Table of frequencies, summaries, and command results

Title stata.com

table — Table of frequencies, summaries, and command results

Description Quick star t Menu Syntax

Options Remarks and examples Stored results Methods and formulas

Also see

Description

table is a ﬂexible command for creating tables of many types—tabulations, tables of summary

statistics, tables of regression results, and more. table can calculate summary statistics to display in

the table. table can also include results from other Stata commands.

Quick start

Two-way tabulation of a1 and a2

table a1 a2

Table of means for v1 and v2 across the levels of a1

table a1, statistic(mean v1 v2)

Two-way table with a1 deﬁning rows and a2 deﬁning columns, with frequencies and pairwise

correlation coefﬁcients between v3 and v4 computed for every cell

table a1 a2, command(pwcorr v3 v4)

Table of regression coefﬁcients with means of the covariates; rows correspond to covariates and

columns correspond to the statistics

table (colname) (statcmd result),

command(regress y x1 x2)

statistic(mean x1 x2)

Same as above, and include standard deviations for the covariates

table (colname) (statcmd result),

command(regress y x1 x2)

statistic(mean x1 x2)

statistic(sd x1 x2)

Statistics > Summaries, tables, and tests > Tables of frequencies, summaries, and command results

2 table — Table of frequencies, summaries, and command results

Syntax

Basic syntax for a one-way table

table rowvar

table () colvar

Basic syntax for a two-way table

table rowvar colvar

Basic syntax for an n-way table

table rowvars colvar

table rowvar (colvars)

table (rowvars) (colvars)

Basic syntax for multiple n-way tables

table (rowvars) (colvars) (tabvars)

Full syntax

table (rowspec) (colspec)



(tabspec)

 

weight

 

, options



rowspec, colspec, and tabspec may be empty or may include variable names or any of the following

keywords:

keyword Description

result requested statistics

stars stars denoting statistical signiﬁcance

var variables from statistic() option

across index across() speciﬁcations

colname column names for matrix statistics

rowname row names for matrix statistics

coleq column equation names for matrix statistics

roweq row equation names for matrix statistics

command index option command()

statcmd index options statistic() and command()

table — Table of frequencies, summaries, and command results 3

options Description

Main

totals(totals) report only the speciﬁed totals

nototals suppress the marginal totals

Statistics

statistic(statspec) statistic to be reported; default is

statistic(frequency) when no weights

are speciﬁed and statistic(sumw) otherwise

Commands

command(cmdspec) collect results from the speciﬁed Stata command

Formats

nformat(% fmt



results



, basestyle



) specify numeric format

sformat(sfmt



results



) specify string format

cidelimiter(char) use character as delimiter for conﬁdence interval limits

cridelimiter(char) use character as delimiter for credible interval limits

Stars

stars(starspec) add stars to denote statistical signiﬁcance

Options

listwise use listwise deletion to handle missing values

missing treat numeric missing values like other values

showcounts show sample size for all variables in statistic()

option

zerocounts report 0 for empty cell counts

name(cname) collect results into a collection named cname

append append results to an existing collection

replace replace results of an existing collection

label(ﬁlename) specify the collection labels

style(ﬁlename



, override



) specify the collection style

markvar(newvar) create newvar that identiﬁes observations used

in the tabulation

noisily display output from each command

fweights, aweights, iweights, and pweights are allowed; see [U] 11.1.6 weight.

strL variables are not allowed; see [U] 12.4.8 strL.

markvar() and noisily do not appear in the dialog box.

4 table — Table of frequencies, summaries, and command results

Options



 

Main



totals(totals) and nototals control which totals are to be displayed in the table. By default, all

totals are reported.

totals(totals) speciﬁes which margin totals to display in the reported table. totals can contain

variables in rowspec, colspec, tabspec, and their interaction. Interactions can be speciﬁed by

using the # operator.

nototals prevents table from displaying any totals.



 

Statistics



statistic(statspec) speciﬁes the statistic to be displayed. Frequency statistics, summary statistics,

and ratio statistics are available by specifying statistic(freqstat), statistic(sumstat varlist),

and statistic(ratiostat



varlist



, ratio options



), respectively.

statistic() may be repeated to request multiple statistics.

statistic(freqstat) speciﬁes that frequencies be computed.

freqstat Deﬁnition

frequency frequency

sumw sum of weights

table — Table of frequencies, summaries, and command results 5

statistic(sumstat varlist) speciﬁes that summary statistic sumstat be computed for the variables

in varlist.

sumstat Deﬁnition

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

fvfrequency frequency of each factor-variable level

fvrawfrequency unweighted frequency of each factor-variable level

fvproportion proportion within each factor-variable level

fvrawproportion unweighted proportion within each factor-variable level

fvpercent percentage within each factor-variable level

fvrawpercent unweighted percentage within each factor-variable level

6 table — Table of frequencies, summaries, and command results

statistic(ratiostat



varlist



, ratio options



) speciﬁes that ratio statistic ratiostat be com-

puted. If varlist is speciﬁed, ratios are computed based on the totals of the speciﬁed variables.

If varlist is not speciﬁed, ratios are computed based on frequencies.

ratiostat Deﬁnition

proportion proportion

percent percentage

rawproportion proportion ignoring optionally speciﬁed weights

rawpercent percentage ignoring optionally speciﬁed weights

ratio options Deﬁnition

across(cellspec) percentages or proportions across levels of

variables or interactions

total compute overall percentages or proportions

cellspec may contain rowvars, colvars, tabvars, or an interaction between any of these variables.

Interactions can be speciﬁed by using the # operator.



 

Commands



command(cmdspec) speciﬁes the Stata commands from which to collect results. command() may be

repeated to collect results from multiple commands.

cmdspec is



explist:



command



arguments



, cmdoptions



explist speciﬁes which results to collect and report in the table. explist may include result

identiﬁers and named expressions.

result identiﬁers are results stored in r() and e() by the command. For instance, result

identiﬁers could be r(mean), r(C), or e(chi2). After estimation commands, result

identiﬁers also include the following:

Identiﬁer Result

r b coefﬁcients or transformed coefﬁcients reported by command

r se standard errors of r b

r z test statistics for r b

r z abs absolute value of r z

r p p-values for r b

r lb lower bounds of conﬁdence intervals for r b

r ub upper bounds of conﬁdence intervals for r b

r ci conﬁdence intervals for r b

r crlb lower bounds of credible intervals for r b

r crub upper bounds of credible intervals for r b

r cri credible intervals for r b

r df degrees of freedom for r b

named expressions are speciﬁed as name = exp, where name may be any valid Stata name and

exp is an expression, typically an expression that involves one or more result identiﬁers.

An example of a named expression is sd = sqrt(r(variance)).

For r-class commands, the default is to include all numeric scalars posted to r() in the table

results. For e-class commands, the default is to include r b in the table results.

table — Table of frequencies, summaries, and command results 7

command is any command that follows standard Stata syntax.

arguments may be anything so long as they do not include an if clause, in range, or weight

speciﬁcation.

Any if or in qualiﬁer and weights should be speciﬁed directly with table, not within the

command() option.

cmdoptions may be anything supported by command.



 

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 may be any statistic named in option statistic() (that is, any freqstat, sumstat, or

ratiostat) or may be any name in the e() or r() results produced by commands speciﬁed in

option command().

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 numeric layout variables (rowspec, colspec, and tabspec)

or the format of factor variables speciﬁed in the statistic() 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)" percent).

results may be any statistic named in option statistic() (that is, any freqstat, sumstat, or

ratiostat) or may be any name in the e() or r() results produced by commands speciﬁed in

option command().

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%%" percent).

This option is repeatable, and when multiple formats apply to one result, the rightmost speciﬁcation

is applied.

cidelimiter(char) changes the delimiter between conﬁdence interval limits to char. The default

is cidelimiter(" "), that is, two spaces.

cridelimiter(char) changes the delimiter between credible interval limits to char. The default is

cridelimiter(" "), that is, two spaces.



 

Stars



stars(starspec) speciﬁes that stars representing statistical signiﬁcance be included in the table.

starspec identiﬁes the result whose values determine signiﬁcance, which characters should represent

each signiﬁcance level, and where these characters should be displayed in the table. starspec is

8 table — Table of frequencies, summaries, and command results

starres



#1 "label1"



#2 "label2"



#3 "label3"



#4 "label4"



#5 "label5"





, attach(attachres) result dimension starsnoteopts



starres is the name of the result whose values determine which characters, typically which

number of stars, are to be displayed.

label1 speciﬁes the characters to be displayed when starres < #1.

label2 speciﬁes the characters to be displayed when starres < #2.

label3 speciﬁes the characters to be displayed when starres < #3.

label4 speciﬁes the characters to be displayed when starres < #4.

label5 speciﬁes the characters to be displayed when starres < #5.

attach(attachres) speciﬁes the name of the result to which the characters deﬁned by label1,

. . . , label5 are to be attached. If attach() is not speciﬁed, a new result named stars is

created and is automatically added to the table.

result and dimension control how collect stars adds items when labeling signiﬁcant

results. These options are mutually exclusive.

result speciﬁes the default behavior, and this option is necessary only if the following

dimension behavior is in effect and you want to change back to the result behavior.

dimension speciﬁes that dimension stars be added to the collection. Items will be tagged

with stars[value], and the labels will be tagged with stars[label]. Use this option

for layouts where results are to be stacked within columns, and use new dimension stars

in the column speciﬁcation of the layout.

starsnoteopts control the display and composition of the stars note.

noshownote and shownote control whether to display the stars note.

increasing and decreasing control the order of p-values in the stars note.

pvname(string) speciﬁes a name for the p-value in the stars note. The default is pvname(p).

delimiter(string) speciﬁes the delimiter between labels in the stars note. The default is

delimiter(",").

nformat(% fmt) speciﬁes the numeric format for the cutoff values in the stars note. The

default is nformat(%9.0g).

prefix(string) speciﬁes the preﬁx for the stars note. The preﬁx is empty by default.

suffix(string) speciﬁes the sufﬁx for the stars note. The sufﬁx is empty by default.

For example, stars( r p 0.01 "***" 0.05 "**" 0.1 "*", attach( r b)) could be

added to a table of regression results to specify that stars be deﬁned based on the p-values in

r p and be attached to the reported coefﬁcients ( r b).



 

Options



listwise handles missing values through listwise deletion, meaning that the entire observation is

omitted from the sample if any variable speciﬁed in a statistic() option is missing for that

observation. By default, table will omit an observation only if all variables speciﬁed in all

statistic() options are missing for that observation.

missing speciﬁes that numeric missing values of any variables speciﬁed in rowspec, colspec, or

tabspec be treated as valid categories. By default, observations with a numeric missing value in

any of these variables are omitted.

table — Table of frequencies, summaries, and command results 9

This option does not apply to factor variables speciﬁed with statistics fvfrequency, fvrawfre-

quency, fvproportion, fvrawproportion, fvpercent, or fvrawpercent.

showcounts speciﬁes that table report the sample size for each variable speciﬁed in option

statistic().

zerocounts speciﬁes that table report a 0 in empty cells for results count, frequency, fvfre-

quency, and fvrawfrequency.

name(cname) speciﬁes that a collection named cname be associated with the collected statistics and

results. The default is name(Table).

append speciﬁes that table append its collection information into the collection named in name().

replace permits table to overwrite an existing collection. This option is implied for name(Table)

when append is not speciﬁed.

label(ﬁlename) speciﬁes the ﬁlename containing the collection labels to use for your table. Labels

in ﬁlename will be loaded for the table, and any labels not speciﬁed in ﬁlename will be taken

from the labels deﬁned in c(collect label). The default is to use only the collection labels

set in c(collect label); see [TABLES] set collect label.

style(ﬁlename



, override



) speciﬁes the ﬁlename containing the collection styles to use for

your table. The default collection styles will be discarded, and only the collection styles in ﬁlename

will be applied.

If you prefer the default 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(table style); see [TABLES] set table style.

The following options are available with table but are not shown in the dialog box:

markvar(newvar) generates an indicator variable that identiﬁes the observations used in the tabulation.

noisily speciﬁes that output from the commands speciﬁed in command() options be displayed. By

default, output from commands is suppressed.

Remarks and examples stata.com

Remarks are presented under the following headings:

Introduction

Specifying the table layout

Advanced table customization

Introduction

The table command can create many customized tables, ranging from simple one-way tabulations

to multiple n-way tables with summary statistics and estimation results. table can compute and

report frequencies, proportions, percentiles, and other summary statistics. It can also run other Stata

commands and include their results in the table. This means you can combine the summary statistics

computed by table with test statistics, correlations, regression coefﬁcients, and other results collected

from Stata commands. In addition to building tables with the desired statistics, you can customize

them by formatting the values in the table and applying predeﬁned styles and labels that affect how

the row headers, column headers, and values are displayed in the table.

10 table — Table of frequencies, summaries, and command results

table can accommodate a variety of layouts. You can deﬁne the rows, columns, and even separate

tables by levels of categorical variables, statistics, or Stata commands.

If your goal is simply to create a table of estimation results or descriptive statistics, you can use the

etable and dtable commands, respectively. These commands allow you to create these tables and

export them to a variety of ﬁle types in a single step. However, unlike table, these commands create

tables with a predeﬁned layout. Therefore, if you want control over the layout or you wish to include

a combination of summary statistics, estimation results, and results from other Stata commands, you

should use the table command.

In the following entries, we provide simpliﬁed syntax, examples, and discussion for specialized

types of tables that can be created using table. If you are interested in creating one of these types

of tables, we suggest reading the corresponding entry.

[R] table oneway One-way tabulation

[R] table twoway Two-way tabulation

[R] table multiway Multiway tables

[R] table summary Table of summary statistics

[R] table hypothesis tests Table of hypothesis tests

[R] table regression Table of regression results

All the concepts demonstrated in the entries above can be combined to create tables including

combinations of tabulations, summary statistics, hypothesis tests, and regression results.

In this entry, we provide additional information on specifying the table layout and which portions

of the layout table will automate for you. In addition, we provide resources for customizing the

table and exporting the results to your preferred format.

Specifying the table layout

A table’s layout is determined by our row, column, and table dimension speciﬁcations. For example,

we specify variable names to deﬁne the rows and place statistics in the columns, or vice versa. Because

we can include so many different statistics, we can specify keywords that we use to identify the

results we have collected from commands and the statistics that table has calculated.

The syntax for specifying the table layout is

table (



rowspec



) (



colspec



) (



tabspec



)

We refer to rowspec, colspec, and tabspec collectively as the “layout”. For some tables, keywords

are required in the layout to uniquely identify the values that we want to include in our table. If you

omit a necessary keyword from the layout, table will ﬁll one in for you.

The rules determining whether a keyword is necessary to uniquely identify values in the table are

as follows:

1. If more than one statistic is speciﬁed, then result is needed in the layout.

2. If more than one variable is speciﬁed in option statistic() and option command() is not

speciﬁed, then var is needed in the layout.

3. If more than one across() speciﬁcation is used for ratio statistics, then across is needed

in the layout.

4. If option command() is speciﬁed, then colname is needed in the layout. If, in addition,

more than one variable is speciﬁed in option statistic(), then colname is needed instead

of var, which was required in 2.

table — Table of frequencies, summaries, and command results 11

5. If multiple command() options are speciﬁed and option statistic() is not speciﬁed, then

command is needed in the layout.

6. If both options command() and statistic() are speciﬁed, then statcmd is needed in the

layout.

If we do not directly specify a necessary keyword in one of rowspec, colspec, or tabspec, the

missing keywords will be automatically added to the layout as follows:

1. If the row speciﬁcation is empty, then put the missing keywords in rowspec.

2. If the row speciﬁcation is not empty but the column speciﬁcation is empty, then put the

missing keywords in colspec.

3. If the row and column speciﬁcations are not empty but the table speciﬁcation is empty and

if result is the only missing keyword and there is only one statistic (result), then put

result in tabspec.

4. Otherwise, append the missing keywords to rowvars.

Below, we demonstrate how missing keywords are added to the layout.

Using auto.dta, we create a table with the minimum and maximum mpg for each level of rep78.

The keyword result identiﬁes the statistics we computed. By listing an empty set of parentheses

followed by rep78, we request that the levels of rep78 be placed on the columns.

. use https://www.stata-press.com/data/r18/auto

(1978 automobile data)

. table () rep78, statistic(min mpg) statistic(max mpg)

Repair record 1978

1 2 3 4 5 Total

Minimum value 18 14 12 14 17 12

Maximum value 24 24 29 30 41 41

Based on rule 1, if we request more than one statistic, result must be in the layout. Based on

situation 1, if the row speciﬁcation is empty, then the missing keyword will be placed in the row

speciﬁcation. We could have created the same table by typing

. table (result) (rep78), statistic(min mpg) statistic(max mpg)

Now, let’s include multiple variables in our statistic() option. We also type rep78 immediately

after table to specify that the levels of rep78 be placed on the rows.

. table rep78, statistic(mean mpg price)

Mileage (mpg) Price

Repair record 1978

1 21 4564.5

2 19.125 5967.625

3 19.43333 6429.233

4 21.66667 6071.5

5 27.36364 5913

Total 21.28986 6146.043

Because we have more than one variable in the statistic() option, then keyword var must

be in the layout (rule 2). If we include a row speciﬁcation but leave the column speciﬁcation empty,

table will treat var as the column identiﬁer. We could have equivalently typed

12 table — Table of frequencies, summaries, and command results

. table (rep78) (var), statistic(mean mpg price)

Next, let’s include both a command() option and a statistic() option with multiple variables

in the same table. We want a table with coefﬁcients and means of the independent variables. We

use the command() option to ﬁt the regression and obtain the means with the statistic() option.

Now, we need both colname and statcmd to uniquely identify the values in the table. Let’s omit

statcmd from our command.

. table (colname) (result[_r_b mean]),

> command(regress mpg turn trunk) statistic(mean turn trunk)

Coefficient Mean

Turn circle (ft.)

regress mpg turn trunk -.7610113

Mean 39.64865

Trunk space (cu. ft.)

regress mpg turn trunk -.3161825

Mean 13.75676

Intercept

regress mpg turn trunk 55.82001

But based on situation 4, table will add statcmd to the row speciﬁcation if we leave it out. So

we could have also typed the following to create the same table:

. table (colname statcmd) (result[_r_b mean]),

command(regress mpg turn trunk) statistic(mean turn trunk)

This table displays each of the statistics that we requested. If we simply wanted to compute some

statistics quickly, it has served its purpose. However, if we wish to share these results with others or

include a table in a report, we will want to make some modiﬁcations.

Advanced table customization

table allows you to customize the results of your table using the stars(), nformat(), sformat(),

cidelimiter(), label(), and style() options. With these, you can add signiﬁcance stars, change

the numeric format, and attach characters such as percent signs or parentheses to values in the

table, use a stored set of labels, or use a predeﬁned style. See [TABLES] Predeﬁned styles for more

information on selecting a style that adjusts elements of the table such as row header alignment,

alignment of values within the cells, and which labels are included in the headers.

Customization can also go beyond the predeﬁned styles and options available to you in the table

command. table stores all of its results in a collection named Table. This means that you can use

the specialized tools available in the collect suite of commands to further customize your table.

With collect, you can modify speciﬁc labels, add borders, change the style of the headers, and the

like. Once you have a publication-ready table, you can use collect export to export your table to

HTML, Word, L

X, PDF, Excel, or another format appropriate for your report.

Stored results

table stores the following in r():

Scalars

r(N) number of observations

table — Table of frequencies, summaries, and command results 13

Methods and formulas

Variables speciﬁed in rowspec, colspec, and tabspec identify groups of observations within the

dataset. These groups are represented in the table by cells and cell margins (totals). For a given cell

or cell margin, let n denote the number of observations (frequency). Let x denote the variable

on which we want to calculate summary statistics, and let x

, i = 1, . . . , n, denote an individual

observation on x. count is the number of nonmissing values of x. first is x

and last is x

. Let

a be the smallest i such that x

is not missing, and then firstnm is x

. Let b be the largest i such

that x

is not missing, and then lastnm is x

Let v

be the weight, and if no weight is speciﬁed, deﬁne v

= 1 for all i. Let v

denote the sum

of the weights (sumw):

i=1

When aweights or pweights are speciﬁed, the normalized weights are given by w

= v

(n/v

)

with w

= n; otherwise, w

= v

and w

= v

The remaining summary statistics are computed according to the following formulas:

total











i=1

if pweights

i=1

otherwise

rawtotal

i=1

mean

x =

i=1

Deﬁne m

as the rth moment about the mean:

i=1

− x)

variance

− 1

i=1

− x)

14 table — Table of frequencies, summaries, and command results

sd (standard deviation)

s =

√

semean (standard error of the mean)

se(x) =

√

sebinomial (standard error of the mean, binomial distribution)

x(1 − x)

sepoisson (standard error of the mean, Poisson distribution)

When pweights are speciﬁed, semean, sebinomial, and sepoisson are all computed as

(x) =

n − 1

i=1



− x)



skewness

−3/2

kurtosis

−2

cv (coefﬁcient of variation)

svycv (coefﬁcient of variation, survey literature)

100

se(x)

|x|

svycv with pweights

100

(x)

|x|

table — Table of frequencies, summaries, and command results 15

geomean (geometric mean)

= exp

i=1

ln x

geosd (geometric standard deviation)

exp





− 1

i=1

( ln x

− ln x

)





Let x

(i)

refer to the x in ascending order, and let w

(i)

refer to the corresponding weights of x

(i)

minimum

(1)

maximum

(n)

range

(n)

− x

(1)

To obtain the pth percentile, which we will denote as x

[p]

, let P = np/100 and

(i)

j=1

(j)

Find the ﬁrst index i such that W

(i)

> P . The pth percentile is then

[p]







(i−1)

+ x

(i)

if W

(i−1)

= P

(i)

otherwise

q1 (ﬁrst quartile)

[25]

q2 (second quartile)

[50]

q3 (third quartile)

[75]

16 table — Table of frequencies, summaries, and command results

iqr (interquartile range)

[75]

− x

[25]

Let f be an indicator for a speciﬁc level of a factor variable and f

denote an individual observation

on f.

fvfrequency (frequency of the factor variable’s level)

i−1

fvrawfrequency (unweighted frequency of the factor variable’s level)

i−1

fvproportion (proportion of the factor variable’s level)

i−1

fvrawproportion (unweighted proportion of the factor variable’s level)

i−1

fvpercent (percentage of the factor variable’s level)

100

i−1

fvrawpercent (unweighted percentage of the factor variable’s level)

100

i−1

proportion is computed from ratios of totals. The numerator is taken from the total for the given

cell or cell margin, and the denominator is taken from the total for a cell margin that contains the

given cell or cell margin. percent is proportion multiplied by 100.

rawproportion and rawpercent are similarly computed using unweighted totals.

table — Table of frequencies, summaries, and command results 17

Also see

[R] table intro — Introduction to tables of frequencies, summaries, and command results

[R] table hypothesis tests — Table of hypothesis tests

[R] table multiway — Multiway tables

[R] table oneway — One-way tabulation

[R] table regression — Table of regression results

[R] table summary — Table of summary statistics

[R] table twoway — Two-way tabulation

[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

 1985–2023 StataCorp LLC, College Station, TX,

For suggested citations, see the FAQ on citing Stata documentation.