This site is an archived version of Voteview.com archived from University of Georgia on May 23, 2017. This point-in-time capture includes all files publicly linked on Voteview.com at that time. We provide access to this content as a service to ensure that past users of Voteview.com have access to historical files. This content will remain online until at least January 1st, 2018. UCLA provides no warranty or guarantee of access to these files.

POLS 6386 MEASUREMENT THEORY
Eleventh (and last!) Assignment
Due 9 May 2001 no later than 12:00 Noon

1. To finish our Lieberman and McCain projects, you need to download all the Senate roll call votes taken by May 1st from:

U.S. Congress Web Site (Thomas)

Repeat the steps you took in part (3) of Homework 7 to get the roll call voting matrix.

1. Run SEN107Y.ORD through PERFL in one and two dimensions. Use SPSS to make a graph of the two dimensional results. Label Senators Lieberman, Helms, McCain, Feingold, Kennedy, Lott, Hagel, Wellstone, and Stabenow in the plot.

2. Below is a Stata file containing my copy of SEN107Y.ORD as of 17 April 2001.

Senate107.dta (Stata 7 file of SEN107Y.ORD)
Senate107a.dta (Stata 6 file of SEN107Y.ORD)

Merge the following into this Stata file:

Be certain that you use Percentages, NOT proportions throughout!

When you have everything merged into the Stata file use the outfile command in Stata to write out the following variables:

outfile statenm party senname rankord bishin bush bushdiff hisp white black using d:\statadat\data11, wide

statenm, party, and senname are variables that are already in the file. rankord is the rank ordering from PERFL, bishin is Ben Bishin's FILTER scores, bush is the Bush percentage of the popular vote in the state, bushdiff is Bush percentage minus the incumbent Senator's percentage from either the 2000, 1998, or 1996 elections, hisp is percent hispanic in the state, white is percent white in the state, and black is percent black in the state. "Outfile" tells Stata that we are writing an output file. Next we list the variables we want to write. "Using" tells Stata the output filename -- in our case its written into D:\statadat (you can use any directory). Note that there is no extension on the file name. Stata writes it as data11.raw. The "Wide" tells Stata to write the file as fixed record text with spaces between the variables.

Use Epsilon to NEATLY FORMAT data11.raw and turn it in.

3. Run probits on the Campaign Finance Reform (S.27) and Budget (H.Con.Res.83) roll calls using the following specifications:

Vote = f(bishin, bush, bushdiff, hisp, white, black)

Vote = f(rankord)

Summarize your results in two tables -- one for the Campaign Finance Reform roll calls, and one for the Budget roll calls. Write a short commentary interpreting your two tables. What are the important differences between the two models and between the two tables?

2. Merge the 107th roll calls into the combined data matrix for the 105th and 106th Senates -- SEN0506.ORD. To do this, use the program LIEBER.EXE and its control card file LIEBER.DAT that you downloaded for homework #8. To create the new SEN0507.ORD file simply change the number in LIEBER.DAT to the current number (again, as of 17 April 2001, 86 roll calls). Create SEN0507A.ORD as described in homework number 10.

1. Run SEN0507A.ORD through W-NOMINATE (WNOM9707.EXE). The control card file, NOMSTART.DAT, should look like this:
```
SEN0507A.ORD
NOMINAL MULTIDIMENSIONAL UNFOLDING OF 105TH TO 107TH SENATES
1370    1    1
2   19
15.0000  0.5000
0.0250   25
(3X,9A1,7X,4A1,2X,6A1,5X,3600I1)
(1x,I4,19A1,1X,4i4,51f7.3)
(1x,I4,19A1,1X,51f7.3)```
The "1370" is the number of roll calls as of 17 April 2001.

Report the following fit statistics for one and two dimensions: Percent Classified, Aggregate PRE, and Geometric Mean Probability.

2. Use Excel to make a histogram of the distribution of roll call vote margins. Make the histogram bars percentages.

3. Use Excel to make a graph of the eigenvalues of the double-centered squared distance matrix computed from the agreement scores from the combined matrix. These are in the output file NOM23.DAT.

4. Use SPSS to make a graph of the two dimensional results. Label Senators Lieberman, Helms, McCain, Feingold, Kennedy, Lott, Hagel, Wellstone, and Stabenow in the plot. Note that we now have two Libermans and two McCains!

3. Run PERFL in one and two dimensions on SEN0507A.ORD.

1. Report the one and two dimensional correct classification percentages.

2. In one dimension, how many rank positions did Senators Lieberman and McCain move after March 2000?

3. Assess the statistical significance of the rank changes of Senators Lieberman and McCain by performing the experiment we discussed in class. Create a copy of SEN0507A.ORD and call it SEN0507B.ORD. Write an Epsilon macro that breaks a Senator into two records exactly as we did for Lieberman and McCain. Create ten matrices with the following pairs of Senators with two records each:

Sarbanes and Specter
Dodd and Cochran
Daschle and Stevens
Hollings and Domenici
Rockefeller and Lott
Biden and Warner
Feinstein and Shelby
Kohl and Santorum
Cleland and Bennett
Breaux and Hatch

Delete the Lieberman and McCain records from each of the ten matrices and run all ten matrices through PERFL in one dimension. Compute the average and standard error of the rank changes by the ten Democrats and perform a one-tail Z-test on Lieberman's shift to the left (i.e., treat the sample mean and sample standard error as the "true" mean, m, and true standard deviation s, respectively). Is it statistically significant? Perform the same test for the Republicans and McCain. Report the means, standard errors, and the calculations of the Z-test.