How Well Do Item Parcels Represent
Conceptually Defined Latent Constructs?
A Two-Facet Approach
Knut A. Hagtvet
Department of Psychology
University of Oslo, Norway
Fadia M. Nasser
School of Education
Tel Aviv University, Israel
This article presents a methodology for examining the content and nature of item par-
cels as indicators of a conceptually defined latent construct. An essential component
of this methodology is the 2-facet measurement model, which includes items and
parcels as facets of construct indicators. The 2-facet model tests assumptions re-
quired for accepting parcels as aggregates of item covariation in representing the la-
tent construct. According to this methodology, parcels are acceptable indicators of
the latent construct if the 2-facet model meets parametric assumptions for unidimen-
sionality and if items and parcels have content validity as measures of the latent con-
struct. The proposed methodology is illustrated using a 1-factor model of the Worry
construct in the test anxiety measurement tradition.
Item parcels, according to Kim (2000), can be traced back to Spearman (1904,
1914), Thurstone (1931), and, later on, to Cattell (1956). However, the use of item
parcels has not been unanimously accepted. Eysenck and Eysenck (1969) sug-
gested that individual items represent the building stones of a questionnaire, and
therefore intercorrelations among items should be the input to factor analysis. Par-
cels, however, have attracted researchers for several reasons: They may help to (a)
STRUCTURAL EQUATION MODELING, 11(2), 168–193
Copyright © 2004, Lawrence Erlbaum Associates, Inc.
Requests for reprints should be sent to Knut A. Hagtvet, Department of Psychology, University of
Oslo, P.O. Box 1094, Blindern, N-0317 Oslo, Norway. E-mail: k.a.hagtvet@psykologi.uio.no; or Fadia
M. Nasser, School of Education, Tel-Aviv University, P.O. Box 39040, Ramat Aviv 69978, Israel.
E-mail: fadia@post.tau.ac.il