INSTRUCTIONS FOR DISCUSSION

1)  Test whether there is an association between a person's gender and the prestige of their occupation.  Use the GSS2008 data set to perform an independent samples t-test on SEX and PRESTG80 (or PRESTG10 depending on the data set).

Mean prestige score for men                                                           ___________________

Mean prestige score for women                                                      ___________________

t-test equality of means significance level (sig.)                                    ___________________

Is the relationship statistically significant  (sig.)                                    Yes                  No

Based on these data, would you say that gender is associated with occupational prestige?  What could explain this relationship (or lack of one)?  Find one research article that either confirms or rejects what the findings suggest.  Be sure to properly cite the article.

How did you know which t-test equality of means significance level (sig.) to use?

2) Perform a paired t-test to compare the White State Prisoner Incarceration Rate per 100,000: 2005 (CRC66) to the Black State Prisoner Incarceration Rate per 100,000: 2005 (CRC67) using the STATES10 data set.

Mean White State Prisoner Incarceration Rate per 100,000: 2005?    __________________Mean Black State Prisoner Incarceration Rate per 100,000: 2005?    __________________Significance (sig.) for the Paired Samples Test?             ___________________Is the relationship statistically significant                                      Yes                  No Are incarceration rates related to race?  What were you expecting to find and did you find it?  Use an academic research article on institutional racism to help illustrate the discrepancy in incarceration rates as well as other findings regarding differences in perceptions by race from Unit 2 (be sure to cite the article, which must be published after 2000).  Be sure to include a definition of institutional racism from the article as well as any necessary supporting arguments/evidence. 

Paired t-tests can also be used to compare means from two different time periods to see if improvement has been made in a particular area for example.

3) Perform a paired t-test to compare Low Birth Weight births per 1,000 births: 2005 (LBWRT05) to the Low Birth Weight births per 1,000 births: 2017 (LBWRT17) using the STATES10 data set.

It should look like this:

Mean Low Birth Weight births per 1,000 births: 2005?    __________________Mean Low Birth Weight births per 1,000 births: 2017?    __________________Significance (sig.) for the Paired Samples Test?             ___________________Is the relationship statistically significant                                      Yes                  No

Has there been significant change?  What kind if any?