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.

Univariate Analysis 589<br />

lation, 68.26% of the statistics for estimat<strong>in</strong>g parameters will fall with<strong>in</strong> one<br />

standard error of the actual parameter; 95% of the estimates will fall between<br />

the mean and 1.96 standard errors; and 99% of the estimates will fall between<br />

the mean and 2.58 standard errors.<br />

In table 19.2, I showed you a small sample of the data from the study that<br />

Ryan, Borgatti, and I did (Bernard et al. 1999) on attitudes about environmental<br />

activism. In that study, we <strong>in</strong>terviewed a random sample of 609 adults from<br />

across the United States. The mean age of respondents <strong>in</strong> our sample was<br />

44.21, sd 15.75. Only 591 respondents agreed to tell us their age, so the standard<br />

error of the mean is:<br />

15.75 / 591 0.648<br />

S<strong>in</strong>ce we have a large sample, we can calculate the 95% confidence limits<br />

us<strong>in</strong>g the z distribution <strong>in</strong> appendix B:<br />

44.211.96(0.648) 44.211.27 42.94 and 44.211.27 45.48<br />

In other words, we expect that 95% of all samples of 591 taken from the<br />

population of adults <strong>in</strong> the United States will fall between 42.94 and 45.48.<br />

As we saw <strong>in</strong> chapter 7, these numbers are the 95% confidence limits of the<br />

mean. As it happens, we know from the U.S. Census Bureau that the real average<br />

age of the adult (over-18) population <strong>in</strong> the United States <strong>in</strong> 1997 (when<br />

the survey was done) was 44.98.<br />

Thus: (1) the sample statistic (x 44.21%) and (2) the parameter ( <br />

44.98%) both fall with<strong>in</strong> the 95% confidence limits, and we can not reject the<br />

null hypothesis that our sample comes from a population whose average age<br />

is equal to 44.98%.<br />

More about z-Scores<br />

As we saw also <strong>in</strong> chapter 7 on sampl<strong>in</strong>g, every real score <strong>in</strong> a distribution<br />

has a z-score, also called a standard score. Az-score tells you how far, <strong>in</strong><br />

standard deviations, a real score is from the mean of the distribution. The formula<br />

for f<strong>in</strong>d<strong>in</strong>g a z-score is:<br />

z <br />

(raw score x)<br />

SD<br />

Formula 19.9<br />

To f<strong>in</strong>d the z-scores of the data on FEMILLIT <strong>in</strong> table 19.7, subtract 19.49<br />

(the mean) from each raw score and divide the result by 20.81 (the sd). Table<br />

19.12 shows these z-scores.

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

Saved successfully!

Ooh no, something went wrong!