{"id":12178,"date":"2018-03-29T12:24:28","date_gmt":"2018-03-29T12:24:28","guid":{"rendered":"https:\/\/data-flair.training\/blogs\/?p=12178"},"modified":"2021-05-09T13:16:29","modified_gmt":"2021-05-09T07:46:29","slug":"sas-t-test","status":"publish","type":"post","link":"https:\/\/data-flair.training\/blogs\/sas-t-test\/","title":{"rendered":"Learn SAS T-Test – SAS PROC T Test"},"content":{"rendered":"

In our last tutorial, we study SAS Linear Regression<\/strong>, Now we\u00a0will discuss another interesting statistical procedure SAS T-TEST that statisticians and scientists use.<\/p>\n

Moreover, we learn SAS PROC T-test with an output. At last, we will see types of T-test in SAS Programming<\/strong>: SAS Simple t-test, SAS paired t-test and SAS\u00a0two-sample t-test.<\/p>\n

So, let’s start to learn SAS T-Test tutorial.<\/p>\n

SAS T-TEST<\/h2>\n

A t-test in SAS Programming is an analytical test used to determine if there is a significant difference between two sets of data, or if the average of a set of data differs significantly from a predicted value.<\/p>\n

For example, consider that an analyst wants to study the amount that Gujaratis and Rajasthanis spend, per month, on clothing.<\/p>\n

It would not be practical to record the spending habits of every individual (or family) in both states, thus a sample of spending habits is taken from a selected group of individuals from each state. The group may be of any small to moderate size.<\/p>\n

The group may be of any small to moderate size \u2014 for this example, assume that the sample group is 200 individuals.<\/p>\n

The average amount of Gujaratis comes out to 10000; the average amount for Rajasthanis is 20000. SAS t-test questions whether the difference between the groups is representative of a true difference between people in Gujarat and people in Rajasthan in general or if it is likely a meaningless statistical difference.<\/p>\n

In this example, if, theoretically, all Gujaratis spent 10000 per month on clothing and all Rajasthanis spent 20000 per month on clothing, it is highly unlikely that 200 randomly selected individuals all spent that exact amount, respective to state.<\/p>\n

Thus, it is safe to conclude that the difference between sample groups is indicative of a significant difference between the populations, as a whole, of each state.<\/p>\n

\"SAS<\/a>

Overview of SAS T-test<\/p><\/div>\n

SAS t-test looks at the t-statistic, the t-distribution and\u00a0degrees of freedom to determine the probability of difference between populations.<\/p>\n

SAS PROC TTEST<\/h2>\n

The SAS PROC TTEST\u00a0is a procedure, which is used to carry out SAS t-test on a single variable and pair of\u00a0variables.<\/p>\n

SAS PROC T TEST Syntax-<\/strong><\/p>\n

PROC TTEST < options > ;<\/pre>\n
CLASS variable ;<\/pre>\n
PAIRED variables ;<\/pre>\n
BY variables ;<\/pre>\n
VAR variables ;<\/pre>\n
FREQ variable ;<\/pre>\n
WEIGHT variable ;\r\n<\/pre>\n
No statement can be used more than once. There is no restriction on the order of the statements after the PROC statement.<\/p>\n
PAIRED variables\u00a0<\/strong>identify the variables<\/strong> to be compared in paired comparisons. \u00a0Use the PAIRED statement only for paired comparisons.<\/p>\n
These PAIRED statements\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0Yield these comparisons <\/strong>
\nPAIRED A*B;\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 A-B
\nPAIRED A*B C*D\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0A-B and C-D
\nPAIRED (A B)*(C D);\u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0 \u00a0A-C, A-D, B-C, and B-D<\/p>\n
SAS One-Sample T-Test<\/h2>\n
Below we see one sample t-test SAS in which find the SAS t-test estimation for the variable weight_loss.<\/p>\n
Example-<\/strong><\/p>\n
PROC TTEST DATA=CLINIC H0=4 ALPHA=0.1;\r\n\u00a0VAR WEIGHT_LOSS ;\r\nTITLE 'TTEST OF H0: MEAN=4 FOR THE CLINIC GROUP';\r\nRUN;<\/pre>\n
ALPHA=p – specifies the confidence interval. By default ALPHA=0.05
\nH0=m – requests tests against m instead of 0. By default H0=0.<\/p>\n
 <\/p>\n
\"SAS<\/a>
SAS One-Sample T-Test – Example<\/p><\/div>\n
 <\/p>\n
\"SAS<\/a>
SAS One-Sample T-Test- Output<\/p><\/div>\n
 <\/p>\n
Two Sample T-Test SAS<\/h2>\n
The SAS PROC TTEST procedure is used to test for the equality of means for a two-sample (independent group) SAS t-test.<\/p>\n