Introduction to SAS: Project 5

Project 5

This project uses column-wise input statements to construct a data set from a file. Categorical response data is analyzed using proc freq. The data is from the file prodpref.dat.

A study was used to determine whether a cost reduced product was equally preferred to the current product. 200 judges were asked their preference on the pair of samples, coded 27 and 45. One half of the group tasted sample 27 first and 45 second, the other half tasted the samples in the reverse order. Each judge was also asked to rate six qualities of these samples on a 0 to 6 scale (0=no response,1=excellent,... ,6 poor). The preferences of each judge were recorded as: 1, prefers first sample tasted 2, prefers second sample tasted 3, no preference. The data are in the file prodpref.dat. and are coded in the form

1   27   45   333424   543424   1
2   27   45   443324   533555   1
3   27   45   543314   543414   1
4   27   45   552232   343333   2

The first entry is the judge number, the second entry is number of the sample tasted first, the third entry is the number of the sample tasted second, the fourth entry is the six ratings for the first sample tasted, the fifth entry is the six ratings of the second sample tasted, and the last entry is the preference.

1. How many of the judges preferred sample 27? You may want to use some if-then statements in constructing an appropriate data set and also the n option in proc means. You could also use proc freq. Turn in a nice printout to back up your answer. There is no need to print out the data set here.

2. Now use chi-square tests to compare the distribution of the rankings of the two products for each of the six aspects of quality. The chi-square test is easily done using proc freq once the data is in a suitable form. The information on column-wise input and do loops may be useful. Are the two products interchangable? Back up your comments with some nice printouts, including a printout of your final program. There is no need to print out the data set here.

Contents Index <<Browse<< >>Browse>>