Keeping only numbers after the decimal

[Erin Mead-Morse]

I am cleaning someone else's data set, and having trouble with one variable: infant's birth weight. It was recorded as lbs.oz. So, for example, a value of "7.80" means 7lbs 8oz, not 7.8 lbs (which would correspond to 7lbs 12.8 oz). I need to fix this so that, for example, a value of "7.80" means 7.8lbs. I used the command "gen bo_wght_lbs=int(bo_wght)" to create a new variable with the lbs (e.g., the "7"). Now I am at a loss as to how to extrapolate the values after the decimal (e.g., the ".80"). Once I can create a new variable with just the ounces after the decimal point, I can convert it to pounds and combine the two variables together into one variable for lbs. Thanks in advance.

[Maarten Buis]

clear
input weirdweight
7.8
end
gen lbs = floor(weirdweight)
gen oz = mod(weirdweight,1)*10
gen realweight = lbs * 0.45359237 + 0.0283495231 * oz

[Maarten Buis]

If you have 7.1 does that mean 7lbs and 1 oz or 7lbs and 10oz (since there are 16oz in lbs)? A consistent scheme like that would mean 7lbs and 8oz would be coded as 7.08 and not 7.8. (Anyone need any convincing metric is king when you want to work with decimal numbers?)

[Erin Mead-Morse]

If you have 7.1 does that mean 7lbs and 1 oz or 7lbs and 10oz (since there are 16oz in lbs)? A consistent scheme like that would mean 7lbs and 8oz would be coded as 7.08 and not 7.8. (Anyone need any convincing metric is king when you want to work with decimal numbers?)

Ugh I hadn't even thought of that. What a pain this data set is. I agree with you about metric. Thank you for pointing this out.