Tyler Cowen [1] refers us to Jay Joseph [2] who refers us to other critics of "twins reared apart" ("TRA") studies:
Critics, however, have pointed to several key methodological problems with TRA studies (TRA study critics include Farber, 1981; Joseph, 2001, 2004, 2010; Kamin, 1974;
Kamin & Goldberger, 2002; Lewontin, Rose, & Kamin, 1984; Taylor, 1980). These problems include (a) the doubtful “separation” of twins, who frequently grew up together
and had contact over much of their lives, (b) similarity bias in the methods of MZA identification and recruitment, (c) the questionable status of “intelligence” and “personality” as valid and quantifiable constructs, (d) the failure of the MISTRA researchers to publish or share raw data and life history information for the twins under study, and (e) the impact that the researchers’ bias in favor of genetic interpretations may have had on their results and conclusions.
and continues:
While these and other issues are important, the main problem with TRA studies such as Bouchard’s MISTRA is clear: the investigators used the wrong control group (MZTs). By using MZTs as controls, they failed to control for several key environmental factors shared by both MZA and MZT pairs (see Joseph, 2004; Rose, 1982). Environmental influences shared by both MZAs and MZTs include but are not limited to the following:
• They are exactly the same age (birth cohort).
• They are always the same sex.
• They are almost always the same ethnicity.
• Their appearance is strikingly similar (which will elicit more similar treatment from the social environment).
• They usually are raised in the same socioeconomic class.
• They usually are raised in the same culture.
• They shared the same prenatal environment.
• Most studied pairs spent a certain amount of time together in the same family environment, were aware of each other’s existence when studied, and often had regular contact over long periods of time (Farber, 1981; Kamin, 1974).
So maybe twin studies aren't the slam dunk they are often presented as.
Critics, however, have pointed to several key methodological problems with TRA studies (TRA study critics include Farber, 1981; Joseph, 2001, 2004, 2010; Kamin, 1974; Kamin & Goldberger, 2002; Lewontin, Rose, & Kamin, 1984; Taylor, 1980). These problems include (a) the doubtful “separation” of twins, who frequently grew up together and had contact over much of their lives, (b) similarity bias in the methods of MZA identification and recruitment, (c) the questionable status of “intelligence” and “personality” as valid and quantifiable constructs, (d) the failure of the MISTRA researchers to publish or share raw data and life history information for the twins under study, and (e) the impact that the researchers’ bias in favor of genetic interpretations may have had on their results and conclusions.
and continues:
While these and other issues are important, the main problem with TRA studies such as Bouchard’s MISTRA is clear: the investigators used the wrong control group (MZTs). By using MZTs as controls, they failed to control for several key environmental factors shared by both MZA and MZT pairs (see Joseph, 2004; Rose, 1982). Environmental influences shared by both MZAs and MZTs include but are not limited to the following:
• They are exactly the same age (birth cohort).
• They are always the same sex.
• They are almost always the same ethnicity.
• Their appearance is strikingly similar (which will elicit more similar treatment from the social environment).
• They usually are raised in the same socioeconomic class.
• They usually are raised in the same culture.
• They shared the same prenatal environment.
• Most studied pairs spent a certain amount of time together in the same family environment, were aware of each other’s existence when studied, and often had regular contact over long periods of time (Farber, 1981; Kamin, 1974).
So maybe twin studies aren't the slam dunk they are often presented as.
[1] http://marginalrevolution.com/marginalrevolution/2011/02/wha...
[2] http://jayjoseph.net/yahoo_site_admin/assets/docs/Claims_and...
Also referenced by Cowen:
http://journals.cambridge.org/action/displayAbstract?fromPag...
http://www.ssc.wisc.edu/econ/archive/wp2001-08.pdf