SAS is the leading Statistical Software and is the tool of choice for statistcal analysis in the Financial, Pharmaceutical and Marketing Industries.
SAS can be used to optimise marketing campaigns, analyse delinquency rates of borrowers and the check bthe efficacy of new drugs. This page goes through a few practical examples of how SAS's statistical procedures are used in a variety of industries.
Consider a simple example: You are a debt collector and wish to determine which of two letters are more effective in making contact with the debtor. Your current letter maybe be written in a black font, but you wish to see if you can get an improved response by using a red font.
You have 1000 debtors to send letters to, you select 200 debtors at random from this 1000 to test the red font. After 2 weeks you are sent a text file with the results of your mailing campaign. A simple file containing a UniqueID, a flag saying whether or not the letter sent to that debtor had a red or black font and another flag saying whether or not the debtor replied to the letter. Since the samples are independent and the test and control groups are a random samples from the same population, it is valid to use a Chi-squared test for association between the categorical variables. This test can be easily performed using Proc Freq, to do this create a cross tab between the two variables you want to test and include the chisq option after the "/" on the table statement.
The output from the code above is shown in the windowbelow. The chi-squared test gives a p-value of 0.0014. This means that the probability is 0.0014, that such large differences in the percentages responding to letters could have been observed by chance, given no real difference in effectiveness of the two letters. With this strong evidence one would normally reject the null hypothesis and conclude that letters using a red font are more effective than letters using a black font.