Abstract
People generally prefer their initials to the other letters of the alphabet, a phenomenon known as the name-letter effect. This effect, researchers have argued, makes people move to certain cities, buy particular brands of consumer products, and choose particular professions (e.g., Angela moves to Los Angeles, Phil buys a Philips TV, and Dennis becomes a dentist). In order to establish such associations between people’s initials and their behavior, researchers typically carry out statistical analyses of large databases. Current methods of analysis ignore the hierarchical structure of the data, do not naturally handle order-restrictions, and are fundamentally incapable of confirming the null hypothesis. Here we outline a Bayesian hierarchical analysis that avoids these limitations and allows coherent inference both on the level of the individual and on the level of the group. To illustrate our method, we re-analyze two data sets that address the question of whether people are disproportionately likely to live in cities that resemble their name.
Highlights
Consistent with the above explanation, Nuttin (1985) first found that people tend to prefer the letters in their names to the other letters of the alphabet, a phenomenon known as the name-letter effect ( nameletter effect (NLE); Nuttin, 1987; Hoorens and Todorova, 1988; Hoorens et al, 1990; Greenwald and Banaji, 1995; Kitayama and Karasawa, 1997; Jones et al, 2002; but see Hodson and Olson, 2005)
For each of the 30 surnames, there were two observations: the proportion of people with that surname deceased in the respective surname city and the proportion of people with that surname deceased in the U.S Pelham et al (2003, p. 803) reported a significant result, t (29) = 2.58, p = 0.015 and concluded that “(. . .) implicit egotism is a highly robust phenomenon.”
We considered the “knowledge-based prior,” p(δ) ∼ N (0, 0.303), a prior proposed by Bem et al (2011) for the effect of extrasensory perception; this prior is a plausible lower bound for the effect sizes expected under the NLE hypothesis
Summary
We show the proportion of people with that name deceased in the U.S (second column), the proportion of people with that name deceased in the respective surname city (third column), and the total number of people deceased in the respective surname city (regardless of their name, fourth column). Gallucci (2003) assumed complete independence and calculated a χ 2 statistic and an associated p-value for each name separately. Overall test, ignoring the fact that the cities may differ from each other. This test ignores the fact that the cities may be similar to each other. In the remainder of this article we propose an alternative, Bayesian method for the analysis of the NLE in large databases. Our Bayesian method accounts for the hierarchical structure of the data and incorporates both the differences and the similarities between cities. Recent introductions for psychologists are given for instance by Hoijtink et al (2008), Kruschke (2010a,b), Lee and Wagenmakers (to appear), and Wagenmakers et al (2010)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
