```-------------------------------------------------------------------------------
help for unitab
-------------------------------------------------------------------------------

Univariate table

unitab depvar [varlist1] [if exp] [in range]] [, level(#) format(%fmt)
categorical(varlist2) exact]

Description

unitab displays a univariate table with the maximum-likelihood estimates of
odds ratio and confidence intervals using logit command and some useful
information using tabulate command.

depvar  binary dependent variable that must be coded as

depvar = 0    or
depvar = #    with # > 0

varlist1  covariates treated as continuous.
varlist2  covariates treated as categorical.

You can specify the same variable as continuous and/or categorical. First
results are displayed for continuous variables and then for categorical
variables.

Explanation of the table

Continuous variable

1 column : summarize depvar if continuous_variable == #

2 column : total observations

3 column : point estimate of odds ratio using maximum likelihood
estimators logit

4-5 columns : lower and upper bound for odds ratio at a certain
level(#)

6 column : statistical significance of the odds ratio using a Wald test

Categorical variable

1 column : tabulate depvar categorical_variable (display only for
depvar = # )

2 column : total observations for each category

3 column : point estimate of odds ratio using maximum likelihood
estimators

4-5 columns : lower and upper bound for odds ratio at a certain
level(#)

6 column : statistical significance of the Pearson's chi-squared test
for the hypothesis that the rows and columns in a two-way
table are independent

Options

level(#) specifies the confidence level, in percent, for calculation of con
> fidence intervals
of the odds ratios; see help level

format(%fmt) specifies the display format for odds ratio and confidence int
> ervals in the univariate table.
format(%4.3f) is the default; format(%6.5f) is a popular al
> ternative.

categorical(varlist2) specifies the variables that you want treat as catego
> rical.

exact displays the significance calculated by Fisher's exact.
We recommend specifying exact whenever samples are small.

Examples

. webuse lbw, clear
. tab low
. su age
. su age if low == 1
. logistic low age
. unitab low age
. xtile ageq = age, nq(4)
. tab ageq low, row nokey chi2
. tab ageq low, row nokey exact
. xi: logistic low i.ageq
. unitab low, c(ageq)
. unitab low, c(ageq) exact
. unitab low age, c(ageq)
. tab race low, row nokey
. xi:logistic low i.race
. unitab low , c(race)
. unitab low age ht ui, c(race smoke ageq)
. unitab low age ht ui, c(race smoke ageq) l(90) f(%6.5f)

Authors

Nicola Orsini, Institute of Environmental Medicine, Karolinska
Institutet, Stockholm, Sweden and Institute of Information Science and
Technology, National Research Council of Italy, Pisa, Italy.

Matteo Bottai, Arnold School of Public Health, University of South
Carolina, Columbia, USA and Institute of Information Science and
Technology, National Research Council of Italy, Pisa, Italy.

Support

Nicola Orsini, http://nicolaorsini.altervista.org, Karolinska
Institutet, Sweden
nicola.orsini@imm.ki.se

Also see

[R] logistic
[R] tabulate

On-line:  help for logit, logistic, table, tabulate, tabstat, tabodds, epitab
```