Following is my contribution to a recent discussion on the American Scientist Open Access Forum on the topic of downloads:
What the institutional repository does with respect to download counts, is to direct traffic to the university's site. This will increase the university's web presence and hence emerging institutional web impact assessments, likely to be of increasing importance in years to come.
Having said that, I would also like to point out that there are serious dangers to scholarship that come with over-emphasis on the numbers. What is popular from a scholarly perspective at one point in time is not necessarily what is important.
One way to think of this: imagine that we humans are like a group of lemmings rushing madly towards a cliff (given the climate crisis and our limited attention to this, I would argue that this is a reasonable comparison). Any lemming that says (or writes) about - how to get to the cliff even faster - is likely to be well-heeded (and cited, if it is an academic lemming). On the other hand, the scholar who looks ahead and sees the cliff and shouts off (or writes up for a peer-reviewed lemming paper): "Hey! Cliff ahead! Should be change direction? " may not get much attention immediately. (Later on, after the early birds have gone over the cliff, could be a different story).
Another important point: one real danger of usage statistics is the potential for usage-based pricing. If we go this route, it is just a matter of time before some of us impose limits on reading. If the undergraduate research project becomes a cost item, there will be a strong incentive to limit research and/or eliminate research projects. When one copy of an article can easily serve anyone, anywhere, it would be a shame to go this route. (Thanks to Andrew Odlyzko for pointing to this danger).
For a more in-depth look at this topic, please see my book chapter, "The implications of usage statistics as an economic factor in scholarly communications", OA copy in the SFU IR at:
http://ir.lib.sfu.ca/handle/1892/1639
Comment: if there is one single possibility that tipped the balance for me to become an open access advocate, it is the spectre of usage-based pricing and the evil that this entails, at least for scholarship.
