27.10.2014 Views

Russel-Research-Method-in-Anthropology

Russel-Research-Method-in-Anthropology

Russel-Research-Method-in-Anthropology

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

Sampl<strong>in</strong>g 149<br />

the potential sampl<strong>in</strong>g error <strong>in</strong> your study. (I’ll discuss how you know this<br />

later on <strong>in</strong> this chapter.) When historical circumstances (the surprise court<br />

judgment, for example) require it, you could <strong>in</strong>terview the same sample of 60<br />

aga<strong>in</strong> (<strong>in</strong> what is known as a panel study), or take another representative sample<br />

of the same size and see what differences there are before and after the<br />

critical event.<br />

In either case, you are better off with the sample than with the whole population.<br />

By the way, there is no guarantee that a week is quick enough to avoid<br />

the problem described here. It’s just less likely to be a problem.<br />

Sampl<strong>in</strong>g Frames<br />

If you can get it, the first th<strong>in</strong>g you need for a good sample is a good sampl<strong>in</strong>g<br />

frame. (I say, ‘‘if you can get it,’’ because a lot of social research is<br />

done on populations for which no sampl<strong>in</strong>g frame exists. More on this at the<br />

end of this chapter.) A sampl<strong>in</strong>g frame is a list of units of analysis from which<br />

you take a sample and to which you generalize.<br />

A sampl<strong>in</strong>g frame may be a telephone directory, or the tax rolls of a community,<br />

or a census of a community that you do yourself. In the United States,<br />

the city directories (published by R. L. Polk and Company) are often adequate<br />

sampl<strong>in</strong>g frames. The directories are available for many small towns at the<br />

local library or Chamber of Commerce. Professional survey researchers <strong>in</strong> the<br />

United States often purchase samples from firms that keep up-to-date databases<br />

just for this purpose.<br />

For many projects, though, especially projects that <strong>in</strong>volve field research,<br />

you have to get your own census of the population you are study<strong>in</strong>g. Whether<br />

you work <strong>in</strong> a village or a hospital, a census gives you a sampl<strong>in</strong>g frame from<br />

which to take many samples dur<strong>in</strong>g a research project. It also gives you a basis<br />

for comparison if you go back to the same community later.<br />

Simple Random Samples<br />

To get a simple random sample of 200 out of 640 people <strong>in</strong> a village, you<br />

number each <strong>in</strong>dividual from 1 to 640 and then take a random grab of 200 out<br />

of the numbers from 1 to 640. The easiest way to take random samples is with<br />

a computer. All of the popular program packages for statistical analysis have<br />

built-<strong>in</strong> random-number generators. Some of the most popular <strong>in</strong>clude SAS,<br />

SPSS, SYSTAT, KWIKSTAT, STATA, and STATMOST. (Internet<br />

addresses for all these programs are given <strong>in</strong> appendix F.) You can also take a

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!