I'm using Census Pulse data (see here) to visualize impacts of COVID-19 on certain variables. The one I'm currently looking at is unemployment insurance.
I began by reading data for all weeks into R, filtering, and pivoting it, so that I am only looking at the percentage of respondents who applied for unemployment insurance, their racial and gender characteristics, and the number of respondents for each category. I am trying to determine how to calculate the margin of error so I can show it as a band around the line graphs I am plotting. I currently have this visualization:
I am trying to determine if I have the correct methodology for calculating the 95% MOE. I have used the following formula to create a MOE 95%
calculated field:
1.96 * ((STDEV([Value])) / SQRT(SUM([N])))
Where Value
is the pivoted field I generated which holds all percentages, and N
is the number of participants. This is a standard MOE calculation, but when comparing it against other data providers (e.g., Urban Institute), my MOEs are showing up different and I cannot access their methodology.
Any assistance would be appreciated —?I'm a bit in over my head!
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…