A survey initially planned to interview 1,000 randomly selected adults but only achieved 640 interviews (64% response rate). The researchers estimate the proportion supporting a policy at $\hat{p} = 0.48$ from the 640 respondents. To account for potential non-response bias, they use post-stratification: the 640 respondents are weighted to match the known population age distribution. This adjustment:

Multiple answers may be correct.