Generating a new variable with an if command

[Chris Rooney]

Hi all,

I have a series of 6 datasets - for ease of reference, let's call them 2015, 2016, 2017, 2018, 2019, 2020.

I have appended the datasets to make my life a bit easier, however, I am having trouble when creating a real wage variable for each year (the inflation differs across each year, hence I can't apply the same inflation rate across all years).

In my appended dataset, I created a variable "year" to indicate which year the observation came from.

I used the following code to generate real wages for 2015

gen realearnings_2015=( s5earnings_employees/74.1)*100 if inlist(year, "2015")

however, I get the following error message:

type mismatch

[Andrew Musau]

Here you have the -if- qualifier and not the if command. The error is consistent with the variable year being a numerical variable and not a string variable as you are treating it with

inlist(year, "2015")

by including the double quotes. No need for the -inlist()- function if you only have one element. Be direct.

gen realearnings_2015=(s5earnings_employees/74.1)*100 if year==2015

[Carlo Lazzaro]

Chris:
try:

gen realearnings_2015=( s5earnings_employees/74.1)*100 if inlist(year, 2015)

[William Lisowski]

Cosider the following approach.

generate factor = .
replace factor = 74.1 if year==2015
// similar replace commands for 2016-2020
generate real_earnings = (s5earnings_employees/factor)*100

The, consistent with what you have for nominal wages (s5earnings_employees), and I expect for all your other variables in the appended dataset, you will have one variable with real wages (real_earnings) in all years.