27.10.2014 Views

Russel-Research-Method-in-Anthropology

Russel-Research-Method-in-Anthropology

Russel-Research-Method-in-Anthropology

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Sampl<strong>in</strong>g 167<br />

25<br />

Number of Households<br />

20<br />

15<br />

10<br />

5<br />

0<br />

0 5 10 15<br />

Children per Household<br />

Figure 6.2b. Distribution of the number of children <strong>in</strong> Ticuna households.<br />

SOURCE: Adapted from data <strong>in</strong> A. Oyuela-Caycedo and J. J. Vieco Albarracín, ‘‘Aproximación cuantitativa a<br />

la organización social de los ticuna del trapecio amazónico colombiano,’’ Revista Colombiana de Antropología,<br />

Vol. 35, pp. 146–79, table 6, p. 159, 1999.<br />

many population subgroups (that is, <strong>in</strong>dependent variables) you want to deal<br />

with simultaneously <strong>in</strong> your analysis; (3) the size of the subgroup that you’re<br />

try<strong>in</strong>g to detect; and (4) how precise you want your sample statistics (or<br />

parameter estimators) to be.<br />

1. Heterogeneity of the population. When all elements of a population have the<br />

same score on some measure, a sample of 1 will do. Ask a lot of people to tell<br />

you how many days there are <strong>in</strong> a week and you’ll soon understand that a big<br />

sample isn’t go<strong>in</strong>g to uncover a lot of heterogeneity. But if you want to know<br />

what the average ideal family size is, you may need to cover a lot of social<br />

ground. People of different ethnicities, religions, <strong>in</strong>comes, genders, and ages may<br />

have very different ideas about this. (In fact, these <strong>in</strong>dependent variables may<br />

<strong>in</strong>teract <strong>in</strong> complex ways. Multivariate analysis tells you about this <strong>in</strong>teraction.<br />

We’ll get to this <strong>in</strong> chapter 21.)<br />

2. The number of subgroups <strong>in</strong> the analysis. Remember the factorial design<br />

problem <strong>in</strong> chapter 5 on experiments? We had three <strong>in</strong>dependent variables, each<br />

with two attributes, so we needed eight groups (2 3 8). It wouldn’t do you much<br />

good to have, say, one experimental subject <strong>in</strong> each of those eight groups. If<br />

you’re go<strong>in</strong>g to analyze all eight of the conditions <strong>in</strong> the experiment, you’ve got<br />

to fill each of the conditions with some reasonable number of subjects. If you

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

Saved successfully!

Ooh no, something went wrong!