An implication in complaints about rising university student:staff ratios is that things are getting worse for students.
We don’t have any measures of student learning, the most important indicator, but we do have from the course experience questionnaire sent to all completing students a series of questions on satisfaction with teaching, which together form the ‘good teaching scale’ (GTS). These questions cover time spent commenting on work, helpfulness of feedback, whether students were motivated to do their best work, how good lecturers were at explaining things, whether lecturers worked hard to make their subjects interesting, and whether staff made an effort to understand difficulties students might be having.
Contrary to what we would expect if SSRs were a major teaching problem, all the GTS scores have improved steadily since 1997, though from a low base. In 1997 on average 39% of completing students gave a clearly satisfied rating to the teaching questions (ie the top two points on a five point scale). By 2009 this was up to 52%.
Curiously, the indicator most contingent on the extra staff time supposedly limited by higher SSRs, whether the ‘staff put a lot of time into commenting on my work’, has shown the most improvement 1997-2009. The two indicators have a very large positive correlation of 0.9 over the 1997-2008 period.
The old social science warning ‘correlation is not causation’ very much applies here. But the fact that trends that were supposed to be negatives are not showing as such is surely significant.
I’ve not seen any detailed analysis that attempts to explain these trends. But I very much doubt that it is coincidence that this is same time period in which universities became reliant on markets for their income. With students free to take their dollars somewhere else, teaching quality became a much bigger issue for universities
Andrew Norton 
Could it potentially be a lowering of expectations? Students expecting good service will probably rate OK service worse than students expecting poor service would.
Though I suspect that the actual reason is at least partly technology changes. Email exchanges or discussion boards allow the students to receive individual help without requiring the same time or work commitment from the lecturers as phone calls, or having to be available in an office for face-to-face chats. This allows a lecturer to help more students, and, if done well, without any decrease in quality.
LikeLike
Sounds like a good question for Glenn Beck…what a tiger that bloke is!
But seriously, the issue is not teacher staff ratios, but that teachers can speak English! My gawd! I used to walk into the tutorials and they would always be taken by some Alien ring-ins. They would be studying their Phds or whatever, and the lecturer would give them tutorials as a way of experience and a little extra money for the poor munchkins.
But they couldn’t speak English properly and it would drive ya bonkers! I would take a half-brained fluent speaker over a braniac gobblely-gook speaker any day of the week.
So in short, the univeristy was putting the needs of the tutors ahead of the students. Outrageous!
LikeLike
Shock! It turns out that Baz is one of the university-educated elite!
Not so Ordinary after all, eh Baz?
LikeLike
Tim – I suspect unis have long suffered in the contrast with schools. But I can’t think of any obvious reason why expectations would have lowered in a pretty consistent way 1997-2009. The schools have been lavishly funded compared to unis in that period, which all other things being equal would raise rather than lower expectations. There have been no major changes in the socioconomic profile of the university population.
I actually think that the most plausible interpretation is the right one: there has been slow but steady improvement over that time period.
Technology is almost certainly a factor, but so too is more teaching training and minimum levels of teaching performance required for promotion. I remember some years ago a senior academic at U of M saying that pulling bad teachers out of classrooms quickly was important change at a faculty level to improve averages (he was talking about U of M internal surveys, but fewer really bad experiences would flow through to improved results in the end of course survey referred to in the chart above).
LikeLike
“and minimum levels of teaching performance required for promotion”
.
Good luck trying to measure it and not changing behavior for the worse (or better for the cheating, who you don’t want to catch, lest they give you a bad report for it, and the stupid and lazy, who you don’t want to fail for the same reason).
LikeLike
Conrad – Fail rates (for domestic commencers; later years not reported) were very stable at around 18% until 2006, after which they started dropping and were 15% in 2009. So I don’t think soft marking at the pass-fail point can explain the trend to 2006 at least.
Your uni remains a bastion of academic standards/bad teaching (choose according to prejudice), with a fail rate of 19%.
LikeLike
Could it potentially be a lowering of expectations? Students expecting good service will probably rate OK service worse than students expecting poor service would. Though I suspect that the actual reason is at least partly technology changes. Email exchanges or discussion boards allow the students to receive individual help without requiring the same time or work commitment from the lecturers as phone calls, or having to be available in an office for face-to-face chats. This allows a lecturer to help more students, and, if done well, without any decrease in quality.
LikeLike