¹ Literacy and numeracy are essential foundational skills that allow young people to succeed in life. The National Assessment Program – Literacy and Numeracy (NAPLAN) testing was introduced in Australia in 2008 to identify whether all students have the literacy and numeracy skills that provide the critical foundation for their learning, and for their productive and rewarding participation in the community.² Parents, carers and teachers play a crucial role in supporting a child's learning - from the early years through to adulthood.³

An enduring goal of educational, economic and social policy in Australia is to improve the educational outcomes of children, increasing the likelihood that they will transition into employment and participate fully in society.⁴ Studies show that students from low socioeconomic status families do not perform as well at school as children from higher socioeconomic status families.⁵ This is not to say that children from disadvantaged families cannot do well at school, as a child's innate ability is also an important influencing factor. However, understanding the socioeconomic factors that relate to educational disadvantage and how these factors interact is an area of ongoing interest.

Following on from an earlier ABS study using data from Tasmania,⁶ data from the Census of Population and Housing has been linked with Queensland government school enrolments and NAPLAN data to produce a powerful dataset that is capable of providing new insights on the role of parental and other socioeconomic characteristics on student achievement. This approach leverages more information from the combined dataset than is possible from the individual datasets taken separately. Importantly, it also enables analysis of specific population sub groups and analysis for small geographic regions.

This article demonstrates how the integrated dataset enhances the evidence base for social, economic and educational policy in Australia and how it can be used to explain the interaction between various socioeconomic factors and educational disadvantage. It also provides a number of starting points for future research.


DATA IN THIS ARTICLE

NAPLAN testing is completed by students in Years 3, 5, 7 and 9. Students are tested in five domains: numeracy, reading, writing, grammar, and spelling. This article looks at 2011 NAPLAN results for numeracy, reading and writing for government school students in Queensland that have been combined with various personal and socioeconomic characteristics collected in the 2011 ABS Census, using data integration techniques.

Data from all year levels has been combined and NAPLAN scores for reading, writing and numeracy have been grouped by whether a student met the national minimum standards for their year level. This article presents the proportion of students whose NAPLAN scores were below the national minimum standard. Students who score below this standard are at risk of being unable to progress satisfactorily at school without targeted intervention.⁷
Analysis was also undertaken to assess the influence on NAPLAN results of particular individual, parental, family and household characteristics. A logistic regression model was built to determine whether each of these characteristics had an effect on the odds of a student obtaining a score at or above the NAPLAN national minimum standard. The regression analysis allows the influence of each factor on a child's NAPLAN performance to be isolated. More information on the regression can be found in the Appendix: Socioeconomic Factors and Student Achievement - Results of Logistic Regression Analysis.

For more detailed information about data sources, definitions and linkage methodologies, see the Explanatory Notes tab.


PERSONAL CHARACTERISTICS AND STUDENT ACHIEVEMENT

Girls outperform boys on reading and writing

Overall, in Queensland government schools in 2011, 6% of students scored below the national minimum standard in numeracy, 9% in reading and 11% in writing.

Girls were substantially less likely than boys to score below the NAPLAN national minimum standard for writing, with 6% of girls scoring below the national minimum standard compared with 16% of boys. Girls also performed better than boys for reading, with 7% below the national minimum standard compared with 11% of boys. Boys and girls performed similarly for numeracy, with 6% of both boys and girls scoring below the national minimum standard.

One in three students with a disability score poorly on writing

ABS Census data includes a range of personal and household characteristics that are not currently available on NAPLAN data. One such data item is a measure of disability that identifies people needing help or assistance in self-care; mobility; or communication due to a long-term health condition or disability. It should be noted that students with significant disability which severely limits their capacity to participate may be exempted from participating in NAPLAN tests.⁸

One in three students with a need for assistance with a core activity scored below the national minimum standard for writing, compared with around one in ten without such a need. Students with a disability also scored more poorly on reading and numeracy, with nearly a quarter (23%) scoring below the national minimum standard for reading compared with 8% for those without a core activity need for assistance and 18% for numeracy compared with 6%.

Aboriginal and Torres Strait Islander students more likely to score below national minimum standard – particularly those living in Very Remote areas

Students who identified as being Aboriginal or Torres Strait Islander were more likely than non-Indigenous students to score below the national minimum standard on all three domains. For reading, 19% of Aboriginal students, 20% of Torres Strait Islander students and 23% of students who identified as both Aboriginal and Torres Strait Islander did not meet the national minimum standard compared with 6% of non-Indigenous students.

Interestingly, only students who identified as Torres Strait Islander scored similarly for all three domains. This was due to girls who identified as being of Torres Strait Islander origin performing considerably better for writing (with only 12% scoring below the national minimum standard) than for numeracy or reading (17% and 16% respectively).


Students living in Major Cities were less likely to score below the national minimum standard for all three domains than students living in Very Remote areas. This was true for students who identified as being Aboriginal and Torres Strait Islander and non-Indigenous students. For reading, 14% of Aboriginal or Torres Strait Islander students living in Major Cities scored below the national minimum standard, less than half the rate of Aboriginal or Torres Strait Islander students in Very Remote areas (34%). For non-Indigenous students, the rates of students scoring below national minimum standard for reading were 7% in Major Cities compared with 12% in Very Remote areas.

Overall, holding all other factors constant, those in Outer Regional, Remote or Very Remote areas were more likely to score below the national minimum standard compared with those in Major Cities. In general, the more remote the location, the higher the likelihood of a student scoring below the national minimum standard.

Children in major cities score more highly for reading

Mapping students' NAPLAN results by the geographic area in which they reside further highlights the effect of remoteness on a student's likelihood of scoring below the national minimum standard.

In 2011, 9% of students scored below the national minimum standard for reading in Queensland. However, students in some areas of Queensland fared better than others, with 'Brisbane Inner - West' having the lowest proportion of students below the national minimum standard (1%), followed closely by 'Sherwood - Indooroopilly' (1.6%). 'Brisbane Inner - East' and 'Kenmore - Brookfield - Moggill' also had very low proportions of students scoring below national minimum standard (both 2.3%).

At the other end of the spectrum, a quarter (25%) of students in the Far North were below the national minimum standard for reading in 2011. There was also a high proportion of students scoring below the national minimum standard in the regions of 'Outback - North' (19%), 'Cairns - South' (15%) and 'Innisfail - Cassowary Coast' (14%).



WHAT ROLE DO PARENTAL AND FAMILY CHARACTERISTICS HAVE ON STUDENT ACHIEVEMENT?

Students with highly educated parents do better on NAPLAN

The education level of their parents has a positive effect on a child's academic achievement. The more highly educated the father or mother, the better the child performs on numeracy, reading and writing. Less than 2% of students whose father had a Postgraduate or Graduate level qualification scored below the national minimum standard for reading, compared with 12% of students whose father's highest educational attainment was Year 11 or below. The same pattern was seen for mothers' qualifications, with 3% of students whose mother had a Postgraduate or Graduate level qualification scoring below the national minimum standard for numeracy, compared with 12% whose mother's highest educational attainment was Year 11 or below.


Children who had two highly educated parents (Certificate III or above) were more likely to perform better in all three domains than those who had only one highly educated parent. Only 2% of children in couple families where both parents had completed a Certificate III or higher scored below the national minimum standard for numeracy, compared with 5% of students in couple families where only one parent had completed a Certificate III or above. Likewise, students in a couple family where both parents had completed Year 12 were less likely to be below the national minimum standard (5%) than those from couple families where only one parent had completed Year 12 (7%).

Students from couple families with only one parent having completed a Certificate III or higher tended to perform slightly better on average than those from lone parent families with similar levels of education.


Students of younger mothers score more poorly for reading and writing

Students whose mother was younger at the time of the child's birth were more likely than children of older mothers to score below the national minimum standard for reading, writing and numeracy. In general, the proportion of students scoring below the NAPLAN national minimum standard declined as their mother's age increased, up to mothers who were aged 40 or more at the child's birth. Children of mothers who were aged 40 or more at the time of their birth tended to do worse as their mother's age increased.

After controlling for other factors (such as parental education), mother's age at the time of the child's birth had an effect on reading and writing scores, with students whose mother was aged in her thirties at the time of the child's birth being more likely to score at or above the national minimum standard than children of younger mothers.


Children of married couples perform best

Children from couple families were less likely to score below the NAPLAN national minimum standard than those from lone parent families (9% compared with 16%).

Children in opposite sex couple families tended to have better NAPLAN results if their parents were in a registered marriage rather than in a de facto relationship. For reading, 6% of children in opposite sex couple families where the couple were married did not meet the national minimum standard, compared with 12% in de facto opposite sex couple families and 8% in same sex couple families.

There was little difference in NAPLAN performance between children in male lone parent families and female lone parent families for numeracy or reading, though writing results were slightly worse for children with a lone father than those with a lone mother.

Much of the difference between family types is explained by other factors such as employment and education levels rather than the family type itself. For example, parents in a registered marriage tend to be more highly educated and are more likely to be employed than those in a de facto couple. However, even after accounting for these factors, children of married couples performed better on average than children of de facto couples for reading, writing and numeracy.


One in five foster children scored below national minimum standard for reading and numeracy

Natural or adopted children and those with a step-parent tended to do considerably better than other children on reading, writing and numeracy. Foster children, those living with their grandparents, otherwise related children (e.g. living with an older sibling or an aunt or uncle) or an unrelated child (e.g. staying with a friend's family) were around twice as likely as natural, adopted or step-children to not meet the national minimum standards for numeracy, reading or writing. For example, of children who were reported in the Census as being foster children, 21% did not meet the national minimum standard for reading, compared with 8% of children who were natural or adopted children of both parents or of a lone parent.

Children in smaller families have better NAPLAN scores

In general, as the number of children in a family grew, NAPLAN scores were more likely to be below the national minimum standard. However, children in one-child families were slightly more likely to score below the national minimum standard in all three domains than students from a two-child family. Nearly one in five children (19%) from families with six or more children achieved results below the national minimum standard for reading. Once controlling for other factors, such as parental education, students from families with four or more children are more likely to score below the national minimum standard than those in families with one or two children.


Students whose parents were born in Australia have poorer NAPLAN scores

Students with at least one parent born overseas tended to perform better on numeracy, reading and writing than those with both parents born in Australia. For example, 7% of those with both parents born in Australia scored below the national minimum standard for numeracy compared with around 5% of those with one or both parents born overseas. A similar pattern can be seen for reading and writing. When other factors were held constant, students with at least one parent born overseas were more likely to score at or above the national minimum standard for reading and writing, compared to students with both parents born in Australia.


The birthplace of the child did not have a consistent effect on NAPLAN results, although students who immigrated within the last five years were less likely to score at or above the national minimum standard than those who were born in Australia, or arrived more than five years ago.

Children with poor English proficiency score lower on NAPLAN

Not surprisingly, students who spoke a language other than English at home and had poor English proficiency were more likely to score below the national minimum standard. For example, 32% of those who spoke English 'not well' or 'not at all' scored below the national minimum standard for reading compared with 8% of those who spoke English 'very well'.

Students who spoke another language at home and spoke English 'very well' were slightly less likely to score below the national minimum standard for all three domains than students who spoke only English at home.

After controlling for the effects of the other variables, having good English skills was strongly associated with scoring at or above the national minimum standard for all three domains, compared with students who spoke English poorly or not at all.



WHAT ROLE DO HOUSEHOLD CHARACTERISTICS HAVE ON STUDENT ACHIEVEMENT?

Household income and location

Socio-Economic Indexes for Areas (SEIFA) are widely used measures of socio-economic advantage and disadvantage.

There is a strong relationship between the socio-economic status of the area in which the child lives, as measured by the SEIFA Index of Relative Socio-economic Advantage and Disadvantage, and NAPLAN performance. Students in the most disadvantaged areas were substantially more likely to score below the national minimum standard for each of the three domains than those in more advantaged areas. For example, 15% of students in the most disadvantaged quintile did not meet the national minimum standard for reading, compared with 4% in the most advantaged quintile.

However, SEIFA provides area-based measures of advantage and disadvantage, and it is important to consider that each household within a specific area has their own level of relative socio-economic advantage and disadvantage. There may be highly advantaged households in an area that is generally disadvantaged, and vice versa. A major benefit of the integrated dataset is that socio-economic measures can be explored at an individual or family level rather than drawing assumptions based on the socio-economic conditions of a particular area.

There was a substantial difference in the proportion of students that did not meet the national minimum standard within a SEIFA quintile, based on household income. In the most disadvantaged areas, 16% of students in low income households (less than $600 per week) scored below the national minimum standard for reading, compared with 11% of those in high income households ($3000 or more per week). Similarly, in the most advantaged quintile, 6% of students in low income households did not meet the national minimum standard, compared with just 2% of high income households. This highlights the importance of household characteristics in understanding NAPLAN performance rather than area-based measures alone.

Nevertheless, even after controlling for other factors using regression analysis, SEIFA still had a consistent influence on whether or not a student meets the national minimum standard for reading, writing and numeracy, suggesting that area-based disadvantage remains an important factor in explaining student performance even when direct measures of socio-economic status are available. In contrast, there was no clear relationship between household income and a child's NAPLAN performance once other factors were held constant. This is most likely because household income is strongly related to parental education, employment, and household composition, which are the underlying factors driving the differences in children's NAPLAN scores.


Parents' employment status influences NAPLAN, regardless of income

Children with employed parents performed considerably better on NAPLAN than those whose parents were not employed. Students with one parent employed (whether in a couple or single parent family) were less likely to score below the national minimum standard (e.g. 6% for numeracy) than those with no parents employed (13% for numeracy). Those in couple families with both parents employed were even less likely to score below the national minimum standard (4% for numeracy). This is true even when the household had the same income. For example, in couple families with a household income of between $2000 and $2499 a week, if one parent was employed 11% of children did not meet the national minimum standard for writing, compared with only 7% if both parents were employed.


Internet access related to better NAPLAN scores

Students living in a dwelling with an internet connection had considerably better NAPLAN scores than students with no internet access (regardless of their remoteness), but those with broadband tended to do best. The proportion of children with broadband who scored below the national minimum standard for reading varied from 7% in Queensland's Major Cities, up to 11% in Remote or Very Remote areas. In contrast, the proportion of students living in a dwelling with no internet connection who scored below the national minimum standard for reading varied from 15% in Major Cities, to 32% in Remote or Very Remote areas.

The relationship between internet access and NAPLAN performance also held after controlling for other factors such as household income, parental education, and Indigenous status.


Children of home owners tend to score better in NAPLAN

For reading, writing and numeracy, students living in a home that was owned (whether outright or with a mortgage) were more likely to score at or above the national minimum standard than those who lived in a rented dwelling. Students whose home was rented from a state or territory housing authority had the largest proportion below the national minimum standard, with 18% for numeracy, 21% for reading and 25% for writing. One in ten students (10%) whose home was rented from other landlord types, such as a housing co-operative or community group, scored below the national minimum standard for numeracy, with 16% for reading and 18% for writing.

Proportion of income spent on housing has little effect on NAPLAN scores

There were small differences in the proportion of students scoring below the national minimum standard, according to the proportion of household income spent on housing costs. For example, for reading, students in households where 30% or more of household income went to mortgage costs were slightly less likely to meet the national minimum standard than those in households where less than 30% went to mortgage costs. Similarly, the proportion of students in rented dwellings who scored below the national minimum standard also varied little with the proportion of rent paid, from 10% for those in households spending less than 30% on rent, to 11% for those spending 30% or more. However, once other factors were controlled for, whether or not a household spent 30% or more of it's income on housing costs did not appear to influence a child's NAPLAN performance.



LOOKING AHEAD

This study has demonstrated a range of personal, parental and socioeconomic information available on the ABS Census that can enhance the evidence base around school achievement. Through integrating the ABS Census with Queensland government school enrolments and NAPLAN data, the Census data items can be used to help understand and explain differences in literacy and numeracy performance between different groups. This is without the additional burden, cost and complexity of collecting this information directly from students, parents or teachers.

Using Queensland administrative data in conjunction with ABS Census data, this article provides further evidence that there is an important relationship between socioeconomic factors, such as parental education, employment and family composition, and NAPLAN scores. There is a consistent trend for children from households with better socioeconomic circumstances to perform better in NAPLAN, confirming that socioeconomic status and parental characteristics are an important factor in student performance. As the different measures of socioeconomic status are strongly related, a logistic regression was used to isolate the factors that contribute to a student's NAPLAN performance. This confirmed the considerable positive contribution of parental education and students' English proficiency. Until now, the potential factors that could improve performance for students from potentially disadvantaged backgrounds have not been able to be comprehensively explored due to data limitations.

Maximising the value of existing administrative data in conjunction with ABS Census data, particularly if undertaken at the national level, has the potential to substantially enhance the evidence base for social, economic and educational policy in Australia. This is a cost effective and efficient way of utilising existing data sources which does not increase collection burden on the general public.

While this report summarises the extent to which various personal, family and household characteristics influence school achievement, detailed reporting of these results over time and across grade levels would be beneficial. Particular areas for further work might include analysing the impact of additional characteristics such as student mobility or previous NAPLAN performance over multiple years. Analysis of later schooling outcomes such as retention rates or post-school outcomes such as further study or employment could also be performed, particularly as more data becomes available over the years.

There is also considerable scope for using the integrated data to provide richer insights into existing analytical findings. For example, the Queensland Government Department of Education, Training and Employment undertake a number of post-school destinations surveys and longitudinal studies that provide insights into the study and employment arrangements of young people after leaving school. The integrated dataset may be used to complement the findings from these destinations surveys, for example by providing information on the outcomes of school leavers with particular characteristics such as students with a disability and students from remote areas.


ENDNOTES

1. Donnelly, Kevin and Wiltshire, Kenneth, 2014, Review of the Australian Curriculum - Final Report, Accessed 14 October 2014
2. ACARA, NAPLAN - The Tests, Accessed 14 October 2014
3. Queensland Government Department of Education, Training and Employment, Literacy and Numeracy at school, Accessed 14 October 2014
4. Australian Government Department of Education, School education, Accessed 14 October 2014
5. Considine, Gillian and Zappala, Gianni, 2001, Factors Influencing the Educational Performance of Students from Disadvantaged Backgrounds, Accessed 14 October 2014
6. Australian Bureau of Statistics, 2014, Educational outcomes, experimental estimates, Tasmania, 2006-2013, Accessed 14 October 2014
7. ACARA, 2014, Standards, Accessed 14 October 2014
8. ACARA, NAPLAN Student Participation, Accessed 18 November 2014