Commenter Fitzroyalty asks whether there is recent data on low SES completions and drop-outs. In general, recent research gives cause for optimism that once low SES students reach university their SES status is not of itself (on average) a negative factor affecting outcomes.
This report based on Longitudinal Study of Australian Youth did find some slightly lower projected completion rates for the children of low education or occupation parents, but that these differences were not statistically significant after controlling for ENTER. In other words, low SES had done all the damage it was going to do at school, and did not do more damage at university.
The 2008 report of the Australian Survey of Student Engagement, released a week or two ago, found that low SES students had very similar rates of considering dropping out (34.6%) to all Australian students (33.1%). The grade point average of low SES students (71.6%) was virtually the same as all Australian students (71.9%).
Recent University of Melbourne analysis showed that students with a range of characteristics suggesting potential disadvantage performed fractionally better than other students in their studies if they met the normal entry criteria, but somewhat worse – with lower grade point averages and higher subject failure rates – if they were admitted on lower school results. This is again consistent with the finding that SES affects school results, but does not do further harm at uni level if good school results are achieved.
The Graduate Pathways Survey also suggests that disadvantage does not persist. It says:
While these students may have arrived at university from disadvantaged backgrounds, there was no difference in their self-reported educational performance. Their learning and development outcomes as well as their average overall grades were on par with those of other students.
In longer term results, they were were very slightly less likely to be working or to be working full-time and ‘tended to report slightly lower levels of participation in occupations classified as professional or managerial after five years (59 compared with 64 per cent).’ Unfortunately, there was no five-years out salary data reported. Overall, however, it looks like higher education is paying off for a majority of this group.
Andrew Norton 
What measure of SES was used?
LikeLike
Shem – LSAY: Separate measures on parental education and occupation.
AUSSE: The postcode measure.
U of M: as stated in link
Graduate Pathways Survey: postcode in primary school, plus neither parent university educated, plus neither parent in a professional occupation.
LikeLike
The other thing those results show is that it is not true that low SES students do better even if they have lower ENTER scores. Obviously, if you equate low SES students on ENTER scores and they do about the same, that can’t be true unless they are still at a disadvantage at uni for reasons not to do with teaching.
LikeLike
Though I thought it was suprising that the GPAs were so similar, given that low SES students would typically arrive with lower ENTER. On the other hand, the GPA for this sample is slightly higher than the GPA for U of M students (which was not in the AuSSE sample for 2008), so maybe there are institutional effects as well.
LikeLike
Thanks Andrew – this is cause for great optimism! My hypothesis has been that the determination and resilience of high achieving low SES students gets them to university despite poor school experiences, and the ability of low SES students to succeed at university may be partially the result of the autonomy and self-motivation that got them there. In comparison, the extensive coaching that many high SES students receive at school has, in my experience, left some of them less prepared to make their own way in a busy indifferent world.
LikeLike
I don’t think the data supports that hypothesis, Fitzroyalty. All it shows is that ENTER scores are a good predictor of university drop-out and completion rates. In other words, there is no case for favouring low-SES entrants over high-SES entrants for reasons of greater “resilience” or otherwise.
LikeLike