Currently computing estimates using the Household Income dataset--B19049 for 2016 5-year estimates.
I am working with the data on a county level. However, when it comes to New York City, we combine the boroughs to obtain estimates for New York City as a whole. When doing this for the Median Household income variable, we took the average of the median values of the boroughs. We were wondering if anyone had an idea on how to find the margin of error for the new estimate i.e. the average of the median household income for the New York City boroughs.
Thank you!
Hi Stacy,
First off, I would advise against using an average of median numbers - that doesn't really make sense. Instead, you can calculate a median using the frequency distribution of household income from table B19001. Below is a link to a PDF that describes the technique, including how to calculate the MOE. Also included is a link to an Excel file that calculates the median. However, it does not have the MOE calculation in it but you should be able to work that out by using the instructions in the PDF file.
http://gisdiva.com/downloads/resources/How_to_Recalculate_a_Median.pdf
http://gisdiva.com/downloads/resources/MedianCalcFromFreqDistribution.xlsx
let me know if you have trouble accessing the files.
-Jami