ASA theory: an empirical study of the attraction proposition

This paper reports an empirical test of Schneider’s (1987) attraction proposition that organizations attract people who share the organization’s values. The values of 621 applicants to nine utility companies in the United Kingdom were compared to (1) the values of people contiguously seeking similar work, (2) the values of employees they might be working alongside, and (3) the values of the organizations’ senior managers. The results show an effect for person–vocation fit, but once this is controlled for all significant effects disappear. These results suggest that applicants choose which organization to apply to based on their desire for a particular type of work rather than their attraction for particular companies, which is contrary to Schneider’s attraction proposition. In a conclusion at the end of the paper, possible reasons for the rejection of Schneider’s attraction proposition are discussed. It is argued that the factors of familiarity, proximity and exposure are critical to applicants’ behavior and should be incorporated into ASA theory. ASA theory was developed by Schneider (1983a, 1983b, 1985, 1987; Schneider, Goldstein & Smith, 1995; Schneider, Smith & Goldstein, 2000; Schneider, Smith, Taylor & Fleenor, 1998) as an explanation of why organizations look, feel and behave as they do. The main idea in this framework is that organizations attract, select, and retain those people who share their values. Schneider (1987) argues that this cycle creates similarity in the type of people employed by the organization and that this similarity limits the actions of the organization owing to the fact that it occupies a constrained niche of like-minded employees sharing similar values, personalities and attitudes. Schneider et al (1995) call this effect the ‘homogeneity hypothesis’ and they predict that it will be dysfunctional for organizations, as they become increasingly ingrown and resistant to change. Schneider’s framework is based on an intuitive cause and effect relationship that has been well-tested in the psychological literature over the past fifty years. Namely, the observation that similarity between people leads to attraction. For example, Hatfield, Traupmann and Walster (1978) reviewed the literature and showed that couples tend to be similar in terms of IQ, education and other characteristics. Newcomb (1961) gathered data from a women’s university college in the USA that had a strong tradition of liberal values. 1 All correspondence concerning this article should be addressed to Jon Billsberry, Centre for Human Resource and Change Management, Open University Business School, The Open University, Milton Keynes MK8 0DA, United Kingdom; +44 (0) 1908 655888; Fax +44 (0) 1908 655898; electronic mail may be sent to [email protected]. 2 The author would like to thank Kevin Daniels, Peter Herriot, Tim Clark, Jo Silvester, and Chris Coupland for their helpful comments. Billsberry Schneider’s Attraction Proposition 2 The researcher found that in order to gain the liking and acceptance of classmates, many students coming from conservative backgrounds adopted liberal attitudes. In reviewing the many ‘bogus stranger’ studies of attraction, Byrne and Nelson (1965) conclude that as the proportion of similar attitudes increases, attraction towards the stranger increases in a linear fashion. From the vocational psychology and job choice literatures, Schneider cites the work of Holland (1976), who showed that people choose to join career environments that they are similar to, and Tom (1971), who contributed to this literature by demonstrating that people’s most preferred work environment has the same ‘personality’ as they do. Hence, Schneider was able to extrapolate from these studies using the idea that similarity leads to attraction. He argues that people want to work with people whom they are similar to and that this effect can be found when people look for jobs (attraction), are selected for jobs (selection), and during employment (attrition). The overall effect of this cycle is that the people within the organization increasingly become more like one another and that this leads to homogeneity in the workforce. Homogeneity hypothesis Schneider’s Attraction–Selection–Attrition cycle (ASA; Schneider, 1987) is one of the most influential models in the person–organization (P–O) fit literature. Yet, despite the centrality of the cycle in the P–O fit literature, and its proposed dire consequences for organizations, there have been relatively few studies of it or its separate propositions (Schneider et al, 1998). Those studies that have investigated whether the homogeneity hypothesis are reviewed below. Jackson, Brett, Sessa, Cooper, Julin and Peyronnin (1991) studied 939 executives in 93 top management teams in large, established American banks and financial holding companies. They found some evidence to suggest that the executives in these top management teams were clustered together into teams that were relatively homogeneous in their demographic composition (i.e. in relation to their age, tenure, level of education, college curriculum, experience outside the industry, and military experience). The researchers also found (1) that top team turnover is greater with greater levels of heterogeneity in the top team, (2) that new recruits to the top team from inside the organization are more homogeneous with existing top team members than recruits from outside the organization, and (3) the more a member of the top team differs to other members, the greater the likelihood that they will leave the top team. This study provides some support for Schneider’s homogeneity hypothesis. This support is weak, however, because the researchers examined homogeneity via demographic variables (age, educational background, military experience etc.), rather than by investigating psychological variables such as values, attitudes, or personality as advocated by Schneider (1987; Schneider et al, 1995, 1998). Jordan, Herriot and Chalmers (1991) investigated three different hypotheses derived from Schneider’s theory: (1) that members of different organizations differ in personality; (2) members of different occupations within organizations differ in aptitude but not personality; and, (3) that there is an interaction between seniority and organization such that the more senior managers are closer to the organizational personality profile than less senior people. Their sample was 344 managers in four British organizations. The researchers found support for the attraction and selection elements of the ASA theory, but not for the attrition stage as senior managers appeared no different to others within each of the organizations. Interestingly, Jordan et al. (1991) found evidence for differences between functions within organizations suggesting that subcultures within organizations reproduce themselves in their own image. Schneider, Smith, Taylor and Fleenor (1998) investigated the homogeneity hypothesis Billsberry Schneider’s Attraction Proposition 3 using data from the archives of the Center for Creative Leadership. Their sample consisted of 12,739 predominantly middle-aged, white, male managers from 142 organizations who had completed the Myers–Briggs Type Indicator (MBTI; Myers & McCaulley, 1985) whilst on a leadership development course at the Center. This sample spanned a broad spectrum of US businesses. The researchers were able to support the hypothesis that organizations are relatively homogeneous with respect to the personality attributes of their managers by demonstrating that the personality profiles are distinctly different to each other. They also found that both industry and organization had a significant effect on the personality characteristics of managers. Finally, Denton (1999) examined the hypothesis that there would be a significant relationship between the homogeneity of personality and organizational tenure. He examined the personality types of 87 female store managers working for the same book retailer, but in different outlets. The participants completed the MBTI personality indicator whilst attending a training course. At the individual level, the researcher was unable to find any statistically significant relationships between the two sets of data. One explanation of this finding could be that the managers of the retail outlets are rarely in contact and operate independently of each other, co-ordinated by the remote head office. As such, there would not appear to be any major factors coalescing the personalities of the 87 dislocated managers. There are many other reasons why no significant findings were found, including, of course, the falsehood of the original premise, but the data seems to suggest that some form of actual personal interaction is a prerequisite for the homogeneity hypothesis. In summary, these studies provide mixed results regarding the homogeneity hypothesis. The studies that have looked at organizational environments in which the respondents have geographical proximity have found the effect, whereas the one study in which the organizational members were geographically dispersed did not. The finding that geographic remoteness of organizational members is relevant to the homogeneity hypothesis (albeit a very weak finding) does suggest that the attraction phase of the ASA cycle might be less influential than hypothesized in the cycle than the selection and attrition phases of the cycle. This might be the case because much selection (acknowledging the caveat of the remote screening of application blanks and CVs) is conducted face-to-face and during employment most employees interact with other members of staff. Applicants, on the other hand, divide into two distinct camps; internal and external applicants that are defined by whether or not they interact with existing members of staff. The external applicants are particularly interesting to consider because they are the people who might disrupt the homogeneity by bringing in new and different values to the organization.