By Sue Wight
Whether we home educate from the start or as a result of school problems, we observe our kids learning firsthand. Most of the time, that’s all the evidence we need. But many of us have occasional doubts and have relatives who want to know if we are limiting the children’s future. Governments and media also want solid data on home education outcomes.
So what information is available on outcomes?
Over 2016-17 the Home Education Network (HEN) ran the largest survey of home educated alumni ever conducted in Australia. The most important outcome revealed in our alumni survey was that grown home educators are happy with their lives, satisfied with their education, and well-prepared for adulthood.
Education Departments in Victoria, New South Wales and South Australia have each collated academic results. In the following cases, sample sizes were small or self-selected, rendering results of only passing interest:
- Home educated students who sat the NAPLAN tests were above state averages.
- The Year 12 results of previously home educated students met (NSW) or exceeded (Vic) those of always-schooled students.
- The rate of previously home educated students starting a bachelor’s degree in Victoria was above the state average.
While those results are interesting, the department studies also include results of significance. These involve home educated kids who had transferred to school and sat tests alongside their school fellows. The results are particularly interesting because:
- the number of students is large;
- the tests are state-run;
- no self-selection was involved; and
- school statistics provide control groups.
In Victoria, the department examined 1879 NAPLAN Reading results across 2013-2015. At Grade 3 level, previously home educated kids scored behind their always-schooled peers. However, results improved with age and, by Year 9, exceeded the always-schooled cohort by a small margin. The 2017 Regulatory Impact Statement dismissed these results as unreliable on the basis the students had spent some time in school. This was a failure of logic — the difference between the two groups was their home education, not schooling.
In NSW, a more detailed study by the Board of Studies (BOSTES) looked at NAPLAN results for 3965 previously home educated students across 2008-2013. The number of students was almost double the size of the largest NSW public school. The result:
Students who were previously home schooled scored significantly higher in the Reading and Grammar & Punctuation tests, generally by about 20-25 marks (or about one-third of a standard deviation) in Reading, and 15-25 marks in Grammar & Punctuation. In Writing, Spelling and Numeracy, the home schooled student averages are generally not statistically significantly different from the overall NSW average. These results were consistent, regardless of the Year Levels being tested, and whether students have been registered for home schooling for shorter or longer periods.
The NSW study also compared over 1000 Year 10 results across English, Science and Maths and found them comparable with state averages if students had been home educated for less than two years. Marks were “significantly higher than the NSW average on the majority of occasions” for students home educated for two or more years.
HEN Alumni Data
While the HEN Alumni Survey has the usual limitations of a self-selected study conducted by a home education group, it offers a significant body of data on outcomes.
HEN would support a large-scale independent study. In the meantime, the results indicate:
- Home educated students complete degrees at a higher rate than the schooled population.
- Home educated students enter a wide range of further study and occupations.
Following our alumni survey and a related article in The Conversation, questions were raised around the socio-economic status (SES) of home educators. Some suggested that perhaps home educated kids are from a higher socio-economic group and would have good results if they were in school.
Let’s put that theory to the test…
Parental education is often taken as an SES measure, and a predicator of student outcomes.
Victorian Department Data:
The Victorian education department study categorised student reading results by parent’s education. There was a perpetual correlation between parental education and student test results for the always-schooled group, but a weaker correlation in the partially home educated group.
When controlled for parent education, comparison of reading results indicated home educated students were:
- behind in 9/12 groups at Grade 3 level;
- ahead in 7/15 groups at Grade 5 level;
- ahead in 8/14 groups at Year 7 level; and
- ahead in 11/14 groups at Year 9 level.
HEN National Alumni Data:
A more highly-educated parent was observed to have a positive effect on student outcomes in the 20-24 and 25-29-year-old age groups, but not as much as in the schooled population. In the 30-34-year-old age group, outcomes were unconnected with parent education.
A further finding was that the advantage of well-educated parents appeared to diminish over longer periods of home education.
Both these studies suggest that the advantage of well-educated parents may be enhanced by home education, whereas having parents with low qualifications appears to be less of a disadvantage in home education than it is for school children.
NSW: Where possible, the BOSTES study geo-coded students into four SES quartiles using 2011 Australian Bureau of Statistics (ABS) percentiles.
It found in every year level:
The socio-economic status of students was spread fairly evenly across the four quartiles, but that in each case, the smallest proportion was in the high-SES group (top 25%). This was the case even in the HSC English group, where the total student population is skewed towards higher SES (32% in top quartile).
A comparison of the SES quartiles for both the Year 10 and Year 12 groups revealed more home educated students were from lower SES postcodes than their fully-schooled peers.
Queensland: The Queensland Home Education Unit publish a full set of postcode data for students registered for home education from 2013-2018. This is the largest available dataset of Australian home educators being for 3232 students. Using the same method as the BOSTES study, 3230 students could be matched to an SES postcode percentile. The result is reasonably consistent with the NSW study.
Victoria: This data-set differs from the Queensland and NSW ones – being families instead of students.
The HEN database was used to create a list of 1587 unique home educating families from Victoria over 2015-18.
Again, a state SES percentile was allocated to each postcode. As with the other states, there was a spread across the four quartiles but Victoria had the largest group in the top 25%. An examination by student would skew the results to a lower SES if (as might be assumed) lower-socio-economic families have more children.
Further Data-sets: It would be interesting to repeat this process state by state with full registration datasets, but at present those are not in the public domain.
Multiple lines of evidence indicate home education has good outcomes.
Home educators are spread across the SES spectrum with SES and parent education having less effect on home education outcome than is evident in school.
We didn’t believe home education success was a matter of privilege — now we have proof.
- NSW SES analysis: BOSTES used ABS Statistical Area Level 1 (SA1) Index of Relative Socio-economic Advantage and Disadvantage, 2011.
- Victorian and Queensland SES analysis was performed using the: Australia Bureau of Statistics (2016). Census of Population and Housing: Socio-Economic Indexes for Areas (SEIFA), Australia, Table 1 Postal Area SEIFA Summary 2016. Retrieved April 25, 2019. http://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/6227.0May%202016?OpenDocument
- Queensland postcodes: Queensland Home Education Unit, Table 2 https://qed.qld.gov.au/publications/reports/statistics/schooling/students