@Thomas - I think that you have already picked some of the holes in using simple percentage points as an indicator of success or otherwise by these agencies. These figures are subject to all the influences that you identify and a great deal more - what about those who choose to carry on their studies -do they count as employed within the field? Or those who work voluntarily? Or those like me a very long time ago who graduated with a Science degree and went to work for a variety of unrelated reasons, one of which was needing the money!, for a Government Department as a Civil Servant in a completely unrelated area?

And the idea that this in some way reflects the quality of learning is not a straightforward one either. As you say political advantage plays a part and even institutional status has a role. Of course there has to be some way of measuring quality or successful learning but these can be very subjective. What about looking at the students perception of their studies and their colleges by virtue of a student satisfaction rating. This can be a little more accurate as it would include those students who choose not to continue in their field and those who, for whatever reason, are not employed at all. If it is made anonymous it may also overcome some of the privacy issues.

I think is a good example of how numbers can be used. I agree with Mark, graduates can find a job in different areas, for example in Mexico educational psychology find job as teachers or clinical psychologist, because socially they don't have a place into education. Some are lucky, or some begin to open doors but it's not because the college gives them skills, it is because they work hard to see their own future.

That was just one example, so in my mind, numbers don't say too much, the interpretations can say something but all of us know that it doesn't sound the same: 35% of students graduate in 3 years than 65% of students graduate in 6 years... you can have numbers and the way you use them.

In a real better research, you would have to go deeper underneath the numbers, but I wonder if someone would do it. At the end, those evaluations are only to pretend someone is doing something.

It's a sensible measure to use for certain programs, mainly narrowly focused purely vocational programs. It's a ridiculous measure to use for others, including most liberal arts programs. How many philosophy majors get jobs as philosophers? Whereas it does seem reasonable to ask Fagins School of Hair Styling how many of its graduates get jobs as hair stylists.

I see the use of employment in the field of study as a measure of quality of higher education as a reduction of the role of the university in society only as a training center for high performance workers.

For example, in Brazil the evaluation system of higher education has increasingly included the evaluation of the "employability" of graduates of the courses, which indirectly directs universities to invest in technological areas over the humanities; courses like philosophy or pedagogy, has gradually reduced the investments. In private universities this is already a reality, even these institutions use these measures as a marketing strategy to attract students.

The placement in the the labor market is an important factor, especially for students who seek higher education in a form of social mobility, however this entry is subject to many economic and social variables that not only the quality of higher education.

Linking the quality of education for employability of graduates demonstrates a commodification of higher education, being a stronger tendency of neoliberal governments.

I think that studying the performance of graduates in the labor market is both useful and interesting. But it should not be done with a narrow focus on the institutions, and on the coherence between field of study and field of employment. I also think that information about graduate´s employment should not be limited to the short term. The comments posted above provide examples and insights of the problems of the indicator.