Lab: Secondary Data Analysis with the General Social Survey

In this lab we will use SPSS to analyze people's attitudes about topics related to the themes. We will be using the General Social Survey.

First, start SPSS (click on programs, SPSS for Windows). Be patient .. while you are waiting, read the rest of this. Open the GSS96 data set. We will first go through some analyses together. Then, each person will do some independent analyses.

As we are doing the analyses, keep SPSS open and do not delete results unless you have made a mistake. In that case, delete only the incorrect results. ASK HOW TO DO THIS.

Make sure to take notes during the time when the class is working together. This is how you will know the correct way to make cross tabs.

Independent Analyses

For the independent analyses you will run crosstabulations with the correct row/column set up and percentages in the correct direction with the variables that follow. You will also run one table with a control (layer) variable. You professor may give you other specific instructions.

Hypotheses:

  1. There is a relationship between sex and your dependent variable.
  2. There is a relationship between race and your dependent variable.
  3. There is a relationship between marital status and your dependent variable.
  4. The relationship between sex and your dependent variable will remain the same if race is controlled for.

Everyone will use the following independent variables: Race, Sex, Marital. For the dependent variable, select the one for your theme:

Religion: PRAYER Work and families: FEKIDS3 Schools and education: SEXEDUC
Medicine:HLTHCARE Immigration: IMMAMECO Age and aging: AGED

Circle the variable for your theme.

To test your hypotheses you will need to run 4 (or if you wish, 5) tables. Instructions for running crosstabs in SPSS are given below.

Go to the GSS web page and look up your variable on the mnemonic list (The mnemonics are in alphabetical order; you will need to select the appropriate letter of the alphabet at the top first). Make sure that you read about what the variable measures! It may not be obvious from the variable name or the response categories.

When you have your tables, you will write up your results.

First, describe your dependent variable. What does it measure?

What, if any, differences do you expect to see in responses to this question between males and females? Blacks, whites and others? Single, married, divorced or widowed people?

Then, write a few sentences about each bivariate table stating what, if any, relationship exists between the independent and dependent variable. Do your tables support the hypotheses 1-3? Were your preedictions supported? Do not list every number. Instead select a few key figures that illustrate your conclusion. You may also use the Chi Square as an aid in interpretation.

Then, look at the table with the control variable. Does the relationship between the variables get stonger, weaker or stay about the same when the control variable is introduced? Is hypothesis 4 supported? Think of at least one other control variable you think would be interesting to use.

At the end of the lab your instructor will come around and check the results on your computer. Do not print until your results have been checked. You will also write a brief summary of your independent results on a sheet of paper or in Word and hand it in at the end of lab. Make sure your name is on the paper. See below for instructions for putting your name on SPSS printouts.

If you are not working in the Carman 220 lab you can open the Gss96 data set by clicking here

Basic guidelines for crosstabs:

To run a cross tab controlling for a third variable

To put your name on the SPSS printout: