T-Test Indep Variable Example 2

T-Test Indep Variable Example 2 - numerator ((n1-1)Var1 +...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
#13.112 Days of Illness Study: Vitamin Supplement Group versus Placebo Group Step1 Same or significantly more illness with Supplements Ha: (MuSupplemenSignificantly Fewer illness days with Supplements Step 2: A 0.05 Compute Degrees of Freed 94 One-Tail Test: (Directional Alternative Hypotheses) Rejection region would be in LEFT TAIL, because it is a 1-tailed test, and the alternative hypothesis suggests the region of rejection is to the LEFT of MU Decision Rule using Tvalue(s): Reject the Null Hypothesis if Ttest is LESS th -1.661 Decision Rule Using Pvalue: Reject the Null Hypothesis if Pvalue is LESS THAN Alpha Step 3 Data is Collected; Next, Compute the Test Statistic t-Test: Two-Sample Assuming Equal Variances Supplement Placebo Mean 19.02 21.85 Variance 41.34 25.49 Observatio 48 48 Pooled Va 33.41 Hypothesi 0 df 94 t Stat -2.4 P(T<=t) on 0.01 t Critical o 1.66 P(T<=t) two 0.02 t Critical tw 1.99 Step 3 Data is Collected; Next, Compute the Test Statistic SupplemePlacebo Mean 19.02 21.85 SDEV 6.43 5.05 n 48 48 VAR= 41.34 25.49 Compute Degrees of Freed94 First Compute Pooled Variance: = Sp2
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: numerator ((n1-1)Var1 + (n2- 3140.96 denominato (n1 + n2 -2) 94 Sp2= 33.41 Compute Observed Difference between Two Sample Means (Xbar1 - Xbar2)=-2.83 Next, compute Ttest for Two Sample means: Ho: (MuSupplemen Ttest = (Xbar1-Xbar2) / SQRT(pooled varian-2.4 Pvalue= 0.01 Step 4 Statistical Decision? Statistical DecisionReject the Null, using either decision rule above Step 5 Conclusion? ConclusioMean days of illness for Supplement Group Significantly less than Placebo group Pvalue is less than Alpha (.05), so we can rule out chance. Conclude that taking supplements significantly reduces number of illness days, compared to taking placebo pills (I.e., nothing at all). If a Decision error w could be Type I error. Potential Consequences of such an error? Would approve Supplement Therapy, when in fact, days of illness are not reduced, and may even be increased. Resulting in wasted medical expenses, and greater illness....
View Full Document

This note was uploaded on 04/08/2008 for the course STAT 2102 taught by Professor Dr.pred during the Spring '07 term at Temple.

Page1 / 2

T-Test Indep Variable Example 2 - numerator ((n1-1)Var1 +...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online