Web Resources

Internet Sources

The Internet is a rich source of tools and information on sampling and survey methods. Following are some examples of the kinds of material you can find.

Instructional videos

If you want a visual illustration of some of the methods described in this book, just Google the name of the method and search for videos. For example, here is a nice illustration of how to draw a simple random sample by assigning permanent random numbers to members of a population, and a nice illustration of how to draw a stratified sample in the same way:

You can find other videos for these subjects as well as systematic sampling, cluster sampling, etc.

For good instructional videos on statistical concepts, check out Khan Academy (https://www.khanacademy.org/). A key feature of this site is a series of brief video clips explaining basic and advanced statistical concepts, the definition and use of each,  how each is calculated, the logic of that calculation and what statistical information is produced. Video simulations further illustrate the meaning and interpretation of each concept. Here are some examples:

Sample size calculators

You can find sample size and power calculators. Here is a sample size calculator from the Australian National Statistical Service with examples. (Note: this calculator only works for estimating proportions.)

Best practices and professional standards

Professional associations sometimes issue reports on methodological issues including sampling. Here are some examples.

From AAPOR (American Association for Public Opinion Research)

From CASRO (Council of American Survey Research Organizations)

From ESOMAR (European Society for Opinion and Market Research)

AAPOR, CASRO, and ESOMAR also have codes of ethics/conduct that govern survey practices.

Bibliographies

You can use Google Scholar (http://scholar.google.com/) to search for academic literature and citations on specific methods. Same for databases to which your library may subscribe; check with your librarian.

The Web Survey Methodology website (http://www.websm.org/) provides extensive information about the methodological issues of Web surveys, as well as modern survey technologies more generally.

  • One interesting feature of this site is their listing of software for web surveys, including free software: Browse software

A number of research journals are available online. This is one directory for locating journals available at no cost.

Sample designs for well known surveys

You may find it interesting and educational to see how sampling is done for large, well-known surveys. Here are some examples:

This is just the start. The Roper Center for Public Opinion Research, Pew Research Center, U.S. Census Bureau, and other sources offer many examples of surveys with design information – and while our focus is on sampling, these sources provide examples of question wording for almost any question you might like to ask. 

Among the many resources on the Census Bureau site is information about the 

  • Population & Housing Census - every 10 years
  • Economic Census - every 5 years
  • Census of Governments - every 5 years
  • American Community Survey - annually
  • U.S. Bureau of Labor Statistics
    Among its many resources, including survey designs and data for labor and economic surveys, the BLS site is a portal with links to principal U.S. federal statistical agencies, international statistical organizations, and national statistical agencies of other countries.
  • Data.Gov
    This is the home of U.S. government data across a wide variety of sources, with data, tools, and resources to conduct secondary analyses, develop web and mobile applications, design data visualizations, and more.

Information about variance estimation for complex samples

You can Google “bootstrapping,” “bootstrapping in SAS,” etc.

Also, the following page contains a summary of available software for the analysis of surveys with complex sample designs, including software that can do variance estimation with such survey data. Information can be viewed by package or comparatively by topic.

Information about the WesVar software for variance estimation may be found here:

WesVar uses the robust and flexible approach of replication variance estimation. Replication methods are simpler conceptually; and the specification of sample design parameters for running WesVar is simpler than in some other approaches. Replication methods apply to sample designs and estimators from the simple to the most complex. Usually, a free earlier version is available. A free demo version is available for download.