5. Results
Table 2 shows Eq. (2)'s estimation in the memory-based task (control)
under the assumption that gender affects the idiosyncratic predisposition
of product dummies only, and females' product perceptions
and choice process do not significantly differ from those of males.
The results presented in Table 2 suggest that gender does not significantly
affect the product dummies (i.e., the cross effects of gender and
hamburger and chicken sandwich dummies are not significant relative
to the cross effect of gender and salad dummy), and therefore the analysis
does not support the assumption that solely cultural, social, and
mating considerations account for gender differences in food choices.
The following section reports testing for significant gender differences
in the perceptions of the three food items.
5.1. Perceptions and perceptual associations
As hypothesized earlier, differences in initial perceptions of the
three products' calorie contents may affect perceptions of the products'
other attributes. Exposure to calorie content and calorie-plus-workouttime
information may change perceptions of products' attributes both
directly (correction effect) and indirectly, via priming on calories,
which implies greater attention to the health and potential weight
gain benefits. Observing gender differences in calorie density perceptions
in the control group supports the argument that males and females
are exposed to differing sources of media information, or that
females who assign greater weight to their health and appearance will
be more sensitive to calorie information, and therefore calorie information
will be more accessible. Observed differences in the patterns of
calorie perception update lend support to the argument that females
are more responsive than are males to calorie information. Table 3
shows the results of these estimations.
In the control group, males and females did not differ in their calorie
perceptions; however, viewing calorie information did expose gender
differences in the perceptions of calories: After viewing calorie information
on the menu, females increased their perceptions of the
hamburger's and the chicken sandwich's calorie densities, while calorie
information did not affect males' perceptions thereof. These changes in
females' perceptions of calories relative to males' constant calorie
valuation generated a significant gender difference in the valuation of
the hamburger's calorie density, and a marginally significant difference
in valuing the chicken sandwich's calorie content. The ordinal ranking
of the three products' caloric values remained static post-exposure to
calorie information.
These results suggest that in the control group, gender differences in
calorie valuations are not the reason for gender differences in choice
processes or entrée choices, if such exist. Calorie information, particularly
calorie information combined with burn time, affects females more