Data Analysis on Political Parties:

Two Exercises using SPSS to analyze a 2010 Pew Research Center study

The survey used in these exercises is from the Pew Research Center, conducted August 25-September 6, 2010 via telephone interviews with a robust national adult sample of 3,509. The dataset is available as a free download. The codebook—an annotated questionnaire that will help you better understand the methodology used to collect these data— is also available. Below are two exercises that will initialize the thought processes in using a public opinion dataset and are intended for introductory instruction.

[BEFORE YOU BEGIN USING SPSS: Remember to properly weight the data analysis; the variable is “WEIGHT”. If you are using the SPSS wizard, choose the menu item “Data” at the top of the Data Editor, select the option “weight cases” at the bottom of the menu list. Enter the variable “WEIGHT” where requested.]

Topic: US Political Parties

Related reading:

Ladd, Everett C. (1989). The Big Shift in Party Strength in the ‘80s. Public Perspective November/December, 1989. p.11-12.


  1. The two major political parties in the US have long established their identities as particularly strong or weak in handling an array of issues the country faces.
    1. Using this dataset, identify the variables that ask Americans about the strengths of each party. [Hint: These questions typically ask specifically, “…which political party handles…<issue> better.”]
    2. What is your hypothesis as to how this sample of Americans answered these questions?
    3. Now using SPSS, run the dataset to find out how people responded—how close were you in your hypothesis? Also, run the party identification question to see what percentage of the sample call themselves Republicans, Democrats, Independents.

[Hint: Go to header ANALYZE -> Descriptive Statistics -> …]

  1. Now, run the crosstabulations that tell you how Democrats and Republicans and Independents answered the same questions. What do you find?
  2. Running other groups, what can you say about how various groups view each of the parties? Other groups might include younger vs. older; various educational levels; men and women; what about the different regions of the country? [Hint: use the recoded age variable for ease of interpretation.]
  3. How do your results compare to the findings expressed in the 25 year old article from 1989?


  1. Recently there has been much talk about the polarization of the parties—this means that the two political parties are very far apart on important issues. This has been demonstrated numerous times recently by congressmen/women of the two key parties in the US Congress.
    1. Are the Americans interviewed in this poll also polarized?
    2. Select some social and economic questions from this dataset and investigate if views of Republicans and Democrats diverge or converge.
    3. Prepare an essay that summarizes your findings. Are there patterns where divergence is predictable?
    4. An advanced investigation: Independents—do they look more like Democrats or Republicans?