destring — Convert string variables to numeric variables and vice versa

Description Quick start Menu

Syntax Options for destring Options for tostring
Remarks and examples Acknowledgment References
Also see

Description
destring converts variables in varlist from string to numeric. If varlist is not specified, destring
will attempt to convert all variables in the dataset from string to numeric. Characters listed in ignore()
are removed. Variables in varlist that are already numeric will not be changed. destring treats both
empty strings “ ” and “.” as indicating sysmiss (.) and interprets the strings “.a”, “.b”, . . . , “.z” as
the extended missing values .a, .b, . . . , .z; see [U] 12.2.1 Missing values. destring also ignores any
leading or trailing spaces so that, for example, “ ” is equivalent to “ ” and “ . ” is equivalent to “.”.
tostring converts variables in varlist from numeric to string. The most compact string format pos-
sible is used. Variables in varlist that are already string will not be converted.

Quick start
Convert strg1 from string to numeric, and place result in num1
destring strg1, generate(num1)
Same as above, but ignore the % character in strg1
destring strg1, generate(num1) ignore(%)
Same as above, but return . for observations with nonnumeric characters
destring strg1, generate(num1) force
Convert num2 from numeric to string, and place result in strg2
tostring num2, generate(strg2)
Same as above, but format with a leading zero and 3 digits after the decimal
tostring num2, generate(strg2) format(%09.3f)

Menu
destring
Data > Create or change data > Other variable-transformation commands > Convert variables from string to
numeric
tostring
Data > Create or change data > Other variable-transformation commands > Convert variables from numeric to
string

1
 destring — Convert string variables to numeric variables and vice versa 2

Syntax
Convert string variables to numeric variables
destring [ varlist ], { generate(newvarlist ) | replace } [ destring options ]

Convert numeric variables to string variables

tostring varlist , { generate(newvarlist ) | replace } [ tostring options ]

destring options Description

∗
generate(newvarlist ) generate newvar1 , . . . , newvar𝑘 for each variable in varlist
∗
replace replace string variables in varlist with numeric variables
ignore(”chars” [ , ignoreopts ]) remove specified nonnumeric characters, as characters or as
bytes, and illegal Unicode characters
force convert nonnumeric strings to missing values
float generate numeric variables as type float
percent convert percent variables to fractional form
dpcomma convert variables with commas as decimals to period-decimal
format
∗ Either generate(newvarlist ) or replace is required.

tostring options Description

∗
generate(newvarlist ) generate newvar1 , . . . , newvar𝑘 for each variable in varlist
∗
replace replace numeric variables in varlist with string variables
force force conversion ignoring information loss
format(format ) convert using specified format
usedisplayformat convert using display format
∗ Either generate(newvarlist ) or replace is required.

Options for destring

Either generate() or replace must be specified. With either option, if any string variable contains
nonnumeric characters not specified with ignore(), then no corresponding variable will be generated,
nor will that variable be replaced (unless force is specified).
generate(newvarlist ) specifies that a new variable be created for each variable in varlist. newvarlist
must contain the same number of new variable names as there are variables in varlist. If varlist is
not specified, destring attempts to generate a numeric variable for each variable in the dataset;
newvarlist must then contain the same number of new variable names as there are variables in the
dataset. Any variable labels or characteristics will be copied to the new variables created.
replace specifies that the variables in varlist be converted to numeric variables. If varlist is not spec-
ified, destring attempts to convert all variables from string to numeric. Any variable labels or
characteristics will be retained.
 destring — Convert string variables to numeric variables and vice versa 3

ignore(”chars” [ , ignoreopts ]) specifies nonnumeric characters be removed. ignoreopts may be

aschars, asbytes, or illegal. The default behavior is to remove characters as characters, which is
the same as specifying aschars. asbytes specifies removal of all bytes included in all characters in
the ignore string, regardless of whether these bytes form complete Unicode characters. illegal spec-
ifies removal of all illegal Unicode characters, which is useful for removing high-ASCII characters.
illegal may not be specified with asbytes. If any string variable still contains any nonnumeric or
illegal Unicode characters after the ignore string has been removed, no action will take place for that
variable unless force is also specified. Note that to Stata the comma is a nonnumeric character; see
also the dpcomma option below.
force specifies that any string values containing nonnumeric characters, in addition to any specified
with ignore(), be treated as indicating missing numeric values.
float specifies that any new numeric variables be created initially as type float. The default is type
double; see [D] Data types. destring attempts automatically to compress each new numeric vari-
able after creation.
percent removes any percent signs found in the values of a variable, and all values of that variable are
divided by 100 to convert the values to fractional form. percent by itself implies that the percent
sign, “ % ”, is an argument to ignore(), but the converse is not true.
dpcomma specifies that variables with commas as decimal values should be converted to have periods as
decimal values.

Options for tostring

Either generate() or replace must be specified. If converting any numeric variable to string would
result in loss of information, no variable will be produced unless force is specified. For more details,
see force below.
generate(newvarlist ) specifies that a new variable be created for each variable in varlist. newvarlist
must contain the same number of new variable names as there are variables in varlist. Any variable
labels or characteristics will be copied to the new variables created.
replace specifies that the variables in varlist be converted to string variables. Any variable labels or
characteristics will be retained.
force specifies that conversions be forced even if they entail loss of information. Loss of information
means one of two circumstances: 1) The result of real(strofreal(varname, ”format ”)) is not
equal to varname; that is, the conversion is not reversible without loss of information; 2) replace
was specified, but a variable has associated value labels. In circumstance 1, it is usually best to
specify usedisplayformat or format(). In circumstance 2, value labels will be ignored in a forced
conversion. decode (see [D] encode) is the standard way to generate a string variable based on value
labels.
format(format ) specifies that a numeric format be used as an argument to the strofreal() function,
which controls the conversion of the numeric variable to string. For example, a format of %7.2f spec-
ifies that numbers are to be rounded to two decimal places before conversion to string. See Remarks
and examples below and [FN] String functions and [D] format. format() cannot be specified with
usedisplayformat.
usedisplayformat specifies that the current display format be used for each variable. For example,
this option could be useful when using US Social Security numbers or daily or other dates with some
%d or %t format assigned. usedisplayformat cannot be specified with format().
 destring — Convert string variables to numeric variables and vice versa 4

Remarks and examples

Remarks are presented under the following headings:
destring
tostring
Saved characteristics
Video example

destring

Example 1
We read in a dataset, but somehow all the variables were created as strings. The variables contain no
nonnumeric characters, and we want to convert them all from string to numeric data types.
. use https://www.stata-press.com/data/r19/destring1
. describe
Contains data from https://www.stata-press.com/data/r19/destring1.dta
Observations: 10
Variables: 5 3 Mar 2024 10:15

Variable Storage Display Value

name type format label Variable label

id str3 %9s
num str3 %9s
code str4 %9s
total str5 %9s
income str5 %9s

Sorted by:
. list

id num code total income

1. 111 243 1234 543 23423

2. 111 123 2345 67854 12654
3. 111 234 3456 345 43658
4. 222 345 4567 57 23546
5. 333 456 5678 23 21432

6. 333 567 6789 23465 12987

7. 333 678 7890 65 9823
8. 444 789 8976 23 32980
9. 444 901 7654 23 18565
10. 555 890 6543 423 19234

. destring, replace
id: all characters numeric; replaced as int
num: all characters numeric; replaced as int
code: all characters numeric; replaced as int
total: all characters numeric; replaced as long
income: all characters numeric; replaced as long
 destring — Convert string variables to numeric variables and vice versa 5

. describe
Contains data from https://www.stata-press.com/data/r19/destring1.dta
Observations: 10
Variables: 5 3 Mar 2024 10:15

Variable Storage Display Value

name type format label Variable label

id int %10.0g
num int %10.0g
code int %10.0g
total long %10.0g
income long %10.0g

Sorted by:
Note: Dataset has changed since last saved.
. list

id num code total income

1. 111 243 1234 543 23423

2. 111 123 2345 67854 12654
3. 111 234 3456 345 43658
4. 222 345 4567 57 23546
5. 333 456 5678 23 21432

6. 333 567 6789 23465 12987

7. 333 678 7890 65 9823
8. 444 789 8976 23 32980
9. 444 901 7654 23 18565
10. 555 890 6543 423 19234

Example 2
Our dataset contains the variables date, price, and percent. These variables were accidentally read
into Stata as string variables because they contain spaces, dollar signs, commas, and percent signs. We
will leave the date variable as a string so that we can use the date() function to convert it to a numeric
date. For price and percent, we want to remove all of the nonnumeric characters and create new
variables containing numeric values. After removing the percent sign, we want to convert the percent
variable to decimal form.
 destring — Convert string variables to numeric variables and vice versa 6

. use https://www.stata-press.com/data/r19/destring2, clear

. describe
Contains data from https://www.stata-press.com/data/r19/destring2.dta
Observations: 10
Variables: 3 3 Mar 2024 22:50

Variable Storage Display Value

name type format label Variable label

date str14 %10s

price str11 %11s
percent str3 %9s

Sorted by:
. list

date price percent

1. 1999 12 10 $2,343.68 34%

2. 2000 07 08 $7,233.44 86%
3. 1997 03 02 $12,442.89 12%
4. 1999 09 00 $233,325.31 6%
5. 1998 10 04 $1,549.23 76%

6. 2000 03 28 $23,517.03 35%

7. 2000 08 08 $2.43 69%
8. 1997 10 20 $9,382.47 32%
9. 1998 01 16 $289,209.32 45%
10. 1999 11 12 $8,282.49 1%

. destring price percent, generate(price2 percent2) ignore(”$,%”) percent

price: characters $ , removed; price2 generated as double
percent: character % removed; percent2 generated as double
. describe
Contains data from https://www.stata-press.com/data/r19/destring2.dta
Observations: 10
Variables: 5 3 Mar 2024 22:50

Variable Storage Display Value

name type format label Variable label

date str14 %10s

price str11 %11s
price2 double %10.0g
percent str3 %9s
percent2 double %10.0g

Sorted by:
Note: Dataset has changed since last saved.
 destring — Convert string variables to numeric variables and vice versa 7

. list

date price price2 percent percent2

1. 1999 12 10 $2,343.68 2343.68 34% .34

2. 2000 07 08 $7,233.44 7233.44 86% .86
3. 1997 03 02 $12,442.89 12442.89 12% .12
4. 1999 09 00 $233,325.31 233325.31 6% .06
5. 1998 10 04 $1,549.23 1549.23 76% .76

6. 2000 03 28 $23,517.03 23517.03 35% .35

7. 2000 08 08 $2.43 2.43 69% .69
8. 1997 10 20 $9,382.47 9382.47 32% .32
9. 1998 01 16 $289,209.32 289209.32 45% .45
10. 1999 11 12 $8,282.49 8282.49 1% .01

tostring
Conversion of numeric data to string equivalents can be problematic. Stata, like most software, holds
numeric data to finite precision and in binary form. See the discussion in [U] 13.12 Precision and
problems therein. If no format() is specified, tostring uses the format %12.0g. This format is, in
particular, sufficient to convert integers held as bytes, ints, or longs to string equivalent without loss of
precision.
However, users will often need to specify a format themselves, especially when the numeric data have
fractional parts and for some reason a conversion to string is required.

Example 3
Our dataset contains a string month variable and numeric year and day variables. We want to convert
the three variables to a %td date.
. use https://www.stata-press.com/data/r19/tostring, clear
. list

id month day year

1. 123456789 jan 10 2001

2. 123456710 mar 20 2001
3. 123456711 may 30 2001
4. 123456712 jun 9 2001
5. 123456713 oct 17 2001

6. 123456714 nov 15 2001

7. 123456715 dec 28 2001
8. 123456716 apr 29 2001
9. 123456717 mar 11 2001
10. 123456718 jul 3 2001

. tostring year day, replace

year was float now str4
day was float now str2
. generate date = month + ”/” + day + ”/” + year
. generate edate = date(date, ”MDY”)
 destring — Convert string variables to numeric variables and vice versa 8

. format edate %td

. list

id month day year date edate

1. 123456789 jan 10 2001 jan/10/2001 10jan2001

2. 123456710 mar 20 2001 mar/20/2001 20mar2001
3. 123456711 may 30 2001 may/30/2001 30may2001
4. 123456712 jun 9 2001 jun/9/2001 09jun2001
5. 123456713 oct 17 2001 oct/17/2001 17oct2001

6. 123456714 nov 15 2001 nov/15/2001 15nov2001

7. 123456715 dec 28 2001 dec/28/2001 28dec2001
8. 123456716 apr 29 2001 apr/29/2001 29apr2001
9. 123456717 mar 11 2001 mar/11/2001 11mar2001
10. 123456718 jul 3 2001 jul/3/2001 03jul2001

Saved characteristics
Each time the destring or tostring commands are issued, an entry is made in the characteristics
list of each converted variable. You can type char list to view these characteristics.
After example 2, we could use char list to find out what characters were removed by the destring
command.
. char list
price2[destring]: Characters removed were: $ ,
price2[destring_cmd]: destring price percent, generate(price2 percent..
percent2[destring]: Character removed was: %
percent2[destring_cmd]: destring price percent, generate(price2 percent..

Video example
How to convert a string variable to a numeric variable

Acknowledgment
destring and tostring were originally written by Nicholas J. Cox of the Department of Geography
at Durham University, UK, who is coeditor of the Stata Journal and author of Speaking Stata Graphics.

References
Cox, N. J. 1999a. dm45.1: Changing string variables to numeric: Update. Stata Technical Bulletin 49: 2. Reprinted in
Stata Technical Bulletin Reprints, vol. 9, p. 14. College Station, TX: Stata Press.
———. 1999b. dm45.2: Changing string variables to numeric: Correction. Stata Technical Bulletin 52: 2. Reprinted in
Stata Technical Bulletin Reprints, vol. 9, p. 14. College Station, TX: Stata Press.
———. 2011. Speaking Stata: MMXI and all that: Handling Roman numerals within Stata. Stata Journal 11: 126–142.
Cox, N. J., and W. W. Gould. 1997. dm45: Changing string variables to numeric. Stata Technical Bulletin 37: 4–6.
Reprinted in Stata Technical Bulletin Reprints, vol. 7, pp. 34–37. College Station, TX: Stata Press.
Cox, N. J., and C. B. Schechter. 2018. Speaking Stata: Seven steps for vexatious string variables. Stata Journal 18:
981–994.
 destring — Convert string variables to numeric variables and vice versa 9

Cox, N. J., and J. B. Wernow. 2000a. dm80: Changing numeric variables to string. Stata Technical Bulletin 56: 8–12.
Reprinted in Stata Technical Bulletin Reprints, vol. 10, pp. 24–28. College Station, TX: Stata Press.
———. 2000b. dm80.1: Update to changing numeric variables to string. Stata Technical Bulletin 57: 2. Reprinted in
Stata Technical Bulletin Reprints, vol. 10, pp. 28–29. College Station, TX: Stata Press.
Jeanty, P. W. 2013. Dealing with identifier variables in data management and analysis. Stata Journal 13: 699–718.

Also see
[D] egen — Extensions to generate
[D] encode — Encode string into numeric and vice versa
[D] generate — Create or change contents of variable
[D] split — Split string variables into parts
[FN] String functions

Stata, Stata Press, and Mata are registered trademarks of StataCorp LLC. Stata and Stata
®
Press are registered trademarks with the World Intellectual Property Organization of the
United Nations. StataNow and NetCourseNow are trademarks of StataCorp LLC. Other
brand and product names are registered trademarks or trademarks of their respective com-
panies. Copyright © 1985–2025 StataCorp LLC, College Station, TX, USA. All rights
reserved.
For suggested citations, see the FAQ on citing Stata documentation.