The good folk over at iPerceptions (creators of the 4Q survey tool with Avinash Kaushik, and - to my mind - hands-down winners of the vendor beauty parade at E-metrics San Francisco) have put out a report into the concepts that are getting the most airplay on the "top" web analytics blogs (which includes this one, to my ego's delight). You can read a blog post about the report here, or download the report (in PDF format) here.
The report uses iPerceptions' proprietary text analysis technology to extract and correlate the concepts (essentially, key words collapsed into groups - so "convert", "conversion" and "converted" all collapse to the "convert" concept) that are appearing on blogs. They've picked 30 web analytics blogs to run this on, though the list isn't public (feel free to share in the comments if your blog was included in the test - I think iPerceptions have e-mail everyone whose blog they analyzed).
The main conclusion of the report is that "social" measurement is one of the key concepts (paired with "metrics") that is found on the blogs, but that conversion measurement is still a popular topic, whilst "engagement" seems to have receded a bit since the furore around this a year or so ago.
It's a nice piece of work, but I have to say it does seem a little... well, thin. Only a few of the concepts appeared enough times to really sustain any kind of statistical analysis (for example, the "Engagement" concept only appeared about 20 times, it would seem), and some of the correlations are working with samples as low as 8. So it's hard to read very much into the findings below the highest level.
iPerceptions say they're going to continue to measure on a bimonthly basis (this report is for May-June) to show trends, but with the core numbers this low, they will be hard-pressed to filter out the noise.
My recommendation for improving this study would be to widen the net. There is much discussion of web analytics going on these days across blogs and news sites (for example, both ClickZ and MediaPost have web-analytics focused articles & newsletters). A better approach might be to devise some kind of way of reliably identifying an article as relating to web analytics, and take the analysis from there. And this would have the advantage of showing what the "mainstream" of the web analytics industry is talking about, rather than us more rarified "expert" bloggers (and besides, I haven't really posted about web analytics since April).