Uncategorized

The same data as in problem 8.3 are to be analyzed. However, they are arranged so that the four ratings from each judge are on one tine. Thus, columns 1-3 are for the judge ID, column 4 is the rating for New York, column 5 for New Orleans, column 6 for Chicago, and column 7 […]

Using the data set from problem 9.8, conduct a logistic regression predicting outcome (case or control) using SMOKING,ASBESTOS. and SES. Use a CLASS statement to create two dummy variables for SES, setting the reference level to ‘2-Mediurn’. (HINT: CLASS SES (PARAM=REF REF=’2-Medium’); PROC LOGISTIC uses the lower value formatted value of the outcome variable (if

Accident data, similar to Section G, are presented here. This time. we recorded accidents that occurred in the past year, based on the presence of a drinking problem and whether the driver had one or more accidents in the previous year. Run a forward stepwise logistic regression on these data, and write the resulting logistic

Repeat problem 9.8, except this time create your own dummy variables (i.e., do not use a CLASS statement). Make the reference level for SES ‘2-Medium’ as before. Problem 9.8 Run the following program to create a SAS data set called SMOKING. Outcome values of 1 represent cases (those with lung disease) and outcome values of

Add the necessary statements to your logistic regression in problem 9.8 to produce an ROC curve (sensitivity versus 1 – specificity). Use the OUTROC MODEL option to create a data set so you can plot the curve. Use PROC GPLOT (or PROC PLOT) to generate this curve.

Run the following program to create the data set PRINCIPAL: Write the statements necessary to run a factor analysis and produce a scree plot. Looking at this plot, decide how many factors you want to maintain, and add an option to the procedure to extract this many factors and to perform a VARIMAX rotation

Run a factor analysis on the questionnaire data in Chapter 3, Section B. Use only the variables PRES, ARMS, and CITIES. Request two factors, VARIMAX rotation method, and set the PRIORS estimate of conununality to SMC. Include the option to generate a scree plot.

Revise your program from problem 10.2 to perform a non-oblique rotation and compare the factor loadings with the results from 10.2. Output a data set containing the factor scores. Use PROC PRINT to list out the first 10 observations from this data set (hint: use the data set option 088=10). Compare these scores with a

Using the test scores in Chapter 11, Section B, run a factor analysis 011 the test, using the scored responses to each of the five questions as the items to factor analyze. Request only one factor and create an output data set containing that one factor score. Do not request any rotations. Use PROC PRINT

Given the following answer key: Write a SAS program to grade the six students whose data are shown here. Provide one listing in Social Security number order and another in decreasing test-score order. Compute both a raw score (the number of items correct) and a percentage score. Student Data: