This study posits three dimensions of self-report measures of readership based on the extent to which a measure reflects a general predisposition to expose oneself to a magazine or is tied to a specific issue. Confirmatory factor analysis of specific indicants related to the three general dimensions was used to test the resulting measurement model with a readership study conducted in Venezuela (N = 500). The results demonstrate that the specified dimensions provided a good fit to the data, with higher loadings and less measurement error associated with the dimension reflecting the most abstract measures. As predicted, the intercorrelations between the dimensions indicated they reflect differing underlying dimensions of readership.
[1]
Daniel Starch,et al.
Measuring advertising readership and results
,
1967
.
[2]
G. Maruyama,et al.
Evaluating causal models: An application of maximum-likelihood analysis of structural equations
,
1980
.
[3]
Richard L. Allen,et al.
The Reliability and Stability of Television Exposure
,
1981
.
[4]
Karl G. Jőreskog,et al.
LISREL: A General Computer Program for Estimating a Linear Structural Equation System Involving Multiple Indicators of Unmeasured Variables.
,
1972
.
[5]
G. Salomon,et al.
On the Meaning and Validity of Television Viewing.
,
1978
.