When we last heard about writer Jonah Lehrer – whose career self-destructed after his writing was found to have numerous instances of plagiarism and factual inaccuracies – he had been paid $20,000 to give a much-criticized speech about journalistic ethics. A few months after that, he was reported to be circulating a book proposal – which also allegedly included plagiarized content. Then….nothing.

And now, very quietly, he’s back.

At the end of March, Lehrer’s website – which had been largely inactive since mid-2013 – was relaunched as a blog, on which, he announced, he would “write about scientific research that interests me”. He also stated that his posts would be approved by the authors whose research he wrote about, or reviewed by an “independent fact-checker”. The relaunch got minimal attention, beyond a supportive Tweet from Malcolm Gladwell – and since then, without much other attention, Lehrer has posted articles about once a week.

The other purpose of the blog is, in Lehrer’s words, “regaining the trust of my readers”. I’ve been following the blog and reading the posts since Lehrer began this new venture. And as one of the critics of the problems in his earlier work, I’m really pleased to see him fully acknowledging his sources, and including links to much of the material he cites.

However, I have some concerns about Lehrer’s presentation and analysis of the research he writes about. The individual articles he discusses are interesting, but the discussions of the research tend to lack context. The reader doesn’t get a sense of how the research findings from each article follow from or add to existing knowledge; when Lehrer mentions other articles on the same topic, they are usually articles by the same authors, which doesn’t fully capture how the range of knowledge about the topic has developed. That makes it difficult to determine whether a particular article is making as big a contribution as Lehrer implicitly suggests.

And while Lehrer obviously knows more about his topics than many journalists, I question whether he has the depth of knowledge in every discipline he writes about to be able to definitely declare a piece of research as “important”, or to conclude that a particular paper “begins to explain” something. His presentation sometimes has whiffs of the same hyperbole that was an issue with his previous work, and that’s a little disquieting.

Lehrer also rarely acknowledges that a single piece of research always has limitations. A research study may have interesting results, but to some degree any research outcomes are unique to the specific circumstances in which they were produced: e.g., the way the hypothesis or research question was worded, the choice of which data to collect and how to collect it, the way the data were analyzed. This is why it’s important to look critically at the study itself, rather than just its results.

For example, in the post entitled “The Too-Much-Talent Effect” (a subject which I have some professional interest in), Lehrer discusses this study, which draws a significant part of its data from the the Voice and Accountability section of the International Monetary Fund’s Worldwide Governance Index. The authors describe this index as “annually employing 31 different qualitative indicators from 13 reputable sources” (p. 14). But the Index has been strongly criticized for not clearly defining the constructs that it measures – including the measures in the Voice and Accountability part of the index – and for confusing measurements of perceptions with measurement of actual indicators (such as measuring perceptions of freedom of the press rather than assessing the amount or type of media regulations affecting press freedoms). This is a limitation that is important to know about in assessing the usefulness of the study’s results. Looking more closely at these kinds of structural and methodological aspects of a research study – rather than just reporting the study’s results –  would really add to the reader’s understanding of what the research is contributing,  and also how research works in general.

In that same post, Lehrer states, “[The study results] suggest[s], for one thing, that much of the national variation in performance – and it doesn’t matter if we’re talking about the soccer pitch or 8th grade math scores – has to do with how well countries utilize their available human capital” – and then “These findings aren’t just relevant for sports teams”. But actually it does “matter”, because what is true for football may not be true for other types of activities. And the authors of the study clearly state this in the article: “Because we examined the impact of egalitarianism on football performance, it remains unclear if and how the present findings extend beyond the football and the sports domain” (p. 34). The authors have made an explicit statement about the limitations of their study’s results, but Lehrer’s commentary ignores that – and, again, slips into the problematic realm of hyperbole.

The blog is certainly a different approach to rebuilding a professional reputation. I don’t know if Lehrer’s blog posts alone are going to be enough to re-establish his credibility, because he has a lot of ground to make up. Nevertheless, the blog seems like a relatively low-risk way for Lehrer to work out the issues in his previous writing. Obviously, I think he could be doing more to avoid the reoccurrence of some of those issues – but I appreciate him being brave enough to risk putting his writing out in public again. It will be interesting to see if the blog can become the place for Lehrer to successfully restart his career.