TY - JOUR
T1 - Within-Group Agreement (rWG)
T2 - Two Theoretical Parameters and their Estimators
AU - Newman, Daniel A.
AU - Sin, Hock Peng
N1 - The authors would like to thank Ben Schneider and Dave Mayer for the data used in the empirical example. This work was presented in a worldwide webcast honoring the career contributions of Larry James on April 26, 2013. An earlier version of the article received the Sage Best Paper Award from the Academy of Management Research Methods Division, and was presented at the annual meetings of the Academy of Management and published in the Academy of Management Proceedings (Newman & Sin, 2008). The author(s) received no financial support for the research, authorship, and/or publication of this article.
PY - 2020/1/1
Y1 - 2020/1/1
N2 - When measuring group-level psychological properties (e.g., organizational climate, leadership, team motivation), researchers typically aggregate individual perceptions to represent the group. L. R. James provided the groundbreaking insight that, in order to justify aggregating individual perceptions to represent a group-level property, one must first establish that there exist shared perceptions—or shared psychological meaning—within the group. Here we label and describe two distinct theoretical parameters that can both be used to define within-group agreement: (a) ψWG (i.e., a parameter that defines within-group agreement as Individual True-Score Consensus), which arises from the theoretical work of L. R. James and colleagues in the 1970s, and (b) ρWG (i.e., a parameter that treats within-group agreement as a Group True-Score Reliability Analog), which forms the theoretical basis for the rWG(J) index. We extend the work of L. R. James by offering a systematic comparison of different estimators of the two within-group agreement parameters (ψWG and ρWG). Recommendations are provided for estimating within-group agreement, to continue the legacy of justified measurement of group-level psychological properties.
AB - When measuring group-level psychological properties (e.g., organizational climate, leadership, team motivation), researchers typically aggregate individual perceptions to represent the group. L. R. James provided the groundbreaking insight that, in order to justify aggregating individual perceptions to represent a group-level property, one must first establish that there exist shared perceptions—or shared psychological meaning—within the group. Here we label and describe two distinct theoretical parameters that can both be used to define within-group agreement: (a) ψWG (i.e., a parameter that defines within-group agreement as Individual True-Score Consensus), which arises from the theoretical work of L. R. James and colleagues in the 1970s, and (b) ρWG (i.e., a parameter that treats within-group agreement as a Group True-Score Reliability Analog), which forms the theoretical basis for the rWG(J) index. We extend the work of L. R. James by offering a systematic comparison of different estimators of the two within-group agreement parameters (ψWG and ρWG). Recommendations are provided for estimating within-group agreement, to continue the legacy of justified measurement of group-level psychological properties.
KW - aggregation
KW - multilevel modeling
KW - r
KW - within-group agreement
KW - ψ
UR - http://www.scopus.com/inward/record.url?scp=85058961899&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85058961899&partnerID=8YFLogxK
U2 - 10.1177/1094428118809504
DO - 10.1177/1094428118809504
M3 - Article
AN - SCOPUS:85058961899
SN - 1094-4281
VL - 23
SP - 30
EP - 64
JO - Organizational Research Methods
JF - Organizational Research Methods
IS - 1
ER -