LEGACY CONTENT. If you are looking for Voteview.com, PLEASE CLICK HERE

This site is an archived version of Voteview.com archived from University of Georgia on May 23, 2017.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.

Homework 3, POLS 8505: MEASUREMENT THEORY
Due 2 February 2015



  1. In this problem we are going to redo the Morse Code Problem with the correct data and the correct version of smacof, namely, with the correct version of smacofSym:

    morse_code.r -- R Program to perform a Non-Metric Multidimensional Scaling of the Morse code data (set up for Lower Half)

    Morse Code Data Lower Half

    Morse Code Data Upper Half

    1. Run morse_code_L1.txt and morse_code_U1.txt through smacof (be sure to use unique output file names) in 1 to 3 dimensions. To do this change:

      ndim=2 to

      ndim=1 and

      ndim=3.

      Report the Stress values for 1 to 3 dimensions for the two halves of the data.

    2. Make two-dimensional Plots for the two halves of the data (put the 26 letters and 10 integers before the corresponding coordinates). You can modify the georgia.r program to do this. Turn in the plots and the R code. Do the plots look the same?

  2. In this problem we are going to use blackbox transpose from the basicspace package in R.

    Download the R programs:

      basicspace_big_blackbox_transpose_1968_Therms.r -- R program that performs a basicspace transpose scaling on the 12 Candidate Feeling Thermometers From the 1968 National Election Study. This version removes the 1500 respondent restriction -- Use with caution! The program reads this R program (courtesy of James Lo):
    Place the two R programs in the same directory as the first calls the second.

    1. Run basicspace_big_blackbox_transpose_1968_Therms.r and turn in the plots that it creates. Adjust the name position vector namepos so that none of the 12 Candidate names overstrike.

    2. Report summary(result) NEATLY FORMATTED and result$fits NEATLY FORMATTED.

  3. We are now going to try blackbox transpose on the NES 1968 Urban Unrest and Vietnam 7 point scales to test how unidimensional they are.

    Download the R programs:

    Place the three R programs in the same directory as the first two call the third.

    1. Run basicspace_big_blackbox_transpose_Urban_Unrest_1968.r. Report summary(result) NEATLY FORMATTED and result$fits NEATLY FORMATTED.

    2. Run basicspace_big_blackbox_transpose_Vietnam_1968.r. Report summary(result) NEATLY FORMATTED and result$fits NEATLY FORMATTED.

    3. In your opinion, are these two scales unidimensional? Explain.

  4. Work Exercise (6) on pages 100-101 of Analyzing Spatial Models of Choice and Judgment with R.