Final answer:
The null hypothesis for comparing two proportions is that there is no difference between the national poll's support rate and the support rate from Michigan's polling; thus, H0: p1 = p2. The alternative hypothesis is Ha: p1 ≠ p2.
Step-by-step explanation:
The correct null hypothesis for testing whether the proportion of voters who support a taxpayer-funded free college program is different from the provided national poll is H0: p1 = p2, where p1 represents the proportion in the national poll (0.48 in favor since 52% oppose) and p2 represents the proportion from the Glangariff Group data (342 out of 600 or 0.57 in favor).
This null hypothesis assumes that the proportion of opposition to making all U.S public colleges free is the same as the proportion of support for a taxpayer-funded free college program.
The alternative hypothesis would be Ha: p1 ≠ p2, indicating that there is a difference in proportions.
When testing at an alpha level of 0.05, and assuming the null hypothesis is true, we would look for evidence that suggests a significant difference between these two proportions to either reject or not reject the null hypothesis based on the calculated p-value from a statistical test.