Hi,
My question is about labelling the missing values.
In the data set I have (imported from SPSS) the missing values were entered from -9 to -4 indicating whether they were left intentionally blank, dont know, prefer not to say etc (from the data dictionary). Now that I have these in Stata it is analysing them as part of the analysis rather than recording them as missing observations. The information I have found is mostly on system generated missing values.
My question is what is the best way to code for these missing values that have been entered into the data as negative numbers so they do not skew the results.
Would it be advisable to recode them as missing?

Please see below how these appear when i run the code: summarize as an example

Variable | Obs Mean Std. dev. Min Max
-------------+---------------------------------------------------------
var1 | 0
var2 | 1,007 1.011716 .1076382 1 2
var3 | 1,007 1.534856 1.145017 -7 5
var4 | 1,007 1.204452 .7073746 -7 5
var5 | 1,007 1.543644 .6141639 -9 2
-------------+---------------------------------------------------------

Ps new user here, I am familiarizing myself with the user guidelines, which I hope this post adheres to, however I apologise in advance if this is not the case.
Thankyou!