The difficulties in doing social science

posted in: MFin, Students | 4

In natural (real) science you can do experiments where you can measure the influence of A on B while holding other things constant. No matter how good a theory is, it’s only taken seriously if there is an experiment that provides data consistent with it. You can’t prove a theory is true, proof is only for maths. You can only keep testing it to see if it stands up to whatever data are available. A famous example is the test of Einstein’s theory of general relativity by British physicist¬†Arthur Eddington who showed that light was indeed “bent” by gravity, as predicted by Einstein, during the solar eclipse of 1919. This offered preliminary confirmation of the theory and later, much more stringent tests have continued to confirm it.

In the social world it’s pretty much impossible to do experiments, other than the slightly artificial ones done in laboratories by psychologists. But they’re better than the alternative, which is to try to infer from actual behaviour what the underlying processes are.

Do MBAs whinge more?

Here’s an example. I’m fairly sure that MBA students are, on average, more likely to complain than other students. It doesn’t mean all or even most MBAs complain, just that in a given sample of MBAs you’ll likely find a higher fraction who do compared with other degree subjects. (The main casualties of this are of course the non-complaining MBAs). ¬†I might be biased, as MFin director, even though I thought this when I was teaching MBAs before the MFin existed.

More useful evidence came to me recently in conversation with someone at Judge who is a sort of service provider to all three premium fee programmes (MBA, Executive MBA and MFin) who volunteered her strong impression that, while the MFin and EMBAs cope with setbacks and problems with a minimum of fuss and complaint, MBA students tend to get in a bit of a state (I’m paraphrasing slightly to avoid the use of offensive language). This is (I think) from an unbiased and independent source with a relatively clear view of the three groups.

But this still leaves lots of questions. For example, is it age? The MFin average age is 26, that of the MBA is 30 and that of the EMBA 37. Perhaps there is a sort of single peaked relationship between age and complaining? Maybe MBAs are going through a sort of early life crisis and the MFin students would have been just as whingy if they’d been a bit older? Perhaps the EMBAs are beyond this stage and are preparing for their mid-life crisis instead.

Even if we could agree that there is, to use economists’ language, a higher marginal propensity to whinge (MPW) among MBAs, we could not conclude that it is necessarily because they’re doing the MBA. Causality is always problematic in social science. Does doing an MBA make you complain? Or are complainers more likely to do an MBA? Here are some competing explanations.

1. It may be the nationality mix. Some nationalities may have a higher MPW. The current class has a higher than usual proportion of North Americans and there is some anecdotal evidence that whinging is above normal levels. But that might a complete coincidence and I might be unfairly maligning North Americans. Only much more data would tell us.

2. Perhaps the students come into the programme with a low MPW but something happens to them in the class that raises it. (There is definitely a seasonality to whinging, with higher levels in the winter and mid-way through the year when everyone is tired and worried about jobs. This seasonality is noticeable in the MFin class as well, but to a lesser extent. EMBAs mostly have a job to return to so may exhibit less seasonality). I do suspect that the MBA team’s customer-focused approach may encourage a higher MPW. We had that problem on the MFin in the early days and we have since tried hard to make everything rule-bound and reduce areas of discretion or negotiation. I can’t be sure if that’s the cause, but we seem to have a nice low-whinge equilibrium now.

3. Maybe, accidentally, the Cambridge MBA selects for high MPW? I’ve no idea why we would do that but it might be that it’s correlated with high quality applicants, so if you want a strong class you have to accept that there will be a few whingers in it.

The ideal experiment would be to take a random selection of MBA applicants of the normal Cambridge standard and feed some into the MBA and leave the others to work for another year, as a control group. You’d need to match the groups by age, sex, nationality, GMAT, career history and anything else that might affect their MPW. If you could do this (it’s of course impossible) you’d learn something about the effect of the MBA itself on MPW but it would still leave open the question of whether ALL MBAs as a group have a higher MPW. For that you’d need a random sample of people who were equally likely to apply to MBAs and MFins. And a lot of them, to allow for statistical controls.

So, in summary, social science is difficult, and poorly grounded conjectures such as the one I started with can live on, because there is no reliable evidence either to confirm or refute them. My future data will likely be strictly constrained now as I’ll probably be booted off MBA teaching for writing this. But I’m fairly confident MBAs have a high MPL (marginal propensity to laugh).


4 Responses

  1. Dijana Maric

    While I am hard pressed not to giggle, I have to note that you appear to have neglected observation of the constants involved – lets term them collectively ‘The Judge Constant’, how they may or may not be different for the described groups and whether there have been any changes in TJC that could have explained the increase in MPW this particular year. Just saying for the sake of being thorough.

    • Simon Taylor

      Good point. The ideal experiment would be a very large panel survey of students over many years applying to different universities, different degrees plus a group that didn’t apply to any masters. We could then identify institutional and birth cohort effects (e.g.are those born in the early and mid-1980s more discontented? Some early reaction to their parents’ clothes perhaps?).

      • Dijana Maric

        Social Anthroppology is not really my field, but I wonder if, were we to correctly identify the reasons for the mass discontent, there would be anything practical to be done about it?

        • Simon Taylor

          There is no evidence of mass discontent. Evidence from polls done after the course finishes shows high levels of satisfaction, both absolutely and relative to other MBA programmes. Second, the majority of MBA students don’t whinge. It’s a minority of complainers, that happens (it seems) to be a larger share than for other similar degree programmes at Cambridge. But it’s still a fairly small minority. The MBA programme team are very vigilant in following up on genuine matters of concern and issues of substance and they consult the students regularly. It’s possible of course that this consultation increases the amount of complaining.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.