Virata manual

destring

Turns a column of text that is really numbers into actual numbers.

What we have

  • generate() or replace
  • ignore(), force, percent, float and dpcomma

Coming later

Nothing known to be missing.

How you write it

destring varlist, {replace | generate(newvars)} [ignore(",") force percent float]

Square brackets mean optional. A vertical bar means choose one.

What it does

Spreadsheets produce these constantly: a column that reads 1,250 or 45% or 37 and arrives as text, so it will not average. destring converts it.

ignore(",") strips the thousands separators, percent handles the sign, and force blanks out anything that still is not a number rather than refusing the whole column.

Watch out. Use force only after you have looked at what it will blank out — it silently turns 'refused' and 'n/a' into missing values, which may or may not be what you want.

Try it

Every example below was run after sysuse vira.sample02, clear.

. tostring age, generate(age_text)
age: generated as str
The same thing in pandas and in R

R

# no R form for `tostring` yet -- see docs/r-equivalents.md
. destring age_text, generate(age_again)
age_text: all characters numeric; generated as byte
The same thing in pandas and in R

pandas

import pandas as pd

df[['age_again']] = df[['age_text']].apply(pd.to_numeric, errors="coerce")

R

# no R form for `destring` yet -- see docs/r-equivalents.md
. describe age_text age_again
Contains data from .../virata/datasets/vira.sample02.dta
 Observations:        6,000
    Variables:            2
------------------------------------------------------------------------------
Variable      Storage   Display    Value
    name         type    format    label      Variable label
------------------------------------------------------------------------------
age_text           str       %9s                 
age_again         byte     %9.0g                 
------------------------------------------------------------------------------
The same thing in pandas and in R

pandas

df.dtypes

R

library(dplyr)
glimpse(df)

# This R has NOT been run against a real R yet: read it as documentation.

Nearby

generate replace egen recode rename order