Title stata.com
dtable — Create a table of descriptive statistics
Description Quick star t Menu Syntax
Options Remarks and examples Methods and formulas Appendix
Acknowledgments Reference Also see
Description
The dtable command allows you to easily create a table of descriptive (summary) statistics,
commonly known as “Table 1”. Optionally, you can add p-values from test statistics comparing
groups on summary statistics. You can also create a table complete with a title, notes, and more, and
then export it to a variety of file types.
Quick start
Create a table describing continuous variables x1 and x2 and factor variables f1 and f2
dtable x1 x2 i.f1 i.f2
Same as above, but report statistics for each level of variable group, with and without the total sample
dtable x1 x2 i.f1 i.f2, by(group)
dtable x1 x2 i.f1 i.f2, by(group, nototals)
Same as above, but test each variable for equality between groups using linear regressions for
continuous variables and Pearson χ
2
tests for factor variables
dtable x1 x2 i.f1 i.f2, by(group, tests)
Same as above, but rather than putting sample frequencies in the first row, put them in the column
header under each group label
dtable x1 x2 i.f1 i.f2, by(group, tests) sample(, place(seplabels))
Create a table of the median and interquartile interval for variables x1-x4
dtable, continuous(x1-x4, statistics(q2 iqi)) ///
define(iqi = q1 q3) sformat("[%s]" iqi)
Same as above, but add the Kruskal –Wallis test for equality between levels of variable group
dtable, continuous(x1-x4, statistics(q2 iqi) test(kwallis)) ///
define(iqi = q1 q3) sformat("[%s]" iqi) by(group, tests)
Create a table to describe variables using survey weights
dtable x* i.f*, svy
Same as above, but restrict the sample to the subpopulation where group equals 4
dtable x* i.f*, svy subpop(if group==4)
Menu
Statistics > Summaries, tables, and tests > Table of descriptive statistics
1
2 dtable — Create a table of descriptive statistics
Syntax
dtable
varlist
if
in
weight
, options
options Description
Main
by(varname
, by opts
) group results by categories of varname
column(col opts) change column header styles and titles
Survey
svy use survey settings
subpop(
varname
if
) identify a survey subpopulation
Sample
nosample suppress sample frequency statistics
sample(
title
, samp opts
) select sample frequency statistics
Continuous
continuous(
varlist
c
, cont opts
) select continuous variable statistics
Factors
factor(
varlist
f
, fact opts
) select factor-variable statistics
Composite
define(name = elements
, def opts
) define a composite result
Formats
nformat(% fmt
results
, basestyle
) specify numeric format
sformat(sfmt
results
) specify string format
Title
title(string) add table title
titlestyles(text styles) change table title styles
Notes
note(string) add table note
notestyles(text styles) change table note styles
Export
export( filename.suffix
, export opts
) export table
Options
no
varlabel display or suppress variable names or labels
no
fvlabel display or suppress factor values or labels
no
listwise specify how to handle missing values
halign(hvalue) specify horizontal alignment of item cells
name(cname) work with collection cname; default is name(DTable)
replace replace the collection
label( filename) specify the collection labels
style( filename
, override
) specify the collection style
warn show collect warnings
warn does not appear in the dialog box.
dtable — Create a table of descriptive statistics 3
by opts Description
no
tests show or suppress tests across groups
no
testnotes show or suppress notes about tests across groups
no
totals show or suppress statistics for the total sample
no
missing show or suppress numeric missing values in varname
col opts Description
summary(
title
, hcol opts
) change header styles for the summary column
by(
by style
, by hcol opts
) change header styles for the by() variable columns;
by style may be label, name, or hide
test(
title
, hcol opts
) change header styles for the test column
total(
title
, hcol opts
) change header styles for the total sample column
hcol opts Description
no
hide show or hide the column header title
halign(hvalue) specify horizontal alignment of column header cells
text styles change column header text styles
nohide and hide do not appear in the dialog box for summary().
by hcol opts Description
no
fvlabel display by() variable values or value labels
halign(hvalue) specify horizontal alignment of column header cells
text styles change column header text styles
text styles Description
font(
fontfamily
, font opts
) specify font style
smcl(smcl) specify formatting for SMCL files
latex(latex) specify L
A
T
E
X macro
shading(sspec) set background color, foreground color, and fill pattern
font opts Description
size(#
unit
) specify font size
color(color) specify font color
variant(variant) specify font variant and capitalization
no
bold specify whether to format text as bold
no
italic specify whether to format text as italic
no
strikeout specify whether to strike out text
no
underline specify whether to underline text
underline(upattern) specify underline pattern for text
4 dtable — Create a table of descriptive statistics
samp opts Description
statistics(nstats) select sample frequency statistics
place(place) select location for sample frequency statistics; place may be
items, inlabels, or seplabels
text styles change row header text styles
cont opts Description
statistics(cstats) select continuous variable statistics
test(ctest) select continuous variable test
text styles change row header text styles
fact opts Description
statistics(fstats) select factor-variable statistics
test(ftest) select factor-variable test
text styles change row header text styles
def opts Description
delimiter(char) use character as delimiter between elements
no
trim preserve or trim extra spaces from numeric formats
no
override preserve or override trim property of elements
replace redefine name if it already exists
suffix fileformat Output format
docx as(docx) Microsoft Word
html as(html) HTML 5 with CSS
pdf as(pdf) PDF
xlsx as(xlsx) Microsoft Excel 2007/2010 or newer
xls as(xls) Microsoft Excel 1997/2003
tex as(latex) L
A
T
E
X
smcl as(smcl) SMCL
txt as(txt) plain text
markdown as(markdown) Markdown
md as(markdown) Markdown
dtable — Create a table of descriptive statistics 5
export opts Description
as(fileformat) specify document type
replace overwrite existing file
docx options available when exporting to .docx files
html options available when exporting to .html files
pdf options available when exporting to .pdf files
excel options available when exporting to .xls and .xlsx files
tex options available when exporting to .tex files
smcl option available when exporting to .smcl files
txt option available when exporting to .txt files
md option available when exporting to .markdown and .md files
docx options Description
noisily show the putdocx commands used to export to the Microsoft
Word file
dofile(filename
, replace
) save the putdocx commands used for exporting to the named
do-file
html options Description
append append to an existing file
tableonly export only the table to the specified file
cssfile(cssfile) define the styles in cssfile instead of filename
prefix(prefix) use prefix to identify style classes
pdf options Description
noisily show the putpdf commands used to export to the PDF file
dofile(filename
, replace
) save the putpdf commands used for exporting to the named
do-file
excel options Description
noisily show the putexcel commands used to export to the Excel file
dofile(filename
, replace
) save the putexcel commands used for exporting to the named
do-file
sheet(sheetname
, replace
) specify the worksheet to use; the default sheet name
is Sheet1
cell(cell) specify the Excel upper-left cell as the starting position to
export the table; the default is cell(A1)
modify modify Excel file
noopen do not open Excel file in memory
noopen does not appear in the dialog box.
6 dtable — Create a table of descriptive statistics
tex options Description
append append to an existing file
tableonly export only the table to the specified file
smcl option Description
append append to an existing file
txt option Description
append append to an existing file
md option Description
append append to an existing file
varlist may contain i. notation, including level selection. Time-series operators and interactions are
not allowed.
fweights, aweights, iweights, and pweights are allowed; see [U] 11.1.6 weight.
results is a list of names taken from nstats, cstats, ctest, fstats, ftest, and composite results.
hvalue may be left, center, or right.
fontfamily specifies a valid font family.
unit may be in (inch), pt (point), or cm (centimeter). An inch is equivalent to 72 points and 2.54
centimeters. The default is pt.
variant may be allcaps, smallcaps, or normal.
variant(allcaps) changes the text to all uppercase letters; applicable when publishing items
from a collection to Microsoft Word, PDF, L
A
T
E
X, and HTML files.
variant(smallcaps) changes the text to use large capitals for uppercase letters and smaller
capitals for lowercase letters; applicable when publishing items from a collection to Microsoft
Word, L
A
T
E
X, and HTML files.
variant(normal) changes the font variant back to normal and leaves the capitalization unchanged
from the original text; applicable when publishing items from a collection to Microsoft Word,
PDF, L
A
T
E
X, and HTML files.
upattern may be any of the patterns listed in the Appendix. For example, underline(none) removes
the underline from the text, and underline(single) underlines the text. All other upatterns are
available only when publishing items from a collection to Microsoft Word.
smcl specifies the name of the SMCL directive to render text for SMCL output. The supported SMCL
directives are input, error, result, and text.
latex specifies the name of a L
A
T
E
X macro to render text for L
A
T
E
X output. Example L
A
T
E
X macro
names are textbf, textsf, textrm, and texttt. Custom L
A
T
E
X macros are also allowed. If text
is to be rendered in a cell, title, or note, then latex is translated to the following when you export
to L
A
T
E
X:
\latex {text}
dtable — Create a table of descriptive statistics 7
sspec is
background(bgcolor) foreground(fgcolor) pattern(fpattern)
bgcolor specifies the background color.
fgcolor specifies the foreground color.
fpattern specifies the fill pattern. A complete list of fill patterns is shown in the Appendix.
bgcolor, fgcolor, and color may be one of the colors listed in the Appendix; a valid RGB value in the
form ### ### ###, for example, 171 248 103; or a valid RRGGBB hex value in the form ######,
for example, ABF867.
sfmt is the specification for a string format in option sformat() and may contain a mix of text and
%s. Here %s refers to the statistic value that is formatted as specified using nformat(). The text
will be placed around the statistic as it is placed around %s in this option. For instance, to place
parentheses around the statistic, you can specify sformat("(%s)").
Two text characters must be specified using a special character sequence if you want them to be
displayed in your collection. To include %, type %%. To include \, type \\. For instance, to place
a percent sign after a statistic, you can specify sformat("%s%%").
Options
Main
by(varname
, by opts
) specifies that the statistics be displayed separately for each unique value
of varname; varname may be numeric or string.
by opts are the following:
tests and notests specify whether to show the tests comparing groups. tests displays the
tests comparing groups; notests suppresses them.
When tests is combined with the missing suboption, the test will be performed with the
missing values composing an additional group. By default, missing values are excluded from
tests.
testnotes and notestnotes specify whether to show a note detailing each test comparing
groups. testnotes displays the notes; notestnotes suppresses them.
totals and nototals specify whether to show the statistics for the total sample. totals
displays the statistics for the total sample; nototals suppresses them.
missing and nomissing specify whether to show numeric missing values in varname. missing
displays the numeric missing values; nomissing suppresses them.
When missing is combined with the tests suboption, missing values will also be included
in the test performed. By default, missing values are excluded from tests.
column(col opts) specifies the column header styles and titles. col opts are one or more of the
following.
summary(
title
,
no
hide halign(hvalue) text styles
) specifies the column header and
styles when global option by() is not specified. The default title is Summary.
hide and nohide specify whether to show the title. hide suppresses the title; nohide displays
the title.
8 dtable — Create a table of descriptive statistics
halign(hvalue) specifies how the column header cell is horizontally aligned. hvalue may be
left, center, or right.
text styles affect the text styles for the column header.
by(
by style
,
no
fvlabel halign(hvalue) text styles
) specifies the by() variable’s
column header and styles.
by style specifies the column header style for the by() variable. by style may be label, name,
or hide.
label specifies that the by() variable’s label be used. If the by() variable does not have
a label, then the name is used.
name specifies that the by() variable’s name be used.
hide specifies that the by() variable not be shown in the column header.
fvlabel and nofvlabel specify whether value labels for the levels of the by() variable
should be displayed. fvlabel displays the value labels; nofvlabel displays the values.
halign(hvalue) specifies how the column header cells are horizontally aligned. hvalue may
be left, center, or right.
text styles affect the text styles for the column header.
test(
title
,
no
hide halign(hvalue) text styles
) specifies the column header and text
styles for the test column. The default title is Test.
hide and nohide specify whether to show the title for the test column. hide suppresses the
title; nohide displays the title.
halign(hvalue) specifies how the test column header cell is horizontally aligned. hvalue may
be left, center, or right.
text styles affect the text styles for the test column header.
total(
title
,
no
hide halign(hvalue) text styles
) specifies the column header and text
styles for the total sample. The default title is Total.
hide and nohide specify whether to show the title for the total sample. hide suppresses the
title; nohide displays the title.
halign(hvalue) specifies how the total sample column header cell is horizontally aligned.
hvalue may be left, center, or right.
text styles affect the text styles for the total sample column header.
text styles are font(
fontfamily
, font opts
), smcl(smcl), latex(latex), and shading(sspec).
font(
fontfamily
, size(#
unit
) color(color) variant(variant)
no
bold
no
italic
no
strikeout
no
underline underline(upattern)
) specifies the font
style. These font style properties are applicable when exporting the table to Microsoft Word,
Microsoft Excel, PDF, HTML, and L
A
T
E
X files, unless otherwise specified.
fontfamily specifies a valid font family. This font style property is applicable when publishing
items from a collection to Microsoft Word, Microsoft Excel, PDF, and HTML files.
size(#
unit
) specifies the font size as a number optionally followed by units. This font
style property is applicable when publishing items from a collection to Microsoft Word,
Microsoft Excel, PDF, and HTML files.
color(color) specifies the text color.
variant(variant) specifies the font variant and capitalization.
dtable — Create a table of descriptive statistics 9
bold and nobold specify the font weight. bold changes the font weight to bold; nobold
changes the font weight back to normal.
italic and noitalic specify the font style. italic changes the font style to italic; noitalic
changes the font style back to normal.
strikeout and nostrikeout specify whether to add a strikeout mark to the text. strikeout
adds a strikeout mark to the text; nostrikeout changes the text back to normal.
underline(upattern), underline, and nounderline specify how to underline the text.
Only one of strikeout or underline is allowed when publishing to HTML files.
smcl(smcl) specifies how to render the text for SMCL output. This style property is applicable
only when publishing items from a collection to a SMCL file.
latex(latex) specifies how to render the text for L
A
T
E
X output. This style property is applicable
only when publishing items from a collection to a L
A
T
E
X file.
shading(sspec) sets the background color, foreground color, and fill pattern. The background
color is applicable when exporting the table to Microsoft Word, Microsoft Excel, PDF, HTML,
and L
A
T
E
X files. The foreground color and fill pattern are applicable when exporting the table
to Microsoft Word and Microsoft Excel.
Survey
svy specifies that dtable use the survey design information from svyset. This means all statistics
will be computed using the specified survey weights, where applicable. This option changes the
list of supported tests in continuous() and factor().
subpop(
varname
if
) specifies that estimates be computed for the single subpopulation identified
by the observations for which varname 6= 0 and that also meet the if conditions. Typically,
varname = 1 defines the subpopulation, and varname = 0 indicates observations not belonging
to the subpopulation. For observations whose subpopulation status is uncertain, varname should
be set to a missing value; such observations are dropped from the estimation sample. This option
implies the svy option.
See [SVY] Subpopulation estimation for more information.
10 dtable — Create a table of descriptive statistics
Sample
nosample and sample() control the display of sample frequency statistics.
nosample prevents reporting any sample frequency statistic.
sample(
title
, statistics(nstats) place(place) text styles
) specifies sample frequency
statistics, their row header title, and how they are composed in the table.
title specifies the row header for the sample frequency statistics. The default title is N.
statistics(nstats) specifies the sample frequency statistics to report.
nstats may include the following sample frequency statistics and any composite result defined
from them.
nstats Definition
frequency frequency
sumw sum of weights
percent percentage
proportion proportion
rawpercent unweighted percentage
rawproportion unweighted proportion
Without weights, the default sample frequency statistic is frequency.
With weights, the default sample frequency statistic is sumw.
With option by(), percent is added to the default sample frequency statistic.
place(place) controls where the sample frequency statistics are reported in the table.
place(items), the default, specifies that the sample frequency statistics be reported in the
first row of items in the table.
place(inlabels) specifies that the sample frequency statistics be added to the column
header labels.
place(seplabels) specifies that the sample frequency statistics be stacked in the column
header as separate labels.
text styles specifies the text styles for title.
dtable — Create a table of descriptive statistics 11
Continuous
continuous(
varlist
c
, statistics(cstats) test(ctest) text styles
) specifies continuous vari-
ables, their row header text styles, their statistics, and their test. If varlist
c
is not specified, then the
specified text styles, statistics, and test become the default for all continuous variables in varlist.
This option is repeatable, and when multiple text styles, statistics, or tests apply to a variable, the
rightmost specification is applied.
statistics(cstats) specifies the statistics to report for varlist
c
.
cstats may contain the following continuous variable statistics and any composite result defined
from them.
cstats Description
mean mean
semean standard error of the mean
sebinomial standard error of the mean, binomial
sepoisson standard error of the mean, Poisson
variance variance
sd standard deviation
skewness skewness
kurtosis kurtosis
cv coefficient of variation
svycv coefficient of variation (svy)
geomean geometric mean
geosd geometric standard deviation
count number of nonmissing values
median median
p# #th percentile
q1 first quartile
q2 second quartile
q3 third quartile
iqr interquartile range
min minimum value
max maximum value
range range
first first value
last last value
firstnm first nonmissing value
lastnm last nonmissing value
total total
rawtotal unweighted total
The default continuous variable statistics are mean and sd.
12 dtable — Create a table of descriptive statistics
test(ctest) specifies that ctest be computed for the continuous variables in varlist
c
. This option
has no effect if the by() option is not specified.
ctest is one of the following:
ctest Description
regress main effects test from a linear regression
poisson main effects test from a Poisson regression
lnormal main effects test from a log-normal regression
kwallis Kruskal–Wallis rank test
kwallis ties Kruskal–Wallis rank test, adjusted for ties
none suppress the test
kwallis and kwallis ties are not allowed with weights or the svy option.
The default test for continuous variables is regress.
For nonsurvey data, regress is equivalent to a one-way ANOVA, and when the by() variable
has two levels, regress is equivalent to a pooled t test.
text styles specifies the row header text styles for varlist
c
.
Factors
factor(
varlist
f
, statistics(fstats) test(ftest) text styles
) specifies factor variables, their
row header text styles, their statistics, and their test. If varlist
f
is not specified, then the specified
text styles, statistics, and test become the default for all factor variables in varlist.
This option is repeatable, and when multiple text styles, statistics, or tests apply to a variable, the
rightmost specification is applied.
statistics(fstats) specifies the statistics to report for varlist
f
.
fstats may contain the following factor-variable statistics and any composite result defined from
them.
fstats Definition
fvfrequency frequency of each factor-variable level
fvpercent percentage within each factor-variable level
fvproportion proportion within each factor-variable level
fvrawfrequency unweighted frequency of each factor-variable level
fvrawpercent unweighted percentage within each factor-variable level
fvrawproportion unweighted proportion within each factor-variable level
The default factor-variable statistics are fvfrequency and fvpercent.
dtable — Create a table of descriptive statistics 13
test(ftest) specifies that ftest be computed for the factor variables in varlist
f
. This option has
no effect if the by() option is not specified.
ftest is one of the following:
ftest Description
pearson Pearson’s χ
2
test
fisher Fisher’s exact test
lrchi2 likelihood-ratio χ
2
test
gamma Goodman and Kruskal’s gamma
kendall Kendall’s τ
b
cramer Cram
´
er’s V
svylr survey-adjusted likelihood-ratio test
svywald survey-adjusted Wald test
svyllwald survey-adjusted log-linear Wald test
none suppress the test
fisher, lrchi2, gamma, kendall, and cramer are not allowed with aweights, iweights,
pweights, or the svy option.
svylr, svywald, and svyllwald are allowed only with the svy option.
The default test for factor variables is pearson.
text styles specifies the row header text styles for varlist
f
.
Composite
define(name=elements
, def opts
) defines a composite result. Composite results give more control
over how statistics are composed and formatted in a table cell.
name is the name you choose for the new composite result.
elements is a list of statistics and other composite results.
def opts are delimiter(char),
no
trim,
no
override, and replace.
delimiter(char) changes the delimiter between elements. The default is delimiter(" ").
notrim and trim control the handling of extra spaces caused by numeric formats applied
to the elements. notrim preserves the extra spaces; trim, the default, removes the extra
spaces.
nooverride and override control handling of the trim property when an element is a
composite result. nooverride, the default, does not change the trim property of elements;
override applies the specified trim property to all elements.
replace permits you to redefine name if it already exists.
Formats
nformat(% fmt
results
, basestyle
) changes the numeric format, such as the number of
decimal places, for specified results. If results are not specified, the numeric format is changed
for all results.
results is a list of names taken from nstats, cstats, ctest, fstats, ftest, and composite results.
14 dtable — Create a table of descriptive statistics
This option is repeatable, and when multiple formats apply to one result, the rightmost specification
is applied.
This option does not affect the format of factor variables specified in varlist or the factor()
option. The default format of these variables is taken from the dataset.
basestyle indicates that the format be applied to results that do not already have their own
format instead of overriding the format for all results.
sformat(sfmt
results
) changes the string format for specified results. You can, for instance, add
symbols or text to the values reported in the table by modifying the string format.
sfmt may contain a mix of text and %s. Here %s refers to the numeric value that is formatted as
specified using nformat(). The text will be placed around the numeric values in your table as it
is placed around %s in this option. For instance, to place parentheses around the percent statistics,
you can specify sformat("(%s)" fvpercent percent).
results is a list of names taken from nstats, cstats, ctest, fstats, ftest, and composite results.
Two text characters must be specified using a special character sequence if you want them to be
displayed in your table. To include %, type %%. To include \, type \\. For instance, to place a
percent sign following percent statistics, you can specify sformat("%s%%" fvpercent percent).
This option is repeatable, and when multiple formats apply to one result, the rightmost specification
is applied.
Title
title(string) adds the text string as a title to the table.
titlestyles(text styles) changes the style for the table title. text styles are the following:
font(
fontfamily
, size(#
unit
) color(color) variant(variant)
no
bold
no
italic
no
strikeout
no
underline
) specifies the font style. These font style
properties are applicable when exporting the table to Microsoft Word, Microsoft Excel, PDF,
HTML, and L
A
T
E
X files, unless otherwise specified.
fontfamily specifies a valid font family. This font style property is applicable when publishing
items from a collection to Microsoft Word, Microsoft Excel, PDF, and HTML files.
size(#
unit
) specifies the font size as a number optionally followed by units. This font
style property is applicable when publishing items from a collection to Microsoft Word,
Microsoft Excel, PDF, and HTML files.
color(color) specifies the text color.
variant(variant) specifies the font variant and capitalization.
bold and nobold specify the font weight. bold changes the font weight to bold; nobold
changes the font weight back to normal.
italic and noitalic specify the font style. italic changes the font style to italic; noitalic
changes the font style back to normal.
strikeout and nostrikeout specify whether to add a strikeout mark to the title. strikeout
adds a strikeout mark to the title; nostrikeout changes the title back to normal.
underline and nounderline specify whether to underline the table title. underline adds
a single line under the title; nounderline removes the underline.
Only one of strikeout or underline is allowed when publishing to HTML files.
dtable — Create a table of descriptive statistics 15
smcl(smcl) specifies how to render the table title for SMCL output. This style property is applicable
only when publishing items from a collection to a SMCL file.
latex(latex) specifies how to render the table title for L
A
T
E
X output. This style property is
applicable only when publishing items from a collection to a L
A
T
E
X file.
shading(sspec) sets the background color, foreground color, and fill pattern. The background
color is applicable when exporting the table to Microsoft Word, Microsoft Excel, PDF, HTML,
and L
A
T
E
X files. The foreground color and fill pattern are applicable when exporting the table
to Microsoft Word and Microsoft Excel.
Notes
note(string) adds the text string as a note to the table. note() may be specified multiple times to
add multiple notes. Each note is placed on a new line.
notestyles(text styles) changes the style for the table notes. text styles are the following:
font(
fontfamily
, size(#
unit
) color(color) variant(variant)
no
bold
no
italic
no
strikeout
no
underline
) specifies the font style. These font style
properties are applicable when exporting the table to Microsoft Word, Microsoft Excel, PDF,
HTML, and L
A
T
E
X files, unless otherwise specified.
fontfamily specifies a valid font family. This font style property is applicable when publishing
items from a collection to Microsoft Word, Microsoft Excel, PDF, and HTML files.
size(#
unit
) specifies the font size as a number optionally followed by units. This font
style property is applicable when publishing items from a collection to Microsoft Word,
Microsoft Excel, PDF, and HTML files.
color(color) specifies the text color.
variant(variant) specifies the font variant and capitalization.
bold and nobold specify the font weight. bold changes the font weight to bold; nobold
changes the font weight back to normal.
italic and noitalic specify the font style. italic changes the font style to italic; noitalic
changes the font style back to normal.
strikeout and nostrikeout specify whether to add a strikeout mark to the notes. strikeout
adds a strikeout mark to the note; nostrikeout changes the note back to normal.
underline and nounderline specify whether to underline the table notes. underline adds
a single line under the notes; nounderline removes the underline.
Only one of strikeout or underline is allowed when publishing to HTML files.
smcl(smcl) specifies how to render the table notes for SMCL output. This style property is
applicable only when publishing items from a collection to a SMCL file.
latex(latex) specifies how to render the table notes for L
A
T
E
X output. This style property is
applicable only when publishing items from a collection to a L
A
T
E
X file.
shading(sspec) sets the background color, foreground color, and fill pattern. The background
color is applicable when exporting the table to Microsoft Word, Microsoft Excel, PDF, HTML,
and L
A
T
E
X files. The foreground color and fill pattern are applicable when exporting the table
to Microsoft Word and Microsoft Excel.
16 dtable — Create a table of descriptive statistics
Export
export(filename.suffix
, export opts
) exports the table to the specified file. export opts are the
following:
as(fileformat) specifies the file format to which the table is to be exported. This option is rarely
specified because, by default, dtable determines the format from the suffix of the file being
created.
replace permits dtable to overwrite an existing file.
noisily specifies that dtable show the commands used to export the table to Microsoft Word,
Microsoft Excel, and PDF files. The putdocx, putexcel, or putpdf command used to export
the table will be displayed.
dofile(filename
, replace
) specifies that dtable save to filename the commands used to
export the table to Microsoft Word, Microsoft Excel, and PDF files.
If filename already exists, it can be overwritten by specifying replace. If filename is specified
without an extension, .do is assumed.
append specifies that dtable append the table to an existing file.
This option is applicable when you export the table to an HTML, a L
A
T
E
X, a SMCL, a txt, or
a Markdown file. When you export to HTML and L
A
T
E
X files, the append option implies the
tableonly option. Furthermore, when you export to HTML files, if the target CSS file already
exists, dtable will also append to it.
tableonly specifies that only the table be exported to the specified HTML or L
A
T
E
X document.
By default, dtable produces complete HTML and L
A
T
E
X documents.
When you export to an HTML file, if the cssfile() option is not specified, a CSS filename is
constructed from filename, with the extension replaced with .css.
cssfile(cssfile) specifies that dtable define the styles in cssfile instead of filename when you
export to HTML.
prefix(prefix) specifies that dtable use prefix to identify style classes when you export to HTML.
sheet(sheetname
, replace
) saves to the worksheet named sheetname. For more information
about this option, see [RPT] putexcel.
cell(cell) specifies an Excel upper-left cell as the starting position to publish the table. The
default is cell(A1).
modify permits putexcel set to modify an Excel file. For more information about this option,
see [RPT] putexcel.
noopen prevents putexcel from opening the Excel file in memory for modification. It does not
appear on the dialog box. For more information about this option, see [RPT] putexcel.
Options
varlabel and novarlabel specify whether variable labels for the continuous and factor variables
should be displayed. varlabel displays the variable label; novarlabel displays variable names.
fvlabel and nofvlabel specify whether value labels for the factor variables should be displayed.
fvlabel displays value labels; nofvlabel displays the values.
listwise and nolistwise specify how to handle missing values across the continuous and factor
variables. listwise handles missing values through listwise deletion, meaning that the entire
observation is omitted from the sample if any continuous or factor variable is missing for that
dtable — Create a table of descriptive statistics 17
observation; nolistwise specifies that dtable omit an observation only if all the continuous and
factor variables are missing for that observation.
halign(hvalue) specifies how the item cells are horizontally aligned. hvalue may be left, center,
or right.
name(cname) specifies the collection for dtable to work with. The default is name(DTable).
replace permits dtable to overwrite the existing collection. This option is implied for
name(DTable).
label(filename) specifies the filename containing the collection labels to use for your table. Labels
in filename will be loaded for the table, and default labels will be used for any labels not specified
in filename.
style(filename
, override
) specifies the filename containing the collection styles to use for
your table. This might be a style you saved with collect style save or a predefined style
shipped with Stata. The dtable collection styles will be discarded, and only the collection styles
in filename will be applied. Note that the layout specification saved in filename will not be applied;
dtable will always use its predefined layout.
If you prefer the dtable collection styles but also want to apply any styles in filename, specify
override. If there are conflicts between the default collection styles and those in filename, the
ones in filename will take precedence.
The default is to use only the collection styles set in c(dtable style); see [TABLES] set
dtable style.
The following option is available with dtable but is not shown in the dialog box:
warn specifies that dtable display warnings from collect. By default, these warnings are suppressed.
Remarks and examples stata.com
Remarks are presented under the following headings:
Introduction
A first example
Describe variables across groups
Survey data
Save your style choices for next time
Composite results
The default style
Introduction
dtable allows you to easily create a table of summary statistics and export it to a variety of file
types, without any knowledge of the collection system. You can specify the continuous and factor
variables that you want described in your table. You can also customize the table by formatting the
results, changing the variable-specific statistics, adding a variable to identify sample groups, testing
each variable for equality between groups, adding a title and notes, and more.
In most cases, you will use dtable to easily create a table and export it to another format.
However, you can customize the table beyond the options that are available with dtable. When you
issue a dtable command, the results are stored in a collection called DTable. This collection is
replaced with each new dtable command, unless you specify a different name for the collection
in the name() option. You can make additional changes to the collection with the collect suite
of commands. To learn more about the collect commands, see [TABLES] Intro and the entries
discussed therein.
18 dtable — Create a table of descriptive statistics
A first example
In its simplest specification, you type dtable and a list of variables you want described in your
table. For our first example, we use data from the Second National Health and Nutrition Examination
Survey (NHANES II) (McDowell et al. 1981). Our table describes continuous variables for systolic
blood pressure, age, weight, and factor variables for race and health status.
. use https://www.stata-press.com/data/r18/nhanes2l
(Second National Health and Nutrition Examination Survey)
. dtable bpsystol age weight i.race i.hlthstat
Summary
N 10,351
Systolic blood pressure 130.882 (23.333)
Age (years) 47.580 (17.215)
Weight (kg) 71.898 (15.356)
Race
White 9,065 (87.6%)
Black 1,086 (10.5%)
Other 200 (1.9%)
Health status
Excellent 2,407 (23.3%)
Very good 2,591 (25.1%)
Good 2,938 (28.4%)
Fair 1,670 (16.2%)
Poor 729 (7.1%)
The default continuous variable statistics are the mean and standard deviation. The default factor-
variable statistics are the factor-level frequencies and percentages. You can change the statistics
reported for each continuous and factor variable; see the continuous() and factor() options for
the list of supported statistics.
Many options are also available to change the formatting and style of your table. Here we set the
format of means and standard deviations to show two decimal places, remove the first row of the
table with the overall sample size, and change the column header to be more descriptive of the table
results.
. dtable bpsystol age weight i.race i.hlthstat,
> nformat(%16.2fc mean sd)
> nosample
> column(summary(M(SD) / n(%)))
M(SD) / n(%)
Systolic blood pressure 130.88 (23.33)
Age (years) 47.58 (17.21)
Weight (kg) 71.90 (15.36)
Race
White 9,065 (87.6%)
Black 1,086 (10.5%)
Other 200 (1.9%)
Health status
Excellent 2,407 (23.3%)
Very good 2,591 (25.1%)
Good 2,938 (28.4%)
Fair 1,670 (16.2%)
Poor 729 (7.1%)
dtable — Create a table of descriptive statistics 19
Additionally, you can complete your table with a title and notes and even export it to another
format. We now add a title and export our table to the file mydtable.html.
. dtable bpsystol age weight i.race i.hlthstat,
> nformat(%16.2fc mean sd)
> nosample
> column(summary(M(SD) / n(%)))
> title(Table 1)
> export(mydtable.html)
Table 1
M(SD) / n(%)
Systolic blood pressure 130.88 (23.33)
Age (years) 47.58 (17.21)
Weight (kg) 71.90 (15.36)
Race
White 9,065 (87.6%)
Black 1,086 (10.5%)
Other 200 (1.9%)
Health status
Excellent 2,407 (23.3%)
Very good 2,591 (25.1%)
Good 2,938 (28.4%)
Fair 1,670 (16.2%)
Poor 729 (7.1%)
(collection DTable exported to file mydtable.html)
We could also export this table to a Microsoft Word, Microsoft Excel, L
A
T
E
X, Markdown, SMCL,
PDF, or plain-text file by specifying the appropriate file extension.
Describe variables across groups
Describing variables across samples in your dataset is easy with the by() option. In the following,
we revisit our original table, but we add the by() option to get summary statistics from the urban,
rural, and total samples.
. dtable bpsystol age weight i.race i.hlthstat, by(rural) sample(N (%))
Rural
Urban Rural Total
N (%) 6,548 (63.3%) 3,803 (36.7%) 10,351 (100.0%)
Systolic blood pressure 130.460 (23.526) 131.607 (22.980) 130.882 (23.333)
Age (years) 46.835 (17.484) 48.862 (16.666) 47.580 (17.215)
Weight (kg) 71.427 (15.363) 72.708 (15.314) 71.898 (15.356)
Race
White 5,419 (82.8%) 3,646 (95.9%) 9,065 (87.6%)
Black 968 (14.8%) 118 (3.1%) 1,086 (10.5%)
Other 161 (2.5%) 39 (1.0%) 200 (1.9%)
Health status
Excellent 1,609 (24.6%) 798 (21.0%) 2,407 (23.3%)
Very good 1,713 (26.2%) 878 (23.1%) 2,591 (25.1%)
Good 1,878 (28.7%) 1,060 (27.9%) 2,938 (28.4%)
Fair 950 (14.5%) 720 (19.0%) 1,670 (16.2%)
Poor 389 (5.9%) 340 (9.0%) 729 (7.1%)
20 dtable — Create a table of descriptive statistics
Next, we add the tests suboption in the by() option to report tests of equality between samples
and the nototals option to suppress the column of statistics from the total sample. We also specify
that we would like sample statistics to be in the column headers rather than the first row of the
table using the sample() option. Here we also add the column(by(hide)) option to suppress the
redundant label from variable rural and add notes that detail what is being reported.
. dtable bpsystol age weight i.race i.hlthstat,
> by(rural, nototals tests)
> column(by(hide))
> sample(, place(seplabels))
> note(Mean (Standard deviation): p-value from a pooled t-test.)
> note(Frequency (Percent%): p-value from Pearson test.)
note: using test across levels of for , , and
.
note: using test across levels of for and .
Urban Rural Test
6,548 (63.3%) 3,803 (36.7%)
Systolic blood pressure 130.460 (23.526) 131.607 (22.980) 0.016
Age (years) 46.835 (17.484) 48.862 (16.666) <0.001
Weight (kg) 71.427 (15.363) 72.708 (15.314) <0.001
Race
White 5,419 (82.8%) 3,646 (95.9%) <0.001
Black 968 (14.8%) 118 (3.1%)
Other 161 (2.5%) 39 (1.0%)
Health status
Excellent 1,609 (24.6%) 798 (21.0%) <0.001
Very good 1,713 (26.2%) 878 (23.1%)
Good 1,878 (28.7%) 1,060 (27.9%)
Fair 950 (14.5%) 720 (19.0%)
Poor 389 (5.9%) 340 (9.0%)
Mean (Standard deviation): p-value from a pooled t-test.
Frequency (Percent%): p-value from Pearson test.
For continuous variables, the Test column reports the p-value from linear regressions, which are
equivalent to pooled t tests because rural has two categories. For factor variables, the Test column
reports the p-values from Pearson χ
2
tests.
Survey data
dtable supports survey data by using the survey data characteristics in your dataset when you
specify the svy option. Your data must be svyset prior to calling dtable with the svy option. Our
data already have survey data characteristics.
. svyset
Sampling weights: finalwgt
VCE: linearized
Single unit: missing
Strata 1: strata
Sampling unit 1: psu
FPC 1: <zero>
The following table reports the survey weighted frequency, mean, standard deviation, factor-level
frequency, and factor-level percent of the specified variables.
dtable — Create a table of descriptive statistics 21
. dtable bpsystol age weight i.race i.hlthstat, svy
Summary
N 117,157,513
Systolic blood pressure 126.946 (21.401)
Age (years) 42.253 (15.502)
Weight (kg) 71.901 (15.433)
Race
White 102,999,549 (87.9%)
Black 11,189,236 (9.6%)
Other 2,968,728 (2.5%)
Health status
Excellent 32,187,335 (27.5%)
Very good 32,176,310 (27.5%)
Good 32,715,395 (28.0%)
Fair 14,380,261 (12.3%)
Poor 5,537,956 (4.7%)
Option by() is allowed with svy. dtable also has options for tests of equality between groups
that are allowed with svy. These tests account for the survey data characteristics in your data. In the
following, we revisit the urban and rural summary table, but we add the svy option, a title, and some
notes. We also add the column(by(hide)) option to suppress the redundant label from variable
rural and add the test() suboption to rename the Test column.
. dtable bpsystol age weight i.race i.hlthstat, svy
> by(rural, nototals tests)
> column(by(hide) test(p-value))
> title(Survey data summary)
> note(Mean (Standard deviation): p-value from linear regression.)
> note(Frequency (Percent%): p-value from Pearson test.)
> note(Statistics computed using the survey weights.)
> note(Tests adjusted for the survey design.)
note: using test across levels of for , , and
.
note: using test across levels of for and .
Survey data summary
Urban Rural p-value
N 79,965,794 (68.3%) 37,191,719 (31.7%)
Systolic blood pressure 126.607 (21.438) 127.675 (21.305) 0.406
Age (years) 41.805 (15.662) 43.215 (15.112) 0.024
Weight (kg) 71.322 (15.371) 73.144 (15.493) <0.001
Race
White 67,579,394 (84.5%) 35,420,155 (95.2%) <0.001
Black 9,936,159 (12.4%) 1,253,077 (3.4%)
Other 2,450,241 (3.1%) 518,487 (1.4%)
Health status
Excellent 22,781,784 (28.5%) 9,405,551 (25.3%) <0.001
Very good 22,867,496 (28.6%) 9,308,814 (25.1%)
Good 22,089,942 (27.7%) 10,625,453 (28.6%)
Fair 8,892,926 (11.1%) 5,487,335 (14.8%)
Poor 3,229,798 (4.0%) 2,308,158 (6.2%)
Mean (Standard deviation): p-value from linear regression.
Frequency (Percent%): p-value from Pearson test.
Statistics computed using the survey weights.
Tests adjusted for the survey design.
22 dtable — Create a table of descriptive statistics
Save your style choices for next time
If you find yourself repeating certain options each time you use dtable, you can save most
of those option choices for future tables. Use collect style save to store your dtable option
choices to a collection style file and then specify this file in the style() option the next time you
use dtable. If you store the collection style file in your PERSONAL folder, you can set that file as
the default style for dtable.
In the following, we use dtable to create a table with some option choices, save our style to a
file in the PERSONAL directory, and set that style file as the default for dtable.
. sysuse auto
(1978 automobile data)
. dtable mpg turn i.rep78,
> sample(Sample, statistic(frequency) place(seplabels))
> sformat("n=%s" frequency)
> by(foreign, tests nototals missing)
> continuous(, test(kwallis))
> factor(, test(pearson))
> title(Table 1. Summary by group with a test.)
> note(Kruskal-Wallis test for continuous variables.)
> note(Pearson’s test for factor variables.)
note: using test across levels of for and .
note: using test across levels of for .
Table 1. Summary by group with a test.
Car origin
Domestic Foreign Test
n=52 n=22
Mileage (mpg) 19.827 (4.743) 24.773 (6.611) 0.002
Turn circle (ft.) 41.442 (3.968) 35.409 (1.501) <0.001
Repair record 1978
1 2 (4.2%) 0 (0.0%) <0.001
2 8 (16.7%) 0 (0.0%)
3 27 (56.2%) 3 (14.3%)
4 9 (18.8%) 9 (42.9%)
5 2 (4.2%) 9 (42.9%)
Kruskal-Wallis test for continuous variables.
Pearson’s test for factor variables.
. personal dir
your personal ado-directory is C:\ado\personal\
(output omitted )
. collect style save "C:\ado\personal\mydtable"
(style from DTable saved to file C:\ado\personal\mydtable.stjson)
. set dtable_style mydtable
. display c(dtable_style)
mydtable
dtable — Create a table of descriptive statistics 23
Now these style choices are the defaults for tables created in the future. Here we create a similarly
styled table but from different data.
. use https://www.stata-press.com/data/r18/lbw
(Hosmer & Lemeshow data)
. dtable age i.smoke i.ht, by(low)
note: using test across levels of for .
note: using test across levels of for and .
Table 1. Summary by group with a test.
Birthweight<2500g
0 1 Test
n=130 n=59
Age of mother 23.662 (5.585) 22.305 (4.511) 0.248
Smoked during pregnancy
Nonsmoker 86 (66.2%) 29 (49.2%) 0.026
Smoker 44 (33.8%) 30 (50.8%)
Has history of hypertension
0 125 (96.2%) 52 (88.1%) 0.036
1 5 (3.8%) 7 (11.9%)
Kruskal-Wallis test for continuous variables.
Pearson’s test for factor variables.
To reset the dtable style back to its original default, type
. set dtable_style dtable
You can add the permanently option to set the default dtable style for future Stata sessions.
Composite results
dtable’s default style does not define any composite results, so you will need to define your own
with the define() option. Stata has a collection style named dtable composites that defines the
following composite results for you.
composite elements Description
idi p10 p90 Interdecile interval
iqi q1 q3 Interquartile interval
rangei min max Range interval
All of these results are bound in square brackets without a delimiter between their elements. To get
access to these composite results, use the style(dtable composites, override) option. You
can then specify idi, iqi, and rangei in the statistic() suboption of continuous(), just as
you would any other statistic.
You may want different formatting for these composite results, or you may want to create composite
results based on other statistics. The define() option allows you to customize your own composite
results. In the following, we use dtable to create a table that defines our own interval results with
some other option choices and then save the collection’s style to a file.
24 dtable — Create a table of descriptive statistics
. sysuse auto
(1978 automobile data)
. dtable, define(idi = p10 p90, delimiter("; "))
> define(iqi = q1 q3, delimiter("; "))
> define(rangei = min max, delimiter("; "))
> sample(Sample, statistic(frequency) place(seplabels))
> sformat("n=%s" frequency)
> sformat("{%s}" idi)
> sformat("[%s]" iqi)
> sformat("<%s>" rangei)
> continuous(mpg, statistic(p50 idi))
> continuous(turn, statistic(q2 iqi))
> continuous(trunk, statistic(mean rangei))
> title(Table 1. Location and interval.)
> note(Median {Interdecile interval})
> note(Median [Interquartile interval])
> note(Mean <Range interval>)
Table 1. Location and interval.
Summary
n=74
Mileage (mpg) 20.000 {14.000; 29.000}
Turn circle (ft.) 40.000 [36.000; 43.000]
Trunk space (cu. ft.) 13.757 <5.000; 23.000>
Median {Interdecile interval}
Median [Interquartile interval]
Mean <Range interval>
. collect style save dtable-intervals
(style from DTable saved to file dtable-intervals.stjson)
Now we can use this style and simply reference our composite results without having to redefine
them.
. dtable, continuous(mpg, statistic(p50 idi))
> continuous(turn, statistic(q2 iqi))
> continuous(trunk, statistic(mean rangei))
> style(dtable-intervals)
Table 1. Location and interval.
Summary
n=74
Mileage (mpg) 20.000 {14.000; 29.000}
Turn circle (ft.) 40.000 [36.000; 43.000]
Trunk space (cu. ft.) 13.757 <5.000; 23.000>
Median {Interdecile interval}
Median [Interquartile interval]
Mean <Range interval>
dtable — Create a table of descriptive statistics 25
The default style
Most of dtable’s option defaults are controllable in its collection style. The following options
are not part of dtable’s collection style:
svy
subpop()
export()
name()
replace
label()
style()
warn
When the by() option is not specified, the default column() header options are
summary("Summary", nohide
halign(center)
font(, nobold
noitalic
nostrikeout
nounderline)
smcl(text))
When a variable is specified in the by() option, its default options are
notest
testnotes
totals
nomissing
and its default column() header options are
by(label, nohide
fvlabel
halign(center)
font(, nobold
noitalic
nostrikeout
nounderline)
smcl(text))
total("Total", nohide
halign(center)
font(, nobold
noitalic
nostrikeout
nounderline)
smcl(text))
test("Test", nohide
halign(center)
font(, nobold
noitalic
nostrikeout
nounderline)
smcl(text))
26 dtable — Create a table of descriptive statistics
The default sample frequency title is “N”. When weights are not specified, the default sample
frequency statistic is frequency; otherwise, sumw is reported. When the by() option is specified,
percent is added to the default sample frequency statistic. The other default sample() options are
font(, nobold
noitalic
nostrikeout
nounderline)
smcl(text)
place(items)
The defaults for the continuous() option are
statistic(mean sd)
test(regress)
font(, nobold
noitalic
nostrikeout
nounderline)
smcl(text)
The defaults for the factor() option are
statistic(fvfrequency fvpercent)
test(pearson)
font(, nobold
noitalic
nostrikeout
nounderline)
smcl(text)
There are no predefined composite results in dtable’s default style.
The default formats are
nformat("%21.3fc", basestyle)
nformat("%21.0fc" N count sumw frequency fvfrequency fvrawfrequency)
nformat("%9.1fc" percent rawpercent fvpercent fvrawpercent)
sformat("(%s%%)" percent rawpercent fvpercent fvrawpercent)
sformat("(%s)" sd)
nformat("%6.3f" regress poisson lnormal kwallis kwallis_ties pearson
fisher lrchi2 svylr svywald svyllwald)
There is no default table title. The defaults for the titlestyles() option are
font(, nobold
noitalic
nostrikeout
nounderline)
smcl(text)
dtable — Create a table of descriptive statistics 27
There are no default table notes. The defaults for the notestyles() option are
font(, nobold
noitalic
nostrikeout
nounderline)
smcl(text)
The remaining default options for dtable are
varlabel
fvlabel
nolistwise
halign(right)
Methods and formulas
See Methods and formulas in [R] table for details on the statistics supported by dtable. When
the svy or subpop() option is specified, the weights from svyset are used to compute statistics
that support weights. In addition, the subpopulation variance bσ
2
is computed as
bσ
2
= s
2
M
n − 1
·
n
s
− 1
M
s
where s
2
is the sample variance as documented in [R] table, M is the estimated population size, n
is the full sample size, M
s
is the estimated subpopulation size, and n
s
is the subpopulation sample
size. The subpopulation standard deviation is computed as bσ =
√
bσ
2
and coefficient of variation (cv)
is computed as bσ/x.
Let y be a continuous variable and by(group, tests) be in effect. For test(regress), dtable
reports the p-value from test [#1] after
regress y i.group
if
in
weight
For test(poisson), dtable reports the p-value from test [#1] after
poisson y i.group
if
in
weight
For test(lnormal), dtable reports the p-value from test [#1] after
gsem y <- i.group
if
in
weight
, family(lognormal)
For test(kwallis) and test(kwallis ties), dtable reports the p-values corresponding with
the χ
2
statistics reported by
kwallis y
if
in
, by(group)
When the svy or subpop() option is specified, the prefix
svy, subpop(
varname
if
) :
is used with regress, poisson, and gsem. For more details, see [R] test, [R] regress, [R] poisson,
[SEM] gsem, [R] kwallis, and [SVY] svy.
28 dtable — Create a table of descriptive statistics
Let f be a factor variable and by(group, tests) be in effect. For nonsurvey data, the factor-
variable test across groups is performed using
tabulate f group
if
in
weight
, opt
where opt is an option of tabulate corresponding with the specified test(ftest) using the following
table:
ftest opt
pearson chi2
fisher exact
lrchi2 lrchi2
gamma gamma
kendall taub
cramer V
For survey data, dtable reports the p-value from
svy, subpop(
varname
if
): tabulate f group
if
in
, opt
where opt is an option of svy: tabulate corresponding with the specified test(ftest) using the
following table:
ftest opt
pearson pearson
svylr lr
svywald wald
svyllwald llwald
pearson, fisher, lrchi2, svylr, svywald, and svyllwald report a p-value. gamma, kendall,
and cramer report the computed measure of association. For more details, see [R] tabulate twoway
and [SVY] svy: tabulate twoway.
dtable — Create a table of descriptive statistics 29
Appendix
Colors
bgcolor, fgcolor, and color
aliceblue darkslategray lightsalmon palevioletred
antiquewhite darkturquoise lightseagreen papayawhip
aqua darkviolet lightskyblue peachpuff
aquamarine deeppink lightslategray peru
azure deepskyblue lightsteelblue pink
beige dimgray lightyellow plum
bisque dodgerblue lime powderblue
black firebrick limegreen purple
blanchedalmond floralwhite linen red
blue forestgreen magenta rosybrown
blueviolet fuchsia maroon royalblue
brown gainsboro mediumaquamarine saddlebrown
burlywood ghostwhite mediumblue salmon
cadetblue gold mediumorchid sandybrown
chartreuse goldenrod mediumpurple seagreen
chocolate gray mediumseagreen seashell
coral green mediumslateblue sienna
cornflowerblue greenyellow mediumspringgreen silver
cornsilk honeydew mediumturquoise skyblue
crimson hotpink mediumvioletred slateblue
cyan indianred midnightblue slategray
darkblue indigo mintcream snow
darkcyan ivory mistyrose springgreen
darkgoldenrod khaki moccasin steelblue
darkgray lavender navajowhite tan
darkgreen lavenderblush navy teal
darkkhaki lawngreen oldlace thistle
darkmagenta lemonchiffon olive tomato
darkolivegreen lightblue olivedrab turquoise
darkorange lightcoral orange violet
darkorchid lightcyan orangered wheat
darkred lightgoldenrodyellow orchid white
darksalmon lightgray palegoldenrod whitesmoke
darkseagreen lightgreen palegreen yellow
darkslateblue lightpink paleturquoise yellowgreen
30 dtable — Create a table of descriptive statistics
Underline patterns
upattern
none dashLong
single dashLongHeavy
words dotDash
double dashDotHeavy
thick dotDotDash
dotted dashDotDotHeavy
dottedHeavy wave
dash wavyHeavy
dashedHeavy wavyDouble
Shading patterns
fpattern
nil pct20
clear pct25
solid pct30
horzStripe pct35
vertStripe pct37
reverseDiagStripe pct40
diagStripe pct45
horzCross pct50
diagCross pct55
thinHorzStripe pct60
thinVertStripe pct62
thinReverseDiagStripe pct65
thinDiagStripe pct70
thinHorzCross pct75
thinDiagCross pct80
pct5 pct85
pct10 pct87
pct12 pct90
pct15 pct95
Acknowledgments
We thank Phil Clayton of the ANZDATA Registry at the Royal Adelaide Hospital, Mark Chatfield
of the University of Queensland, and Niels Henrik Bruun of Region Nordjylland for writing similar
commands for creating tables of descriptive statistics. We also thank Joanna Dipnall of Monash
University, Australia, for suggesting a unified framework for tables.
dtable — Create a table of descriptive statistics 31
Reference
McDowell, A., A. Engel, J. T. Massey, and K. Maurer. 1981. Plan and operation of the Second National Health and
Nutrition Examination Survey, 1976–1980. Vital and Health Statistics 1(15): 1–144.
Also see
[R] table intro — Introduction to tables of frequencies, summaries, and command results
[R] table summary — Table of summary statistics
[TABLES] Intro — Introduction
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 companies. Copyright
c
1985–2023 StataCorp LLC, College Station, TX,
USA. All rights reserved.
®
For suggested citations, see the FAQ on citing Stata documentation.
