In pursuit of what makes a difference in higher ed, and chasing rainbows....

The first thing to make visible:
"the minimum goal of education, when rightly expressed, should be for all students to make at least one year’s progress for one year’s input, no matter where they start" (Hattie, 2015)

The second concern Hattie argues is that unless an impact passes the 0.4 level, grounds for support are seriously dubious.
He makes this highly visible in the visual guide on what does or doesnt have impact  and its worth then considering what translates across to a higher ed sector.

The teacher having credibility, clarity and opportunity of discussion all score highly.
Providing timely and specific feedback scores well.
The size of a class, does not. Class size has a 0.21 impact.
Pre term birth weight is a more reliable indicator!
Often argued is that it is what the teacher does as makes the difference. Notably, this is not John Hattie's argument. He points out there is so much more going on; its the lecturers and the structures or wrap around services, the 'culture' of the learning environment, where learners (and teaching staff) are nourished and supported by the system as a whole.
In a PhD study by Molly Neville, it came down to feeling like I (the student) matters.

How come then, we get so distracted by quality measures that bear little resemblance to what makes a difference?

Informed by Hattie's meta-analysis, this is where we might be better focussed:
Instead of predetermined student outcomes, might we focus on what are the student's needs?
From this, what then are the lecturer needs in relation to the student needs.
Identify the required actions or behaviours for implementing.
Evaluate the impact, and
then do it all again, continuously...

Starting from know thy impact- what is it we do, that we would want to keep doing, and is there an evidence base for this in terms of desired impact on student learning?
Is it good enough, and is there evidence that this is 'good enough'?

THEN, we might consider what progress in this area might look like.