The SAS data set RESULT contains the following variables: Region (GrpA or GrpB) Sales (dollars per year) Which SAS programs can be used to find the p-value for comparing GrpA sales with GrpB sales? (C

Question

Options

• Aproc ttest data = RESULT; class Region; var Sales; run;
• Bproc ttest data = RESULT; class Region; model Sales = Region; run;
• Cproc glm data = RESULT; class Region; model Sales = Region; run;
• Dproc glm data = RESULT; class Sales; model Sales = Region; run;

• A
  71% (17)
• B
  8% (2)
• C
  17% (4)
• D
  4% (1)

Explanation

Option A is correct because PROC TTEST requires a CLASS statement (to define the two groups) and a VAR statement (to specify the continuous response variable) - this is the exact valid syntax for a two-sample t-test comparing GrpA vs. GrpB sales.

Option C is also correct (the answer key likely intends A and C for "choose two"): PROC GLM uses CLASS for the grouping variable and MODEL dependent = predictor syntax, which is valid and produces an equivalent F-test p-value for a two-group comparison.

Option B is wrong because PROC TTEST does not accept a MODEL statement - that syntax belongs to PROC GLM/PROC REG; using it in PROC TTEST causes an error.

Option D is wrong because CLASS in PROC GLM must identify the categorical grouping variable (Region), not the continuous response variable (Sales) - putting Sales in CLASS while also using it as the dependent variable in MODEL is logically contradictory.

Memory tip: Think "T-test = VAR" (TTEST gets a VAR statement, not a MODEL), and "GLM = MODEL Y = X" (GLM always uses model notation with the response on the left). Any option that mixes these conventions is wrong.

No community discussion yet for this question.