mStats (version 3.4.0)

expandtbl: Expand 2x2 table into data.frame

Description

expandtbl() generates a data.frame based on vectors.

Usage

expandtbl(
  ...,
  exp_name = "exp",
  exp_lvl = c("exposed", "unexposed"),
  case_name = "case",
  case_lvl = c("case", "control"),
  strata_name = "strata"
)
expandfreq(data, freq)

Arguments

...

vectors

exp_name

Name of exp Variable

exp_lvl

Names of two categories in the order of Exposed and non-exposed

case_name

Name of Case variable

case_lvl

names of two categories in the order of

strata_name

Name of stratified variable

data

frequency table in data.frame

freq

name of variable for the weighted frequency

Value

data.frame

Functions

• expandfreq: expandfreq() expands a frequency-weighted table into a data.frame.

Details

expandtbl

uses the vectors of 2x2 tables and generates a data frame of at least two columns: exp and case.

expandtbl(c(100, 200, 100, 200))

Strata

Multiple tables can be used to construct a dataset by specifying strata_name as follow. Strata can be included using multiple named vectors.

expandtbl(
             strata1 = c(100, 200, 100, 200),
             strata2 = c(100, 200, 100, 200),
             strata3 = c(100, 200, 100, 200),
             exp_name = "exp",
             exp_lvl = c("exposed", "unexposed"),
             case_name = "case",
             case_lvl = c("case", "control"),
             strata_name = "Strata"
)

Labels for variables

If names or levels of variables are not specified, the followings are applied.

1. exp Name: exp

2. exp levels: exposed and unexposed

3. case Name: case

4. case levels: case and control

5. Strata Name: strata

6. Note: Strata levels are not considered as vectors must be named.

expandfreq() uses the weighted frequencies in data.frame format and construct another data.frame based on the frequency weight. The name of the frequency weighted variable can be specified by freq argument.

Examples

# NOT RUN {

## Asthma Example from Essential Medical Statistics
## page 160
asthma <- expandtbl(c(81, 995, 57, 867),
              exp_name = "sex",
              exp_lvl = c("woman", "man"),
              case_name = "asthma",
              case_lvl = c("yes", "no"))

# }
# NOT RUN {
## label variable and dataset
asthma <- label(asthma, "Hypothetical Data of Asthma Prevalence")
asthma <- label(asthma, sex = "Man or Woman",
                           asthma = "Asthma or No Asthma")

## Checking codebook
codebook(asthma)


## simple tabulation
tab(asthma)

## cross-tabulation
tab(asthma, sex, by = asthma)
# }
# NOT RUN {

## Example for expanding frequency weighted data

## Example from UCLA website
## you can download the dataset here:
## https://stats.idre.ucla.edu/stat/stata/examples/icda/afterlife.dta

x <- data.frame(gender = c(1, 1, 0, 0),
                 aftlife = c(1, 0, 1, 0),
                 freq = c(435, 147, 375, 134))
y <- expandfreq(x, freq)

## check the numbers by tabulation
## tab(y, gender, by = aftlife)

# }