28
spending were less important in shaping attitudes toward the domestic costs of the war
than were regional stakes in the global economy.
As was the case with the 1954 CCCL data, some of the estimated effects of
agricultural trade orientation were anomalous, but these were not as large as the impact of
the trade orientation of the larger manufacturing sector. Agricultural export orientation
was associated with less willingness to bear the domestic costs of the military.
Movement from the smallest to the largest regional observation on this variable would
move a respondent downward across 21 percent of the range of the first dimension. On
the other hand, when individual-level variables are added to the model, the effects of
agricultural trade orientation are in line with theoretical expectations. Overall, like the
other datasets examined here, the Vietnam study suggests that regional agricultural
interests were simply not as important as those of manufacturing in shaping attitudes
toward military spending.
The results concerning individual self-interest were also similar to those produced
by the 1954 CCCL study. Education, family income, and professional occupational
status were not statistically significant predictors of the first dimension. However,
individuals with greater education were more willing to accept the costs of war on the
second dimension. Other things equal, individuals with the highest category of education
scored 9 percent higher on the second dimension than individuals in the lowest
educational category. Family income had the predicted sign on the second dimension,
but narrowly missed the threshold for statistical significance. Although these results are
weaker than those concerning regional economic interest, holders of human capital
appeared to be more willing to pay the domestic costs of the war. As in all the models
Convention