Model one was able to accurately predict support for increasing cash assistance
62.1 % of the time. The independent variable of interest (Does welfare lead women to
have more children?) is a strong and statistically significant predictor of opinion about
spending on assistance. People who think that welfare does lead women to have more
children are less likely to support increases in cash assistance. In addition, some of my
demographic variables were significant predictors. Single respondents are more likely
than married people to support increases. Democrats and Independents are more likely
than Republicans to support increases. Respondents with fewer years of formal education
are more likely than respondents with more formal education to support increases in cash
assistance. Gender and race did not demonstrate statistical significance in this model.
Hypothesis Two: Respondents who think Welfare recipients have low moral
standards will be less likely to support increases in Welfare spending. To test this
hypothesis, I replaced the independent variable “Does welfare lead women to have more
children?” in the model above with another question about Welfare recipients. Question
28 asks respondents “In general, do you think people on welfare have higher, lower, or
about the same moral values as other Americans?” All other variables in model two are
the same as in model one.
Convention