Path Analysis
Christy Lleras
Pennsylvania State University, University Park, Pennsylvania, USA
Glossary causal relationships. However, given the complexity of
social life, disentangling the interrelationships among
endogenous variable A variable whose variation is explained variables is often a difficult task. Path analysis is
by one or more variables within the model. a methodological tool that helps researchers using quan-
exogenous variable A variable whose variation is explained titative (correlational) data to disentangle the various
by factors outside the model and which also explains other (causal) processes underlying a particular outcome. The
variables within the model. path analytic method is an extension of multiple
path analysis A statistical method used to examine hypothe- regression analysis and estimates the magnitude and
sized (causal) relationships between two or more variables. strength of effects within a hypothesized causal system.
recursive model A path model where all the causal relation- In addition, path analysis can be used to test the fit be-
ships flow in a single direction with no reciprocal effects or
tween two or more causal models, which are hypothesized
feedback loops.
by the researcher to fit the data.
specification error Error that occurs when significant causal
variables are excluded from the path model. Since path analysis assesses the comparative strength
spurious effects When part of the association between two of different effects on an outcome, the relationships be-
variables is due to shared causal influences. tween variables in the path model are expressed in terms
standardized path coefficient Coefficients that have been of correlations and represent hypotheses proposed by the
converted into standardized z-scores which allow research- researcher. Therefore, the relationships or pathways
ers to compare the relative magnitude of the effects of cannot be statistically tested for directionality and the
different explanatory variables in the path model. models themselves cannot prove causation. However,
path models do reflect theories about causation and
can inform the researcher as to which hypothesized causal
model best fits the pattern of correlations found within the
data set. One of the advantages of using path analysis is
Path analysis is a statistical technique used primarily to
that it forces researchers to explicitly specify how the
examine the comparative strength of direct and indirect
variables relate to one another and thus encourages the
relationships among variables. A series of parameters are
development of clear and logical theories about the
estimated by solving one or more structural equations in
processes influencing a particular outcome. Path analysis
order to test the fit of the correlation matrix between two
is also advantageous in that it allows researchers to break
or more causal models, which are hypothesized by the
apart or decompose the various factors affecting an out-
researcher to fit the data.
come into direct effects and indirect components.
Introduction History of Path Analysis
Path Analytic Methods
Path analysis was originally developed by geneticist
One of the primary goals of social scientific research is Sewall Wright in the 1920s to examine the effects of
to understand social systems through the explication of hypothesized models in phylogenetic studies. Wright’s
Encyclopedia of Social Measurement, Volume 3 Ó2005, Elsevier Inc. All Rights Reserved. 25
