Quality & Quantity 32: 109–117, 1998.
© 1998 Kluwer Academic Publishers. Printed in the Netherlands.
Levels of Aggregation: A Conceptual Model
A. BEN OUMLIL
& JOSEPH L. BALLOUN
Department of Management & Marketing, The University of Dayton, Dayton, Ohio 45469, U.S.A.;
School of Business and Entrepreneurship, Nova Southeastern University, 3100 SW 9th Avenue,
Forth Lauderdale, FL 33315-3025, U.S.A.
Abstract. Aggregate analysis has been established as a standard method on the study of market
response behavior for a long time. Aggregation has advanced our understanding of the linkages
among social characteristics and aggregate response behavior. However, aggregate analysis has been
hindered by fragmentary and unsystematic procedures to determine the most appropriate level of
aggregation. The general objective of this paper is to provide a conceptual framework to determine
the level of aggregation of variables in data analysis. In addition, statistical procedures are suggested
in this framework to verify and to determine the level of aggregation represented by a variable. The
conceptual framework is useful for deciding if the variables are to be analyzed from micro-analysis
focus or macro-analysis focus. The statistical procedures enable the researcher to systematically
identify and verify the level(s) of aggregation of variables in an existing data set.
Modeling market response behavior has been a signiﬁcant topic in the marketing
research ﬁeld. In modeling marketing response behavior the emphasis is placed
on either individual or aggregate behavior. For example, multidimensional scal-
ing techniques [MDS] are often used to determine key underlying dimensions of
consumer’s assessments of objects (e.g., product, service, ﬁrms). In dealing with
perceptions of stimuli on a proximity basis of measurement, the researcher can
either scale data on a subject-by-subject basis or strive to create fewer maps through
some aggregation procedure (Hair et al. 1987).
Individual Level Data
In modeling marketing response behavior at the individual level, one application is
to estimate individual utility function for a group of objects (e.g., products, occu-
pations). Various reserchers (e.g., Johnson, 1974; Wittink and Montgomery, 1979)
have demonstrated, throught the application of individual utility estimate functions,
that the individual models have good predictive power. In Johnson’s study (1974),
the median correlation between predicated and reported indiviual preferences was
0.80. Wittink and Montgomery (1979) have succeeded in predicting 63 percent of