Summary of survey software: Epi Info (CSAMPLE procedure)

NOTE: The following information was summarized from information in the program documentation, and was not directly provided by the developer. Therefore, it may not be entirely accurate. Potential users are advised to download the documentation for further information.

Vendor

Centers for Disease Control and World Health Organization.

Types of designs that can be accommodated.

Stratified sampling, with or without clustering; multistage samples; unequal-probability (e.g. pps) samples. Does not calculate finite population corrections, so either sampling fraction must be small or sampling must be with replacement.

Types of estimands and statistical analyses that can be accommodated.

Means, proportions, odds ratios, risk ratios, risk differences.
EpiInfo also includes a wide variety of other estimation modules, not necessarily designed for survey data estimation, and there is a related mapping program, EpiMap.

Restrictions on number of variables or observations.

Primary methods used for variance estimation.

Taylor linearized deviation approach.

General description of the "feel" of the software.

The CSAMPLE program is a module of the Epi Info package, which contains a full range of facilities for entry, management, manipulation, and analysis of data, specifically oriented toward public health applications. (There are translations of Epi Info into 11 languages other than English.) Epi Info can be used with either a menu interface or a command-driven (keyword) interface.

Data can be imported from a variety of other formats.

Platforms on which the software can be run.

Most recent version (EpiInfo 2000) requires Windows (95 or more recent version). EpiInfo 6 requires any computer that can run a DOS application. Note that the CSAMPLE module for analysis of complex survey is only available in the DOS version, although it can be installed to be accessed from the Windows version. From the on-line documentation:

The [standard, outside CSAMPLE]commands assume that the data were collected using simple random (or unbiased systematic) sampling. If the data were collected using a more complex design, such as cluster sampling, then methods used to analyze the data need to take the design into account. Epi Info for DOS, available from the Epi Info web site mentioned on the title page of this manual, provides the CSample program for the analysis of data collected using complex designs, and this can be linked to the Epi Info 2000 menu if desired for use after exporting data to an Epi Info 6 file.

Pricing and terms.

The software itself and manuals are free, and available over the Internet.
Printed manuals and disks are sold by a distributors; see http://www.cdc.gov/epiinfo/EIvendor.htm

Contact information.

Additional information.

This software is discussed in the review article from The Survey Statistician.


Return to main page for survey software


DISCLAIMER