Tuesday, April 17, 2007
An argument for action research in the design of student fees policy
Peter Knight writes on how he got his predictions wrong on the impact of student fees on prospective students shopping around for better deals in education; although he does not say it explicitly, the beginning of his article is surely an invitation to consider action research (read 'stakeholder input', for those who like a different type of jargon) into the design of policy instruments.
Subscribe to:
Post Comments (Atom)
4 comments:
There was an idea expressed in the Guardian article that students may choose more utilitarian courses because of the new loan system in the UK.
I do not think this has be the case, because there is a limit on what students re-pay, based on their post-graduate income. However, students may suffer from a pschyological bias in this instance and possibly swing towards a course that they think will lead to stable employment after college.
If this is the case, I would be interested to know how much this affects the "match between" what students want to do in their careers and what they actually end up doing after college.
A lot of my thoughts to date have been focused on whether there is "mismatch" between under-graduate degree and the nature of subsequent employment, or whether there is "mismatch" between PhD degree and the nature of subsequent employment.
However, I am coming more to think that the most enlightening answers to the question of "(mis)match" would come from a comparison of students' labour market outcomes with their vocational preferences from several points in time.
The first point at which we would ideally measure vocational preferences is when students start to make choices that will impact on their future career - this happens when Leaving Cert subjects are chosen. However, some of the choice at this juncture may be arbitrary, in the sense that it is not based on any distinguishable preferences.
We could deal with this issue by asking students if they have any preferences at each juncture, and then modelling comparisons separately for students that have vocational preferences and for those that don't.
The second point at which we would ideally measure vocational preferences is entry to higher education (or further education, or apprenticeship, or type of post Leaving Cert employment). This is much more complicated as if students have a distinguishable vocational preference(s), then they set up a menus for their preferences and take the best offer from their menu of preferences. This doesn't just apply to the CAO application process; students may want to work in a certain type of job after the Leaving Cert, but not end up in their preferred employment situation.
One can already see how utterly complex the analysis of "(mis)match" is becoming. There is an unbelievable amount of inter-temporal individual heterogeneity e.g. whether students have vocational preferences, at what point do they form them, how close is their vocational outcome to the highest preference on their menu at each juncture.
This goes on and on, with other suggested points for measuring vocational preferences being post-graduate employment or "post-PhD" employment. The other complexity in this model that Liam has made me aware of is that once vocational preferences are formed, they may change over time. This is another factor that one would want to control for in this model. In fact, it has just occurred to me that preferences that were formed at an early stage may in fact be later dropped with no preferences being in place thereafter.
There is so much to piece together here that is simply crying out for a formal theoretical model, and that is definitely what I will get stuck into come this autumn. I'm really keen on tracing out a newly branded line of enquiry, call it "The Extent of (Mis)Match Between Vocational Outcomes and Inter-Temporal Vocational Preferences", and to trace this enquiry from when students start to make choices that will impact on their future career. I hypothesise that this happens when Leaving Cert subjects are chosen.
Getting back to the Guardian article, and the idea that the UK's higher education loan system may affect college course choice (as distinct from college course preference), an economist might find it hard to criticise any government initiative that would lead to individual career choices that would benefit the economy (in the greater social good).
However, government initiatives that throw people off the track of matching their vocational preferences with their vocational outcomes - deny people the chance to realise their full individual freedom. But even if one isn't a libertarian, mass mismatch between vocational preferences and vocational outcomes surely can't be good for aggregate job satiscation, and associated well-being.
It seems to be coming together really well. I get the feeling that this work has the potential to reveal insightful and thoroughly useful information about key issues that permeates and mediates social, economic and psychological realities.
The Guardian article you quote does seem to have an intuitive logic but attempting to understand vocational choice or indeed the development of those preferences is a major challenge. This is relevant to the extent that you wish to model the change in these preferences inter-temporally because it begs the question, as you already mentioned, when do preferences begin to form? You mention the choice of leaving cert subjects, and although this is significant I think that it is more a point of departure. I know I was fairly adamant about becoming a shopkeeper when I was five and a pilot when I seven … (and an economist when I was nine).
An issue that is coming to mind is the choice of Secondary school. This is probably less significant in Ireland than in other countries due to our anonymous points-only criterion for entry to third-level. Take the situation in the UK where the choice of entry to a Grammar school is an imperative requirement for some lines of academic achievement. Other interesting dynamics are obvious in the States - where choice of school (from pre-school through prep-school and college) can be vital to certain avenues of progression. This potentially brings a parental element to your model.
I'll caht to you later, good work!
In relation to what Peter and Martin have said, there should be a formal model of information in the shape of the career choices taken up (and thus visible) to by siblings and parents. I have afeeling there's quite a bit of info on this in medical education journals,m
The impact of sibling, parental (and other peer/network) influences on preference formation and evolution is very interesting.
I mentioned Oxoby (2007), in an earlier post on this blog. His work on “Skill Uncertainty and Social Inference” (IZA WP-2567), suggests that individuals without familiar reference points for academic achievement may under-invest in their human capital due to uncertain expectations about their academic ability.
The mechanism that this works through could be one where the formation of preferences is in line with pre-existing social backround. So there is a "(social background) reference dependency" that impacts greatly on the formation of preferences.
Post a Comment