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.

582 Chapter 19<br />

Notice that we need to square the difference of each observation from the<br />

mean and then take the square root later. As we saw <strong>in</strong> calculat<strong>in</strong>g the mean,<br />

xx0. That is, the simple sum of all the deviations from the mean is<br />

zero. Squar<strong>in</strong>g each difference (each x x) gets rid of the negative numbers.<br />

Variance describes <strong>in</strong> a s<strong>in</strong>gle statistic how homogeneous or heterogeneous<br />

a set of data is, and, by extension, how similar or different the units of analysis<br />

described by those data are. Consider the set of scores <strong>in</strong> table 19.2 on the<br />

variable called REDUCE. These scores show people’s support for the idea<br />

that ‘‘Americans are go<strong>in</strong>g to have to drastically reduce their consumption<br />

over the next few years.’’ Suppose that for each level of education, you could<br />

predict the level of support for that attitud<strong>in</strong>al item about cutt<strong>in</strong>g back on consumption.<br />

If you could do this <strong>in</strong> 100% of all cases, then you would speak of<br />

‘‘expla<strong>in</strong><strong>in</strong>g all the variance’’ <strong>in</strong> the dependent variable.<br />

I’ve never encountered this strength of association between two variables <strong>in</strong><br />

the social sciences, but some th<strong>in</strong>gs come pretty close, and <strong>in</strong> any event, the<br />

pr<strong>in</strong>ciple is what’s important.<br />

The standard deviation, s, is the square root of the variance, s 2 . The formula<br />

for the standard deviation is:<br />

s 2<br />

xx<br />

n1<br />

Formula 19.5<br />

Table 19.10 shows how to calculate the standard deviation for the data on<br />

female illiteracy <strong>in</strong> table 19.7. The sum of the (x x) 2 is 3,896.11 and n <br />

10.<br />

Substitut<strong>in</strong>g <strong>in</strong> the formula for standard deviation, we get:<br />

s <br />

3,896.11<br />

9<br />

20.81<br />

If we were report<strong>in</strong>g these data, we would say that ‘‘the average percentage of<br />

adult female illiteracy is 19.49 with sd 20.81.’’<br />

For grouped data, we take the midpo<strong>in</strong>t of each <strong>in</strong>terval as the raw score<br />

and use formula 19.6 to calculate the standard deviation:<br />

s 2<br />

fxx<br />

n1<br />

Formula 19.6<br />

Table 19.11 shows the procedure for calculat<strong>in</strong>g the standard deviation for the<br />

grouped data <strong>in</strong> table 19.4. We know from table 19.6 that x 45 for the data<br />

<strong>in</strong> table 19.4.

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

Saved successfully!

Ooh no, something went wrong!