Do birds of a feather flock, fly, and continue to fly together? The differential and cumulative effects of attraction, selection, and attrition on personality‐based within‐organization homogeneity and between‐organization heterogeneity progression over time

• Date: 01 December 2018
• Author: Joo Hun Han,In‐Sue Oh,Brian Holtz,Seongsu Kim,You Jin Kim
• DOI: http://doi.org/10.1002/job.2304
• Published date: 01 December 2018

RESEARCH ARTICLE

Do birds of a feather flock, fly, and continue to fly together?

The differential and cumulative effects of attraction, selection,

and attrition on personality‐based within‐organization

homogeneity and between‐organization heterogeneity

progression over time

In‐Sue Oh

1

|Joo Hun Han

2

|Brian Holtz

1

|You Jin Kim

1

|Seongsu Kim

3

1

Fox School of Business, Temple University,

Philadelphia, Pennsylvania

2

School of Management and Labor Relations,

Rutgers, The State University of New Jersey,

Piscataway, New Jersey

3

Graduate School of Business, Seoul National

University, Seoul, Korea

Correspondence

In‐Sue Oh, Department of Human Resource

Management, Fox School of Business, Temple

University, Philadelphia, PA.

Email: insue.oh@temple.edu

Funding information

Ministry of Education of the Republic of Korea

and the National Research Foundation of

Korea, Grant/Award Number: NRF‐

2015S1A5A2A03047915

Summary

The most fundamental proposition of attraction–selection–attrition (ASA) theory is

that organizations progress toward homogeneity of personality over time through

the processes of attraction, selection, and attrition. However, critical aspects of that

proposition have remained untested, thus limiting a fuller understanding of the ASA

theory. To close the gap, this multiwave, multisample, and multi‐inventory study

examines the extent to which each of the ASA processes and different personality

traits contribute to within‐organization homogeneity progression as well as between‐

organization heterogeneity progression over time. Our findings suggest that both

within‐organization homogeneity and between‐organization heterogeneity emerge

to varying degrees over time and that selection, among the ASA processes, is most

responsible for the within‐organization homogenization, whereas attraction contrib-

utes most to between‐organization heterogeneity. We also found that within‐

organization homogeneity progression overall operates more strongly on extraversion

than on the other personality traits, whereas between‐organization heterogeneity

progression operates more strongly on neuroticism than on the other personality

traits. Overall, this study provides an important extension to the ASA theory and use-

ful insights into the organization‐level emergent process of personality‐based human

capital resources. We discuss theoretical and practical implications of this study along

with study limitations and future research directions.

KEYWORDS

attraction–selection–attrition,heterogeneity, homogeneity, human capital resources, personality

1|INTRODUCTION

Schneider's(1987) attraction–selection–attrition(ASA) model posits that

the formation,change, failure, and successof an organization are at least

partiallydetermined by attributes (i.e.,personalities, goals, and values)of

individuals withinthe organization, and therefore “the peoplemake the

place.”Consistent with the literature on person–organization (P‐O) fit

(e.g., Kristof, 1996; Kristof‐Brown, Zimmerman, & Johnson, 2005), the

ASA theory suggests thatindividuals are more likely to be attracted to,

selected by, and stay longer in organizations that fit their personality.

That is, “the centralproposition of ASA theory is that organizations tend

In‐Sue Oh and Joo Hun Han contributed equally. The authorship order is

arbitrary.

Received: 10 February 2017 Revised: 25 May 2018 Accepted: 31 May 2018

DOI: 10.1002/job.2304

J Organ Behav. 2018;39:1347–1366. © 2018 John Wiley & Sons, Ltd.wileyonlinelibrary.com/journal/job 1347

toward homogeneity of personality”over time through attraction,

selection,and attrition (Schneider,Smith, & Goldstein,2000, p. 74). If this

proposition is accurate, then there should be a gradual shrinkage in

variabilityon the basis of individualdifferences among employeeswithin

an organization overtime. Moreover, if within‐organizationhomogene-

ity

1

does emerge in this manner, then ASA processes may generate

sufficient between‐organization variability to permit the reliable differ-

entiationof organizations (Schneider& Bartram, 2017; Schneider,Smith,

Taylor, & Fleenor, 1998).The ASA processes as a whole, therefore, can

contribute to sustained competitive advantage of the organization by

driving the emergence of human capital resources defined as the

aggregate traits of an organization's workforce that have the potential

to contribute to organizational effectiveness (e.g., Oh, Kim, & Van

Iddekinge, 2015;Ployhart, 2006; Schneider & Bartram, 2017).

Subsequent studies that empirically tested Schneider's (1987)ASA

theory (e.g., Jordan, Herriot, & Chalmers, 1991; Oh et al., 2015;

Ployhart, Weekley, & Baughman, 2006; Schneider et al., 1998)

conducted a multivariate analysis of variance (MANOVA) on the basis

of cross‐sectional personalitydata collected from multipleorganizations.

These studies generally found that between‐organizationheterogeneity

was greater thanwithin‐organization homogeneity, thus suggestingthat

“members of the same organization (individuals within organizations)

should be more similar in shared personality than members of different

organizations(individuals between organizations)”(Ployhart et al., 2006,

p. 662). That is, prior research has shown that, as a result of the com-

bined effect of ASA, organizations can be reliably distinguished from

each other on the basis of organization‐level personality emergence.

Despite a great deal of research drawing on the ASA theory, impor-

tant theoretical and empirical questions have yet to be answered. First,

prior research focusing on between‐organization heterogeneity using

MANOVA has not explicitly examined the degree of homogeneity that

exists within organizations. Schmidt, Shaffer, and Oh (2008) pointed

out that as far as there was a fairly large amount of between‐organiza-

tion variability, the results of a MANOVA would seem to support the

ASA homogeneity progression hypothesis regardless of the actual

degree of within‐organization homogeneity. Thus, MANOVA‐based

studies cannot provide any direct evidence regarding within‐organiza-

tion homogeneity. Relatedly, although it is the fundamental proposition

of the ASA theory that organizations become more homogeneous in

terms of individual traits (e.g., personality) over time (Schneider et al.,

2000), this critical issue has not been properly tested and remains an

open empirical question. Specifically, despite the fact that “time is

central to the ASA model's description of how organizations evolve”

(Jackson & Chung, 2008, p. 52; see also Schaubroeck, Ganster, & Jones,

1998, and Mitchell & James, 2001), existing empirical tests of the ASA

theory have solely relied on cross‐sectional data that do not allow the

researchers to test the within‐organization homogeneity progression

over time (Bradley‐Geist & Landis, 2012). Further, as noted by Ones

and Viswesvaran (2003, p. 571; Vinson, Ones, & Connelly, 2008),

“empirical tests of ASA theory have tended to (exclusively) focus on

across‐organizations variability”(e.g., see Jordan et al., 1991; Ployhart

et al., 2006; Schneider et al., 1998), without testing the trajectory of

within‐organization homogeneity progression over time. This is prob-

lematic because between‐organization variability does not necessarily

establish within‐organization homogeneity (Bliese, 2000; George &

James, 1993). Thus, the direct examination of within‐organization

homogeneity emergence using longitudinal data can provide an

important empirical test of the ASA theory.

Second, existing theory and research based on the ASA theory

(Schneider, 1987; Schneider et al., 1998) has predominantly focused

on the combined effects of the three ASA processes together as a

whole. However, as a hypothetical example, it is possible that the

selection process may yield higher within‐organization homogeneity

than do the other two ASA processes, whereas the attraction process

may yield higher between‐organization heterogeneity than do the

other two ASA processes. That is, despite its clear importance in the

ASA theory, a more fine‐grained understanding of the unique contri-

bution of each separate ASA process to within‐organization homoge-

neity and between‐organization heterogeneity progression is lacking

in the current literature. Thus, we believe that examining the unique

effects of each ASA process may shed light on more potent driver(s)

for the emergence of human capital resources, help to refine the

integration of the ASA theory with Barney's (1991) resource‐based

theory (RBT), and ultimately lead to a better understanding of how

organizations create human capital resources for sustainable competi-

tive advantage. In other words, an empirical test of the unique contri-

bution of each ASA process to within‐organization homogeneity and

between‐organization heterogeneity progression can further advance

both the ASA theory and RBT.

Lastly, although the ASA theory concerns within‐organization

homogeneity and between‐organization heterogeneity in terms of var-

ious kinds of traits, it does not specify whether ASA processes operate

similarly across different traits. One reason for the lack of such research

is that organizations may differ in the aggregate traits from the begin-

ning (given that different founders/managers may find different traits

more attractive), and thus there are no specific traits that consistently

drive within‐organization homogeneity (particularly through attraction)

across organizations and likewise drive between‐organization hetero-

geneity in different organizations. In addition, organizations and

employees may have different levels of preference (vs. tolerance) for

fit (vs. misfit) depending on the particular trait in question. As a result,

ASA processes may lead to different levels of within‐organization

homogeneity and between‐organization heterogeneity across different

traits (Schneider & Bartram, 2017; Vinson et al., 2008), though extant

theory and empirical research have not yet provided a clear answer to

this issue. Thus, it is an open question as to whether certain individual

traits may be more or less important in determining within‐organization

homogeneity and between‐organization heterogeneity across the ASA

processes (Kristof‐Brown & Guay, 2010). Thus, investigating the nature

and magnitude of within‐organization homogeneity and between‐orga-

nization heterogeneity progression in terms of different individual

traits across the ASA processes should provide a more sophisticated

understanding of ASA processes, as well as extend RBT‐based

research by testing potential differential roles of individual traits as

microfoundations of human capital resources.

1

Throughout the paper, homogeneity (heterogeneity) refers to relative homoge-

neity (heterogeneity) because we are interested in examining which of the ASA

processes and which of personality traits contribute more or less to personality‐

based within‐organization homogeneity (and between‐organization heterogene-

ity) progression over time.

1348 OH ET AL.