summary2: Summarize edsurvey.data.frame Variables

In EdSurvey: Analysis of NCES Education Survey and Assessment Data

Summarize edsurvey.data.frame Variables

Description

Summarizes edsurvey.data.frame variables.

Usage

summary2(
  data,
  variable,
  weightVar = attr(getAttributes(data, "weights"), "default"),
  dropOmittedLevels = FALSE,
  omittedLevels = deprecated()
)

Arguments

data	an edsurvey.data.frame, an edsurvey.data.frame.list, or light.edsurvey.data.frame
variable	character vector of variable names
weightVar	character weight variable name. Default is the default weight of data if it exists. If the given survey data do not have a default weight, the function will produce unweighted statistics instead. Can be set to NULL to return unweighted statistics.
dropOmittedLevels	a logical value. When set to TRUE, drops those levels of the specified variable. Use print on an edsurvey.data.frame to see the omitted levels. Defaults to FALSE.
omittedLevels	this argument is deprecated. Use dropOmittedLevels

Value

summary of weighted or unweighted statistics of a given variable in an edsurvey.data.frame

For categorical variables, the summary results are a crosstab of all variables and include the following:

[variable name]	level of the variable in the column name that the row regards. There is one column per element of variable.
N	number of cases for each category. Weighted N also is produced if users choose to produce weighted statistics.
Percent	percentage of each category. Weighted percent also is produced if users choose to produce weighted statistics.
SE	standard error of the percentage statistics

For continuous variables, the summary results are by variable and include the following:

Variable	name of the variable the row regards
N	total number of cases (both valid and invalid cases)
Min.	smallest value of the variable
1st Qu.	first quantile of the variable
Median	median value of the variable
Mean	mean of the variable
3rd Qu.	third quantile of the variable
Max.	largest value of the variable
SD	standard deviation or weighted standard deviation
NA's	number of NA in variable and in weight variables
Zero weights	number of zero weight cases if users choose to produce weighted statistics

If the weight option is chosen, the function produces weighted percentile and standard deviation. Refer to the vignette titled Statistical Methods Used in EdSurvey and the vignette titled Methods Used for Estimating Percentiles in EdSurvey for how the function calculates these statistics (with and without plausible values).

Author(s)

Paul Bailey and Trang Nguyen

See Also

percentile

Examples

## Not run: 
# read in the example data (generated, not real student data)
sdf <- readNAEP(path=system.file("extdata/data", "M36NT2PM.dat", package = "NAEPprimer"))

# print out summary of weighted statistics of a continuous variable
summary2(data=sdf, variable="composite")
# print out summary of weighted statistics of a variable, including omitted levels
summary2(data=sdf, variable="b017451", omittedLevels = FALSE)
# make a crosstab
summary2(data=sdf, variable=c("b017451", "dsex"), omittedLevels = FALSE)

# print out summary of unweighted statistics of a variable
summary2(data=sdf, variable="composite", weightVar = NULL)

## End(Not run)

EdSurvey documentation built on June 27, 2024, 5:10 p.m.