In a special issue of Journal of Research in Crime and Delinquency on criminal career research, Daniel Nagin wrote an essay about the contribution of group-based trajectory modeling (GBTM). Appropriately, he also refers to the controversies about the applications of this methodology, where he contends that all earlier critique is just based on a couple of misunderstandings. I am of course honored that my own critique is found important enough to be mentioned (although as one of those having misunderstood the point). I suppose that means I have made some sort of impact. It would have been nice, though, if my actual arguments had been met with rational clear arguments instead of just being dismissed. In fact, other advocates of GBTM have actually responded to my work, but without pointing out any mistakes on my part.
It is worth pointing out that Nagin also refers to another important critique of GBTM by Daniel J. Bauer, who it seems, according to Nagin is based on the same misconceptions. (Other critics could have been mentioned as well). To my knowledge, none has pointed out mistakes in the arguments made by Bauer. On the contrary, Nagin and Odgers (p. 118) have previously acknowledged the importance of a simulation study by Bauer and Curran:
Their work serves as a useful a caution against the quixotic quest to identify the true number of groups in either GMM or GBTM analyses. Perhaps most importantly, this work reinforces the need to move away from interpretations of trajectory groups as literally distinct entities.
So, Nagin has previously agreed that the groups have been interpreted as distinct entities, at least by some, and we should move away from such interpretations. Yet, reading his recent essay, one gets the impression that those criticizing such interpretations have just misunderstood the point. This seems like a contradiction to me.
I do not mind the disagreement, but it would have moved the academic debate forward if those accusing others of being misguided could meet the actual arguments or point out errors in the premises etc. I am still waiting for someone to point out the mistakes in my article on GBTM.
P.S. I have never seen any of the advocates for GBTM criticizing any interpretation of GBTM.
P.P.S Actually, Brame et al did point out a mistake of mine, of which I agree. I had written that Moffitt’s taxonomy was “spurred” by GBTM. That was clearly the wrong word, and I can only blame my bad English as a non-native speaker. I should have written that the popularity of the taxonomic approach was “fueled” by the development of GBTM. Not a major point, though.
The post An update on group-based trajectory modeling in criminology appeared on The Grumpy Criminologist 2016-06-30 09:06:25 by Torbjørn.