POLI 279 MEASUREMENT THEORY
Sixth Assignment
Due 2 June 2006


  1. In this problem we are going to use R and KYST to replicate the analysis done by Herbert F. Weisberg and Jerrold G. Rusk in their classic 1970 paper "Dimensions of Candidate Evaluation," American Political Science Review, 64 (Dec. 1970): 1167-1185. Download the following R program and data file:

    Weisberg_and_Rusk_Shepard_Plot.r -- R Program to Replicate Weisberg and Rusk
    Here is what Weisberg_and_Rusk_Shepard_Plot.r looks like:
    
    #
    #
    #
    #  Weisberg_and_Rusk.r -- Replicates Famous 1970 W&R paper in APSR
    #
    #   ***See Homework #5 2003 Houston MEASUREMENT THEORY Course***
    #             http://voteview.org/measure.htm
    #
    # wallace        wallace therm
    # humphrey       humphrey thermometer
    # nixon          nixon thermometer
    # mccarthy       mccarthy thermometer
    # reagan         reagan thermometer
    # rockefeller    rockefeller thermometer
    # lbj            lbj thermometer
    # romney         romney thermometer
    # kennedy        robert kennedy thermometer
    # muskie         muskie thermometer
    # agnew          agnew thermometer
    # lemay          "bombs away with Curtis LeMay" thermometer
    #
    #  Read Just The Thermometer Data From 1968 Survey
    #  
    #  Remove all objects just to be safe
    #
    rm(list=ls(all=TRUE))
    library(MASS)
    library(stats)
    #
    T <- matrix(scan("C:/ucsd_homework_6/therm68.txt",0),ncol=12,byrow=TRUE)
    #
    nrow <- length(T[,1])
    ncol <- length(T[1,])
    #
    #
    #  Labels For Candidates
    #
    junk <- NULL
    junk[1] <- "Wallace"
    junk[2] <- "Humphrey"
    junk[3] <- "Nixon"
    junk[4] <- "McCarthy"
    junk[5] <- "Reagan"
    junk[6] <- "Rockefeller"
    junk[7] <- "LBJ"
    junk[8] <- "Romney"
    junk[9] <- "Kennedy"
    junk[10] <- "Muskie"
    junk[11] <- "Agnew"
    junk[12] <- "LeMay"
    #
    # The range of the 1968 Feeling Thermometers was 0 to 97 -- 98 and 99 
    # were used as missing values. You need to tell R that 98 and 99 are 
    # missing. To do this, use the following command:
    #
    # If T[i,j] = 98 or 99 set TT[i,j] = NA (missing data)
    # else if set TT[i,j] = T[i,j]
    #
    TT <- ifelse(T==98 | T==99,NA,T)  
    #                                  
    # To compute the correlation matrix use the command below:
    # The "pairwise.complete.obs" tells R to throw away missing 
    #  data pair-wise, not list-wise (that is, the whole row of data!).
    #
    R <- cor(TT,use="pairwise.complete.obs")
    RR <- R
    #
    #
    # Transform the Correlation Matrix to Distance Matrix
    #
    i <- 0
    while (i < ncol) {
      i <- i + 1
      j <- 0
      while (j < ncol) {
         j <- j + 1
    #
    #  This is the Normal Transformation
    #
         RR[i,j] <- (1 - R[i,j])
    #
    #
      }
    }
    #
    #  Call Classical Kruskal-Young-Shepard-Torgerson Non-Metric 
    #           Multidimensional Scaling Program
    #
    # RR -- Input
    # dim=2 -- number of dimensions
    # y -- starting configuration (generated internally)
    #
    #  Scale in two dimensions
    #
    kdim <- 2
    y <- rep(0,kdim*ncol)
    dim(y) <- c(ncol,kdim)
    #
    #  Call Kruskal NonMetric MDS
    #
    wandrnmds <- isoMDS(RR,y=cmdscale(RR,kdim),kdim, maxit=50)
    #                                                    as.dist transforms a square symmetric matrix of distances 
    RR.dist <- as.dist(RR,diag=FALSE,upper=FALSE)          into a distance object.  This is necessary for the Shepard(...) routine
    shepardcheck.sh <- Shepard(RR.dist,wandrnmds$points)           Shepard is an undocumented function in R
    plot(shepardcheck.sh, type="n",asp=1,main="",xlab="",ylab="",    I figured this out by trial and error!!
             xlim=c(0,2),ylim=c(0,2),font=2)                         I copied the syntax from the bottom of the isoMDS page!!
    points(shepardcheck.sh$x, shepardcheck.sh$y,pch=16,col="red")    The key is to use the lower left triangle distance matrix from 
    #                                                                   the as.dist function.
    #  The "S" means "stair-steps"
    #
    lines(shepardcheck.sh$x, shepardcheck.sh$yf,lty=1,lwd=2,type = "S",font=2)
    #
    # Main title
    mtext("Shepard Diagram for MDS Solution\nWeisberg and Rusk",side=3,line=1.00,cex=1.2,font=2)
    # x-axis title
    mtext("Observed Dissimilarities",side=1,line=2.75,font=2,cex=1.2)
    # y-axis title
    mtext("Estimated Dissimilarities (D-hats)",side=2,line=2.75,font=2,cex=1.2)
    #
    #
    #  This creates another Graphics Device in R
    #
    windows()
    xmax <- c(wandrnmds$points[,1],wandrnmds$points[,2])
    axismax <- max(abs(xmax))
    #
    plot(wandrnmds$points[,1],wandrnmds$points[,2],type="n",asp=1,
         main="",
         xlab="",ylab="",
         xlim=c(-axismax,axismax),ylim=c(-axismax,axismax),font=2)
    #
    # Main title
    mtext("The 1968 Candidate Configuration\nFrom MDS Program in R",side=3,line=1.00,cex=1.2,font=2)
    # x-axis title
    mtext("Liberal - Conservative",side=1,line=2.75,font=2,cex=1.2)
    # y-axis title
    mtext("Anti-Wallace   Wallace",side=2,line=2.75,font=2,cex=1.2)
    #
    #
    # pos -- a position specifier for the text. Values of 1, 2, 3 and 4, 
    # respectively indicate positions below, to the left of, above and 
    # to the right of the specified coordinates 
    #
    namepos <- NULL
    namepos[1] <- 2   # Wallace
    namepos[2] <- 3   # Humphrey
    namepos[3] <- 4   # Nixon
    namepos[4] <- 2   # McCarthy
    namepos[5] <- 4   # Reagan
    namepos[6] <- 2   # Rockefeller
    namepos[7] <- 3   # LBJ   
    namepos[8] <- 2   # Romney
    namepos[9] <- 4   # Kennedy
    namepos[10] <- 2  # Muskie
    namepos[11] <- 4  # Agnew
    namepos[12] <- 2  # LeMay
    #
    text(wandrnmds$points[,1],wandrnmds$points[,2],junk,pos=namepos,offset=0.5,col="blue",font=2)
    points(wandrnmds$points[,1],wandrnmds$points[,2],pch=16,col="red",font=2)
    text(-0.7,1.0,paste("Stress = ",  
                    0.01*round(wandrnmds$stress, 2)),col="purple",font=2)
    #
    #  This creates another Graphics Device in R
    #
    windows()
    #
    # Create Data For Gradient of Generalization Plots
    #
    #
    y <- rep(0,((ncol*(ncol-1))/2)*2)
    dim(y) <- c(((ncol*(ncol-1))/2),2)
    #
    i <- 1
    kk <- 0
    while (i <= ncol) {
      j <- i
      while (j <= ncol) {
         k <- 0
         dist <- 0.0
         while (k < kdim) {
           k <- k+1
           dist <- dist + (wandrnmds$points[i,k]-wandrnmds$points[j,k])^2
         }
         if(i != j) {
         kk <- kk +1
         y[kk,1] <- dist
         y[kk,2] <- R[i,j]
    #     y[kk,2] <- RR[i,j]
         }
      j <- j + 1
      }
      i <- i + 1
    }
    #
    #
    ylow <- min(y[,2])
    yhigh <- max(y[,2])
    plot(y[,1],y[,2],ylim=c(ylow,yhigh),
         xlab="",ylab="",col="red",font=2)
    lines(lowess(y[,1],y[,2],f=.2),lwd=3)
    text(10,80,"Line estimated \nUsing Lowess")
    #
    # Main title
    mtext("Shepard's Theory of Generalization\nWeisberg and Rusk Data",side=3,line=1.00,cex=1.2,font=2)
    # x-axis title
    mtext("Psychological Distance",side=1,line=2.75,font=2,cex=1.2)
    # y-axis title
    mtext("Observed Similarity",side=2,line=2.75,font=2,cex=1.2)
    #
    #
    #  Save The Shepard Graph Data Sorted Ascending
    #
    kp <- order(y[,2])
    shepard <- cbind(y[kp,1],y[kp,2])
    #
      Run Weisberg_and_Rusk_Shepard_Plot.r and you will get the following three graphs:





    1. Turn in these three graphs.

    2. We are now going to compare our results from R with those from KYST. To do this, we need to get the Pearson Correlation matrix stored in the matrix R out into a file that we can then submit to KYST. At the command line in R type:

      write.matrix(R,"c:/ucsd_homework_6/R-table2.txt") -- write.matrix writes a matrix to a file nicely formatted

      Bring the file up in Epsilon. It should look like this:



      Type in the names of the 12 candidates so that your file looks like this:



      Now, use Epsilon to stack the following commands on top of the correlation matrix:
      
      TORSCA
      PRE-ITERATIONS=3
      DIMMAX=3,DIMMIN=1
      PRINT HISTORY,PRINT DISTANCES                                                
      COORDINATES=ROTATE
      ITERATIONS=50
      REGRESSION=DESCENDING
      DATA,LOWERHALFMATRIX,DIAGONAL=PRESENT,CUTOFF=-2.00
      1968 FEELING THERMOMETER CORRELATION MATRIX
       12  1  1
      (12X,12F12.8)
      and do not forget to put COMPUTE and STOP on the bottom!

      Report the Stress Values for 1, 2, and 3 dimensions and use R to graph the results in two dimensions.

    3. Use R to produce a Shepard Diagram for the two dimensional solution in the same format as Homework 2 question 1.

    4. Make a graph of the eigenvalues of the correlation matrix like the one you did for Homework 5 question 2.e. Note that you can get the eigenvalues by using the eigen(..) command in R.

  2. The aim of this problem is to familarize you with the Aldrich-McKelvey scaling program. Download the program and the 1968 data and "control card" files:

    MCKALNEW.EXE -- Aldrich-McKelvey scaling program
    and place them in the same folder on a WINTEL machine. Read the Aldrich-McKelvey scaling program page to see how to run the program.

    Here are the first three lines of OLS68A.DAT.
    
     1681  0 10  1  1  1  1  1 63  4  4  5  7  1  2  2  3  7  1  1
     1124  0 10  1  0  0  0  1 82  1  1  4  4  1  1  1  1  4  1  5
       78  5 10  1  0  1  1  1 78  2  1  5  7  4  5  5  6  6  7  5
    
    The variables, in order, are:
    
        RESPONDENT ID      = unique 4 digit number
        PARTY ID           = 0 to 6 -- 0 = Strong Democrat
                                       1 = Weak Democrat
                                       2 = Lean Democrat
                                       3 = Independent
                                       4 = Lean Republican
                                       5 = Weak Republican
                                       6 = Strong Republican
        RAW INCOME         = **do not use**
        FAMILY INCOME      = income quintile 1 - 5
        SEX                = 0 Man, 1 Woman
        RACE               = 0 White, 1 Black
        SOUTH              = 0 North, 1 South
        EDUCATION          = 1 High School or less, 2 Some College, 3 College
        AGE                = In Years
        URBAN UNREST SCALE = Johnson, Humphrey, Nixon, Wallace, Self-Placement
        VIETNAM SCALE      = Johnson, Humphrey, Nixon, Wallace, Self-Placement
        VOTED              = 1 Voted, 5 Did Not Vote
    
    Run the Aldrich-McKelvey scaling procedure using both the urban unrest and vietnam files. In particular, for the urban unrest scale here are the commands:

    MCKALNEW
    Control Card File Name? URBAN68.CTL
    Data File Name? OLS68A.DAT
    Output File Name? URBAN.PRN
    Coordinate Output File Name? URBAN.DAT


    The program reads URBAN68.CTL and OLS68A.DAT and writes the output files to disk. The coordinates for the political stimuli are in URBAN.PRN. In particular, here is the relevant portion of URBAN.PRN:
        etc etc etc
     ******************************************************************************
     PERFORMANCE INDEX=    0
     EIGENVALUES
      -1070.0801
       -936.8338
       -251.0831
          0.0012
     ******************************************************************************
     ******************************************************************************
     STIMULUS COORDINATES
        LBJ      HHH      NIXON    WALLA
      -0.3978  -0.4262   0.0116   0.8124  These are the coordinates
     STIMULUS COORDINATES RAW DATA
      -0.4087  -0.4229   0.0232   0.8084
     ******************************************************************************
     CORRECTED GOODNESS OF FIT AND RAW FIT
          0.0919      0.5075
             etc etc etc
    1. Graph the scaled stimuli coordinates against each other using R -- that is, use the Vietnam coordinates as the horizontal axis and the Urban Unrest coordinates as the vertical axis. Use the names of the candidates and label the graph and the axes appropriately. Why do you think the plot looks the way it does?

    2. Use Epsilon to merge the party ID variable from OLS68A.DAT into the coordinate output file. For example, here are the first few lines of the coordinate output file for the urban unrest scale:
      
       LINE #  CASE # R POS   ALPHA    BETA    SCALED POS     RSQ
            1   1681   1.0   -2.6243    0.5249   -2.0994    0.9994    0.9997
            2   1124   1.0   -0.8831    0.3532   -0.5298    0.6790    0.8240
            3     78   4.0   -0.9588    0.2557    0.0639    0.8992    0.9483
            4    553   4.0   -1.2302    0.2895   -0.0724    0.6460    0.8037
      The second column is the respondent ID number. Use the respondent ID number to match OLS68A.DAT with the output file and insert the party ID code into the output file. After you have inserted the party code you can delete all the columns except the party code, BETA (you will need that for graphing), and the Scaled Position. If you have done everything correctly the first few lines of your file should look like this:
      
        0     0.5249   -2.0994
        0     0.3532   -0.5298
        5     0.2557    0.0639
        1     0.2895   -0.0724
        1     0.2763    0.6907
        1     1.3930   -0.3482
        0     1.0597   -0.5298
        0     0.2322   -0.3482
        0     0.4371   -0.8742
        5     0.3901    0.6827
        0     0.0033    0.0050
        0     0.2119   -0.5298
        1     0.2624    0.2624
        2     0.2938   -0.5142
        1     0.2745    0.1373
             etc.
             etc.
             etc.
      Write an Epsilon text macro that inserts the party variable into the coordinate file. In the macro, use a split screen and place the coordinate file in the top screen and OLS68A.DAT in the bottom screen. When you begin it should look like this:



      The beginning of the text macro should look like this:
      
      (define-macro "hw6" "C-U20C-F
      C-U5C-BC-KC-YC-DC-AC-XC-Y
      C-U3C-F
      C-U4C-BC-KC-YC-DC-AC-XC-Y")
      Open up another window and put the above macro fragment in it and you should be here:



      Finish constructing the macro and turn in the listing.

    3. Read the above file into R and make smoothed histograms of the scaled positions of the respondents by party ID. For example, to make smoothed histograms of the Strong Democrats and Strong Republicans use this R Program:

      Smoothed_Histogram_hw_6_2006.r -- R Program to Plot Strong Democrats and Strong Republicans on Urban Unrest Scale

      Here is what Smoothed_Histogram_hw_6_2006.r looks like:
      #
      #
      #  Smoothed_Histogram_hw_6_2006.r -- Plots Strong Democrats and Strong Republicans on 
      #                            1968 Urban Unrest Scale
      #
      #
      rm(list=ls(all=TRUE))
      #
      #
      library(MASS)
      #
      T <- matrix(scan("C:/UCSD_Homework_6/urban_hw6.txt",0),ncol=3,byrow=TRUE)
      #
      #  Gore and Bush Voters
      #
      strong.democrat <- T[T[,1]==0 & T[,2] > 0,3]    Select Strong Democrats With Positive Betas
      strong.republican <- T[T[,1]==6 & T[,2] > 0,3]  Select Strong Republicans With Positive Betas
      #
      DemShare <- length(strong.democrat)/(length(strong.democrat)+length(strong.republican))   These two commands just compute
      RepShare <- length(strong.republican)/(length(strong.democrat)+length(strong.republican))  the proportions for the two Parties
      #
      demdens <- density(strong.democrat)   density computes kernel density estimates. (Also see bandwidth.)
      demdens$y <- demdens$y*DemShare       This is a trick so that the two densities....
      #
      repdens   <- density(strong.republican)
      repdens$y <- repdens$y*RepShare       ...will add to 1.0
      #
      ymax1 <- max(demdens$y)
      ymax2 <- max(repdens$y)
      ymax <- 1.1*max(ymax1,ymax2)
      #
      plot(demdens,main="",
             xlab="",
             ylab="",
             xlim=c(-1.5,1.5),ylim=c(0,ymax),font=2)
      lines(demdens,lwd=3,col="red")
      lines(repdens,lwd=3,col="blue")
      #
      text( .50,0.800,"Red = Strong Democrats",col="red",font=2,cex=1.2)
      text( .50,0.725,"Blue = Strong Republicans",col="blue",font=2,cex=1.2)
      # Main title
      mtext("Strong Party Identifiers\nFrom 1968 Urban Unrest 7-Point Scale",side=3,line=1.50,cex=1.2,font=2)
      # x-axis title
      mtext("Urban Unrest Scale Value",side=1,line=2.75,cex=1.2)
      # y-axis title
      mtext("Density",side=2,line=2.5,cex=1.2)
      #
      arrows(-.398, 0.06,-.398,0.0,length=0.1,lwd=3,col="red")
      text(-.308,.08,"LBJ",font=2)
      arrows(-.426, 0.06,-.426,0.0,length=0.1,lwd=3,col="red")
      text(-.516,.08,"HHH",font=2)
      arrows( .012, 0.13, .012,0.0,length=0.1,lwd=3,col="blue")
      text( .000,.16,"Nixon",font=2)
      arrows( .812, 0.13, .812,0.0,length=0.1,lwd=3,col="green")
      text( .812,.18,"Wallace",font=2)
      #
      #    LBJ      HHH      NIXON    WALLA
      #  -0.3978  -0.4262   0.0116   0.8124
      Here is the graph it produces:



      Turn in this plot.

    4. Similar to the above, make a graph for all Democrats (0,1,2) and all Republicans (4,5,6). Adjust the labeling accordingly and make certain that everything is neatly presented.

    5. Similar to the above, make a graph for all Democrats (0,1,2) and all Republicans (4,5,6) for the Vietnam scale. Adjust the labeling accordingly and make certain that everything is neatly presented.

    6. Similar to the above, make a graph for all Democrats (0,1,2), all Republicans (4,5,6), and Independents (4) for the Vietnam scale. Adjust the labeling accordingly and make certain that everything is neatly presented. To do this note that you will have to add code to compute IndShare and adjust everything so that the three smoothed histograms add up to 1.

  3. The aim of this problem is to show you how to use metric unfolding to analyze thermometer scores. To do this you need to run a program that unfolds the thermometer scores. We are going to analyze the 1968 feeling thermometers. Download the the program, control card file, and data file and place them in the same directory.

    MLSMU6.EXE -- Metric Unfolding Program
    The 1968 Election Data file contains the same variables that we used above plus the voting information for the respondents. The variables are:
    
    idno           respondent id number
    partyid        strength of party id -- 0 to 6
    income         raw income category
    incomeq        income quintile -- 1 to 5
    race           0 = white, 1 = black
    sex            0 = man, 1 = woman
    south          0 = north, 1 = south
    education      1=HS, 2=SC, 3=College
    age            age in years
    uulbj          lbj position urban unrest
    uuhhh          humphrey pos urban unrest
    uunixon        nixon position urban unrest
    uuwallace      wallace pos urban unrest
    uuself         self placement urban unrest
    vnmlbj         lbj pos vietnam
    vnmhhh         hhh pos vietnam
    vnmnixon       nixon pos vietnam
    vnmwallace     wallace pos vietnam
    vnmself        self placement vietnam
    voted          1=voted, 5=did not vote
    votedfor       who voted for -- 1 = humphrey, 2= nixon, 3=wallace
    wallace        wallace therm
    humphrey       humphrey thermometer
    nixon          nixon thermometer
    mccarthy       mccarthy thermometer
    reagan         reagan thermometer
    rockefeller    rockefeller thermometer
    lbj            lbj thermometer
    romney         romney thermometer
    kennedy        robert kennedy thermometer
    muskie         muskie thermometer
    agnew          agnew thermometer
    lemay          "bombs away with Curtis LeMay" thermometer
    
    The control card file for the metric unfolding procedure is shown below. The first line has the name of the data file. The first number in the second line is the number of stimuli, the next two numbers are the minimum and maximum number of dimensions to estimate, and the "10" is the number of iterations.

    The third line contains some "antique" options we will never use. The only numbers that matter on this line are the "4" which indicates the number of identifying characters to read off each line of the data file (e.g., the respondent id number), and the "2" at the end. This is the number of missing data codes which appear in the sixth line.

    The first number in the fourth line is a tolerance value -- leave it as is. The next three numbers are parameters to transform the input data into squared distances. In this case, let amx=-.02, bmx=2.0, and cmx=2.0. The following equation transforms the thermometers into squared distances:

    d2 = (amx*t+bmx)cmx

    where t = input data. This formula takes a linear transformation of the input data to the power cmx. With amx = -.02, bmx = 2.0, and cmx = 2.0, this is equivalent to subtracting the thermometer score from 100, dividing by 50, and then squaring. This converts t from a 0-100 scale to a 4-0 scale. If the data, t, are distances, set amx = 1.0, bmx = 0.0, and cmx = 2.0. If the data are correlations, set amx = -1.0, bmx = 1.0, and cmx=2.0 or 1.0 if the correlations are initially regarded as unsquared or squared distances respectively.

    The next value, "1.5", is the maximum absolute expected coordinate value on any dimension. It is used for plotting purposes. If the squared distances are confined to a 4-0 scale, xmax=1.5 is usually sufficient. The last two numbers, "0.0" and "100.0", are the minimum and maximum expected values of the input data. These are used to catch coding errors in the input data. Anything out of range is treated as missing data.

    The fifth line is the format of the data file and the sixth line contains the missing data codes.

    Finally, the last 12 lines are labels for the stimuli.
    OLS68B.DAT
       12    2    2   10    0    0
        1    1    0    4    2
        .001  -0.02    2.0     2.0     1.5     0.0   100.0
    (1X,4A1,60X,12F3.0)
     98 99
    WALLACE
    HUMPHREY
    NIXON   
    MCCARTHY
    REAGAN
    ROCKEFELLER
    LBJ   
    ROMNEY 
    R.KENNEDY
    MUSKIE   
    AGNEW
    LEMAY   
    1. Run MLSMU6. It will produce an output file called FORT.22. The first 20 lines look like this:
      
       WALLACE          1.2646    0.5154  217.4823    0.5541 1242.0000
       HUMPHREY        -0.5559    0.3738  114.7892    0.6968 1252.0000
       NIXON            0.1480   -0.5415  123.2209    0.5319 1250.0000
       MCCARTHY        -0.6251   -0.4938  151.8926    0.3854 1204.0000
       REAGAN           0.3080   -0.8895  131.8091    0.4380 1212.0000
       ROCKEFELLER     -0.5579   -0.5995  148.1413    0.3724 1229.0000
       LBJ             -0.5223    0.4905  147.0334    0.5573 1253.0000
       ROMNEY          -0.4736   -0.7866  111.3147    0.3434 1167.0000
       R.KENNEDY       -0.4245    0.2351  148.8571    0.5418 1242.0000
       MUSKIE          -0.6611    0.1660  126.0836    0.4862 1177.0000
       AGNEW            0.2341   -0.8706  114.1418    0.4675 1180.0000
       LEMAY            1.1901    0.4267  174.3242    0.4601 1188.0000
       1681            -0.0285    0.2555    0.7918    0.6824   12.0000
       1124            -0.1768    0.2692    1.4788    0.6992   12.0000
         78             0.5707   -0.1514    3.5611    0.2141   12.0000
        553             0.1376    0.1064    0.1597    0.7047    9.0000
          7             0.2542    0.1235    1.2634    0.0116   12.0000
        412             0.2781    0.0867    0.1024    0.6197   12.0000
        631             0.5017    0.1088    1.1196    0.0742   12.0000
       1316             0.2175   -0.5842    1.1568    0.8577   12.0000
                             etc etc etc
                             etc etc etc
      The first two columns after the names are the two dimensional coordinates. The first 12 lines are the coordinates for the political candidates and lines 13 onward are the coordinates for the respondents. Use R to plot the 12 candidates in two dimensions. This plot should be very similar to the one you did for Question 1 above.

    2. Use Epsilon to insert the voted and voted for variables into FORT.22 (strip off the candidate coordinates first). Turn in the Epsilon macro you used to do the insertion and the first 20 lines of the file.

    3. Use R to make two-dimensional plots of the Voters, Non-Voters, Humphrey Voters only, Nixon Voters only, and Wallace Voters only. For example, your Humphrey Voter plot should look something like this:



      Label each plot appropriately and use solid dots to plot the respondents. Turn in all these plots.

  4. The aim of this problem is to analyze the 2000 thermometer scores using metric unfolding in the same fashion as above. Download the control card file and data file and place them in the same directory.

    UNFOLD_2000.CTL -- Control Card File for Metric Unfolding Program
    ELEC2000.DAT -- 2000 Election Data

    Here are the first four lines of ELEC2000.DAT.
    
        10787  4  8  0  0  0  2 49  1   0  65  60  30  40  70  50 998 998  40  59  75  63  65  6  1  3  6  4  0  2
        21271  2  6  0  1  0  2 35  1  50  50  50  50 997  50   0  50 997  50 100   0 100   0  4  4  2  6  8  0  0
        40285  2  6  0  0  0  2 63  0  70  55  55  60  65  55  55  65  50  60  70  65  65  90  6  5  5  5  5  0  1
        50191  6  6  0  1  0  2 40  1  50  40  80  60  60  80  70  50  70   0  20  90  70  70  6  2  2  6  4  0  2
    
    The variables, in order, are:
    
        RESPONDENT ID      = unique 8 digit number
        PARTY ID           = 0 to 6 -- 0 = Strong Democrat
                                       1 = Weak Democrat
                                       2 = Lean Democrat
                                       3 = Independent
                                       4 = Lean Republican
                                       5 = Weak Republican
                                       6 = Strong Republican
        FAMILY INCOME      = 1 to 22 - 1.   A. NONE OR LESS THAN $4,999
                                       2.   B. $5,000-$9,999
                                       3.   C. $10,000-$14,999
                                       4.   D. $15,000-$24,999
                                       5.   E. $25,000-$34,999
                                       6.   F. $35,000-$49,999
                                       7.   G. $50,000-$64,999
                                       8.   H. $65,000-$74,999
                                       9.   J. $75,000-$84,999
                                       10.  K. $85,000-$94,999
                                       11.  M. $95,000-$104,999
                                       12.  N. $105,000-$114,999
                                       13.  P. $115,000-$124,999
                                       14.  Q. $125,000-$134,999
                                       15.  R. $135,000-$144,999
                                       16.  S. $145,000-$154,999
                                       17.  T. $155,000-$164,999
                                       18.  U. $165,000-$174,999
                                       19.  V. $175,000-$184,999
                                       20.  W. $185,000-$194,999
                                       21.  X. $195,000-$199,999
                                       22.  Y. $200,000 and over
                                
        RACE               = 0 White, 1 Black
        SEX                = 0 Man, 1 Woman
        SOUTH              = 0 North, 1 South
        EDUCATION          = 1 High School or less, 2 Some College, 3 College
        AGE                = In Years
        MARRIED            = 0 Single, 1 Married
    
        FEELING THERMOMETERS   (0 TO 100)
    
                           = CLINTON  
                           = GORE     
                           = BUSH     
                           = BUCHANAN 
                           = NADER    
                           = MCCAIN   
                           = BRADLEY  
                           = LIEBERMAN
                           = CHENEY   
                           = HILLARY CLINTON
                           = DEMOCRATIC PARTY
                           = REPUBLICAN PARTY
                           = REFORM PARTY
                           = PARTIES IN GENERAL
        
        LIBERAL-CONSERVATIVE  SCALE  (1=EXTREMELY LIBERAL, 2=LIBERAL, 3=SLIGHTLY LIBERAL,
                                 4=MODERATE; MIDDLE OF THE ROAD, 5=SLIGHTLY CONSERVATIVE,
                                 6=CONSERVATIVE, 7=EXTREMELY CONSERVATIVE)
    
    
                           = SELF-PLACEMENT
                           = CLINTON
                           = GORE
                           = BUSH
                           = BUCHANAN
    
        PRE-POST INTERVIEW = 1 IF PRE-ELECTION INTERVIEW ONLY
        VOTE CHOICE        = 0 NON-VOTER
                           = 1 GORE
                           = 2 BUSH
                           = 3 3RD PARTY
    
    MLSMU6 expects to read UNFOLD.CTL!! Consequently, rename your current UNFOLD.CTL to UNFOLD_1968.CTL and then you can copy UNFOLD_2000.CTL to UNFOLD.CTL.

    1. Run MLSMU6. It will produce an output file called FORT.22. The first 20 lines look like this:
      
       CLINTON         -0.7879   -0.0317  153.1404    0.7198 1477.0000
       GORE            -0.7133   -0.1701  112.1776    0.7061 1468.0000
       BUSH             0.8234   -0.2492  149.4325    0.5889 1458.0000
       BUCHANAN         0.4576    1.0536  178.0645    0.3114 1246.0000
       NADER           -0.2737    0.7599  174.6307    0.2645 1094.0000
       MCCAIN           0.2850   -0.6498  122.5794    0.3691 1182.0000
       BRADLEY         -0.0780   -0.7509  106.1498    0.3689 1088.0000
       LIEBERMAN       -0.3428   -0.6394  107.1314    0.4758 1096.0000
       CHENEY           0.7002   -0.4687  107.9753    0.5099 1147.0000
       HILLARY         -0.8617    0.0625  203.7540    0.6459 1466.0000
       DEMPARTY        -0.6788   -0.1713  112.8208    0.6861 1453.0000
       REPUBPARTY       0.8235   -0.3286  142.8395    0.5546 1447.0000
       REFORMPTY        0.1644    1.0398  132.9094    0.3140 1128.0000
       PARTIES          0.1949   -0.7946  158.0865    0.2290 1413.0000
          1             0.3666   -0.0534    1.2943    0.3584   12.0000
          2            -0.2740    0.6767    2.5447    0.5465   12.0000
          4             0.0094    0.0645    0.8008    0.0000   14.0000
          5             0.5073   -0.0010    1.3888    0.6249   14.0000
          7             0.2719    0.0294    0.3600    0.5458   14.0000
          8            -0.6582   -0.2931    1.8168    0.7683   14.0000
                                   etc etc etc
                                   etc etc etc
      Use R to plot the 14 stimuli in two dimensions. This plot should be similar in format to the ones you did for the 1968 configuration above.

    2. Use Epsilon to insert the VOTE CHOICE variable into FORT.22 (strip off the candidate coordinates first). Turn in the Epsilon macro you used to do the insertion and the first 20 lines of the file.

    3. Use R to make two-dimensional plots of the Voters, Non-Voters, Gore Voters only, and Bush Voters only. For example, your Gore Voter plot should look something like this:



      Label each plot appropriately and use solid dots to plot the respondents. Turn in all these plots.

    4. Use R to make smoothed histograms -- using the first dimension from the thermometer scaling -- of the Voters and Non-Voters only, and the Gore Voters, Bush Voters, and Non-Voters. For example, your Bush-Gore-NonVoters plot should look something like this:



      Here is the trick for getting the percentage breakdown in the plot:
      #
      text(-1.0,0.53,paste("Gore Voters ",  
                      100.0*round(goreShare, 3)),col="red",font=2)
      text(-1.0,0.5,paste("Bush Voters ",
                      100.0*round(bushShare, 3)),col="blue",font=2)
      text(-1.0,0.47,paste("Non Voters  ",
                      100.0*round(nonShare, 3)),col="black",font=2)
      #
      Turn in the R code and the plots.

  5. In this problem we are going to use the classic W-NOMINATE program to analyze the 104th House and Senates. Download the program, control card files, and data files and place them in the same directory.

    WNOM9707 -- W-NOMINATE Program
    W-NOMINATE is discussed in detail with several examples on the W-NOMINATE Page.

    1. Run W-NOMINATE on the 104th House. Turn in the NOM21.DAT output file.

    2. The legislator coordinates are in the output file NOM31.DAT. The first few lines should look similar to this:
          11049990999 0USA     100  CLINTON       55  17   6  99  0.733 -0.587  0.065
          21041509041 1ALABAMA 20000CALLAHAN     563  58  15 503  0.853  0.729  0.043
          31042930041 2ALABAMA 20000EVERETT      571  70  21 502  0.825  0.746  0.041
          41041563241 3ALABAMA 10000BROWDER      459  89 120 462  0.664 -0.037  0.015
          51041100041 4ALABAMA 10000BEVILL       438 121 108 462  0.660 -0.081  0.015
          61042910041 5ALABAMA 10000CRAMER       470  97 116 480  0.662 -0.023  0.014
          71042930141 6ALABAMA 20000BACHUS       602  32  34 488  0.856  0.689  0.036
          81042930241 7ALABAMA 10000HILLIARD     457 121  34 504  0.726 -0.643  0.023
          91041406681 1ALASKA  20000YOUNG, DON   519  49  31 466  0.814  0.592  0.031
         101042950061 1ARIZONA 20000SALMON       615  37  45 468  0.840  0.816  0.045
                                  etc etc etc
                                  etc etc etc
      The legislator coordinates are in the next to the last column (shown in red). For example, former President Clinton's coordinate is -0.587. Use R to make a smoothed histogram of the Republicans and Democrats in the 104th House using the estimated coordinates above (the party code is shown in blue). Use arrows to show the locations of former President Clinton and former Speaker of the House Newt Gingrich.

    3. Use Epsilon to change the number of dimensions to "2" in NOMSTART.DAT and run the program again (be sure to save NOM31.DAT from the one dimensional run as it will be overwritten). NOMSTART.DAT (NOMSTART.H104) looks like this:
      HOU104KH.ORD                Name of Data File
      NOMINAL MULTIDIMENSIONAL UNFOLDING   Title of Run
       1321    1    5             Number RCs, Left on 1st, Up on 2nd
          1   36                  Number Dimensions, Number Characters to Read From Header
       15.0000  0.5000            Starting Values for BETA and WEIGHT
        0.0250   20               RC Min. Cutoff, Number RCs for Legislator
      (36A1,3600I1)               Format for Roll Call File
      (1x,I4,36A1,1X,4i4,51f7.3)  Format for Legislator Coordinate File -- NOM31.DAT
      (1x,I4,36A1,1X,51f7.3)      Format for H-S Coordinate File -- FORT.34
      The red "1" in the fourth line is the number of dimensions. This is the number you should change to "2". Leave everything else in the file the same! Run W-NOMINATE on the 104th House in two dimensions. Turn in the NOM21.DAT output file.

    4. The two-dimensional coordinate file looks like this:
          11049990999 0USA     100  CLINTON       54  17   7  99  0.729 -0.594 -0.116  0.066  0.129
          21041509041 1ALABAMA 20000CALLAHAN     561  36  17 525  0.880  0.758  0.653  0.042  0.132
          31042930041 2ALABAMA 20000EVERETT      569  50  23 522  0.852  0.774  0.634  0.040  0.132
          41041563241 3ALABAMA 10000BROWDER      473  35 106 516  0.733 -0.034  0.842  0.015  0.083
          51041100041 4ALABAMA 10000BEVILL       448  57  98 526  0.723 -0.079  0.759  0.015  0.075
          61042910041 5ALABAMA 10000CRAMER       474  46 112 531  0.710 -0.018  0.688  0.015  0.064
          71042930141 6ALABAMA 20000BACHUS       597  25  39 495  0.861  0.695  0.232  0.035  0.078
          81042930241 7ALABAMA 10000HILLIARD     445 117  46 508  0.748 -0.642  0.417  0.023  0.046
          91041406681 1ALASKA  20000YOUNG, DON   517  34  33 481  0.839  0.600  0.583  0.031  0.122
         101042950061 1ARIZONA 20000SALMON       623  35  37 470  0.842  0.826 -0.119  0.045  0.075
                                  etc etc etc
                                  etc etc etc
      The legislator coordinates are are the third and fourth columns from the end (shown in red). For example, former President Clinton's coordinates are -0.594 and -0.116. Use R to plot the legislators in two dimensions. Use "D" for Non-Southern Democrats, "S" for Southern Democrats, "R" for Republicans, and "P" for President Clinton. This graph should be in the same format as the one you did for question 2.f of Homework 5.

    5. Run W-NOMINATE on the 104th Senate. Turn in the NOM21.DAT output file.

    6. Use R to make a smoothed histogram of the Republicans and Democrats in the 104th Senate using the estimated coordinates from the NOM31.DAT file. Use arrows to show the locations of former President Clinton, and Senators Kennedy and Helms.

    7. Use Epsilon to change the number of dimensions to "2" in NOMSTART.DAT and run the program again (be sure to save NOM31.DAT from the one dimensional run as it will be overwritten). Use R to plot the legislators in two dimensions. Use "D" for Non-Southern Democrats, "S" for Southern Democrats, "R" for Republicans, and "P" for President Clinton. This graph should be in the same format as the one you did for question 2.f of Homework 5.