BIOSTAT212_SU20: Missing versus excluded data

For Question 5e - I thought I did this correctly, then found that there is both Excluded Record data and Missing data, such that for the variable hivstatus there are 4 values rather than 3.

Am I supposed to combine Excluded Records with Missing data into one value, or how else do I account for the Missing data values?

Thanks!

Re: Missing versus excluded data

by Taylor Howell - Sunday, August 23, 2020, 12:48 PM

Hi Sara,

Yes, you could recode the "ExcludedRecord" values to Missing ".". You could then also check to see if you need to relabel.

Re: Missing versus excluded data

by Sara LaHue - Sunday, August 23, 2020, 3:42 PM

Thanks! Looks like I was doing this correctly but was getting tripped up by what I was seeing when I did "..., miss" - thanks so much for verifying!

Re: Missing versus excluded data

by Sujata Mathur - Monday, August 24, 2020, 9:42 AM

i do am stuck in this. can anyone guide me through the steps

Re: Missing versus excluded data

by Taylor Howell - Monday, August 24, 2020, 7:39 PM

For Question 5e, you are asked to recode the hivstatus variable so that 0=HIV Negative, 1=HIV Positive, and .=Excluded Record (missing).

I would first check how hivstatus is currently coded in the dataset to see if you need to convert the variable. Then, I would recode the variable using the rule above. You could then also check to see if you need to relabel.

Hope this helps!