Focus on Data linkage in education research


Data collection and analysis is a useful tool in identifying areas of need, improvement, trends and informing policy development. In education research, data analysis can provide insight, for example, on education outcomes and return on education, and transitions and pathways to further study and employment. Integrated datasets, or linked data, provide more information than what can be gleaned from single datasets, creating opportunity for wider analysis. Links can be made across different types of data, and between two or more datasets.

This Focus on... presents a sample of linked dataset research relating to vocational education and training (VET) in schools and dataset quality, tertiary education outcomes, and adult learning.

VET in schools research and dataset quality

The Australian Bureau of Statistics (ABS) has conducted studies on integrating vocational education and training in schools (VETiS) data with ABS Census data to determine the quality of a linked dataset. The 2013 feasibility study linked data from the ABS 2011 Census of Population and Housing with 2011 VETiS enrolment unit record level data provided by the National Centre for Vocational Education Research (NCVER). The study describes the data, the linkage process, and evaluation of the linkage. The evaluation determined that although the linkage method didn't produce a dataset of acceptable quality for analysis, it provided insight into improvements, including using more detailed geographic identifiers as well as weighting and calibrating the linked dataset for future linkage studies.

The ABS continued investigations of data linkage quality and in 2014 released Assessing the effect of geographic information on data linkage quality using vocational education and training in schools data, 2014, utilising improved methods of data linkage. The study linked three datasets: 2011 Queensland Curriculum and Assessment Authority (QCAA) VET in Schools, 2011 NCVER VET in Schools (Queensland records only), and 2011 ABS Census of Population and Housing (Census). The study showed that linkage rates rose when linking on small-area geography compared to large-area geography. Following the development of these improved data linkage methods, the ABS published Outcomes from vocational education and training in schools, experimental estimates, Australia, 2006-2011, which used a linked dataset of 2011 Census of Population and Housed integrated with 2006 VET in Schools data collected by NCVER, to provide insight into post-school destinations and outcomes of participating students.

In 2017, NCVER published VET in Schools students: characteristics and post-school employment and training experiences, which seeks to provide insights into both the uptake of VETiS programs since the mid-1990s and the further destination of VETiS students, with a focus on employment and training outcomes five years after completion. This research also used linked data from NCVER's 2006 national VET in Schools Collection with the data from the ABS 2011 Census of Population and Housing data. The report outlines the methodology and findings of the research.

More on this...

Tertiary education outcomes research

In Labor market returns to college education with vocational qualification, the author links data from the adult cohort of the National Educational Panel Study (NEPS-SC6-ADIAB) with administrative earnings data from the Sample of Integrated Labour Market Biographies (SIAB) to investigate the kind of labour market returns German non-traditional students can expect after enrolling in college instead of pursuing a vocational training-based career.

The adult cohort provides detailed life course information from birth to adult life for over 17,000 individuals born between 1944 and 1986, with an additional five waves covering employment and educational activities. Ninety per cent of the survey participants agreed to have their data linked, of which 74 per cent were linkable to SIAB, which covers the period 1975 to 2010.

What do students earn after their tertiary education? looks at a group of nearly 30,000 students who last enrolled in a tertiary education institution in New Zealand in 2003, and examines the influence of their tertiary education on their one-year and three-year post-study earnings. The study used data known as the Employment Outcomes of Tertiary Education (EOTE) dataset. EOTE was created by linking data on participation and achievement in institution-based and workplace-based tertiary education with employment data from the Linked Employer-Employee Database (LEED), which holds information on both employees and firms. LEED itself is a linked dataset, the result of linking longitudinal employment and income data on individuals with information on the firms they work for.

In Post-schooling outcomes of university graduates: a tax data linkage approach, a dataset of administrative data on students who graduated from the University of Ottawa was linked with Canadian tax record data. The authors report the analysis not only yielded an understanding of higher education (HE) learnings premia and returns to HE, but could also help inform young people in making schooling choices, and provide valuable information to individual HE institutions and policy makers.

How English domiciled graduate earnings vary with gender, institution attended, subject and socioeconomic background uses tax and student loan administrative data to measure how the earnings of English graduates around 10 years into the labour market vary with gender, institution attended, subject and socioeconomic background. A central focus for the research is what the diversity of graduates' earnings observed in the data may mean for social mobility, defined as the relationship between parental background (measured by an indicator of family income) and a child's eventual labour market success (measured by their income up to ten years after graduation).

More on this...

Adult learning research

The research conducted for the 'Labour Market Outcomes of Education and Training during Incarceration' project utilised the Western Australia prisoner education and welfare dataset, comprising five years of linked prison history, correctional education and income support payments. The research and final report, Welfare and recidivism outcomes of in-prison education and training, found prisoners who have up-skilled are less likely to recidivate (in terms of increased offence seriousness), and an increased number of successful classes will also reduce recidivism. In addition, ex-prisoners who are best able to remain in the community for longer have studied and successfully completed all their classes.

Other publications related to this project include Prisoner education and training, and other characteristics: Western Australia, July 2005 to June 2010 and Study in prison reduces recidivism and welfare dependence: a case study from Western Australia 2005-2010.

The labour market returns to further education for working adults investigates the labour market benefits obtained by working adults aged 25 to 64 years, who enrolled with a tertiary education provider and completed a certificate or diploma at levels 1-6, from 2003 to 2005. The main data source for this study was the Employment Outcomes of Tertiary Education (EOTE) dataset, created by Statistics New Zealand by linking administrative data on participation and achievement in the publicly funded tertiary education system, with administrative data on individuals' earnings and incomes from the Linked Employer-Employee Database (LEED). Survey of Families, Incomes and Employment (SoFIE), a longitudinal household survey that contains measures of educational achievement and hours worked that are not available in EOTE, was used as a secondary data source.

More on this...

Further research


Published: May 2018