Professor Story, POLS 3310
- NATIONAL ELECTION STUDIES
- You will use two studies:
- 2000 National Election Study from the ICPSR. Every one will download the 2000 National Election Study from the ICPSR (www.icpsr.org). This will be the first National Election Study that you will use.
- "Run" the POR file (runs the program and creates the SPSS data file). To do so, you will:
- Open SPSS, select Open File, for "File Type" choose SPSS Portable, find this particular file (on your jump drive or whatever), and double click on this portable file. This creates the SPSS data file. Save the SPSS data file.
- Create an Excel. To do so, you will:
- Save. Select your location. Save as Type: Excel (latest version). Type in a file Name. Save. Insert a column (farthest left). Type your last name and copy your last name down the column through all valid rows.
- Use One of the three following data sets:
- No. 1: U.S. National Election
Survey, 2004 (from Pollock)
- No. 2: U.S. National Election
Survey, 2008, ICPSR 25383
- No. 3: U.S. National Election
Survey, 2008 (from Kirkpatrick/Kidd)
- Save the SPSS data file on your drive and create an Excel file (with your last name in the first column) as above.
- Select 4 hypotheses from the 2000 National Election Study and 4 hypotheses from the one other chosen from the three) National Election Study (2004 or either of the 2008 versions). If possible, it might be interesting to use similar hypotheses to determine any possible variations over time.
- From the codebooks, select four hypotheses from all available variagles (which might be relevant, interesting, plausible, etc.). You may not use presidential vote and gender as one of your hypotheses. The 2000 study/codebook is already organized in "chapters"--making it easier to find variables. Consult either the codebooks or the "Variable View" for the other studies.
- For each of your eight hypotheses (four from each study), you will produce one cross-tab with the Chi Square statistic. Use the Chi Square statistic with the lowest value for "Asymp. Sig. (2-sided)."
- SURVEY INSTRUMENT
- Administer a survey instrument
- Select a convenience sample of 5 respondents. Ask them the following questions and record the answers.
- What is your age in years?
- What is your current annual income (do not record decimal places, but do record all integer digits)?
- What is your ethnicity? 1=white; 2=african-american; 3=hispanic; 4=other
- What is your gender? 1=male; 2=female
- What is your level of education? 1=less than high school; 2=high school; 3=college; 4=undergraduate degree; 5=beyond undergraduate degree
- What do you consider your party identification? 1=democrat; 2=republican; 3=independent; 4=other
- Do you consider yourself: 1=strongly liberal; 2=liberal; 3=moderate; 4=conservative; 5=strongly conservative?
- For whom did you vote in the 2004 presidential election: 1=bush; 2=kerry; 3=nader; 4=other; 5=did not vote?
- For whom did you vote in the 2008 presidential election: 1=mccain; 2=obama; 3=other; 5=did not vote?
- For whom did you vote in the 2012 presidential election: 1=obama; 2=romney; 3=other; 5=did not vote?
- Record the answers in an Excel file. For any refusals to answer, record all of these as 0. The rows will be each respondent and the columns will be each variable. Do not include a row with variable names (or headers). The first column will be your last name (in all five rows). The second column will be the respondent id# (simply 1, 2, 3, 4, and 5). Each subsequent column would record the info
from the variables above (keep them in exactly that order).; Thus, you should
have 12 columns and 5 rows (beginning in the top, leftmost cell).
- If you do
not have Excel software already, it is available on machines in student labs on
campus (and you can find many sources that sell software to students at very
- Save your Excel file.
and any of the data sets (use at least two different data sets) on my web
site (not the National Election Studies) to determine the following
2 hypotheses (usually micro-level studies);
2 hypotheses (usually micro-level studies);
Correlation Coefficient for 2 hypotheses (usually macro-level studies);
for 2 hypotheses (usually macro-level studies);
Correlations for 2 hypotheses (usually macro-level studies); and
Regression for 2 hypotheses (usually macro-level studies)
error-free SPSS output.
Submit in one Word file the following (in this order--for non-Word output, copy
and paste into ONE final Word file). EMAIL THE FILE TO ME:
Copy of the first page
only of your Excel files for:
The one data set from
either the 2004 Study, the ICPSR 2008 Study, or the Kirkpatrick/Kidd
instrument from the 5-person convenience sample
Summaries of all
hypotheses--and conclusions. For the bivariate/multivariate
statistics, summarize which data sets you are using.
SPSS output (no errors)
for all of the above statistics. Highlight the key statistics on your
SPSS output. Feel free to delete unnecessary tables, etc. in your
Be sure to organize your