Question

Here is the scenario that I am working on:

The efficacy of a new addiction medication was evaluated in a randomized, placebo-controlled, doubleblind study. The medication in question, Antaquil, is intended to moderate the symptoms of alcohol withdrawal and craving with minimum side effects. Over the course of three weeks, a sample of 36 individuals who were recovering from alcohol addiction were randomly assigned to two groups: one administered the medication and one administered a placebo. At the end of the designated period, participants were administered the Obsessive Compulsive Drinking Scale (OCDS), an instrument that provides a global measure of thoughts about alcohol during nondrinking periods. Scores can range from 0 to 40 with higher scores signaling higher levels of rumination about alcohol. Prior to participation participants were all informed of the nature of Antaquil and were told they could leave the study at any time. Outcome data on the OCDS are shown below (also found in the Data Set Scenario 4 Excel file).

treatment group 40, 35, 27,18,30,28,11,23,30,13,16,17,26,22,19,17,29,10

placebo group 37,35,34,24,29,14,23,25,32,37,30,30,29,22,23,31,28,20

Hypothesis: Whether One Mean is Higher?

Answer 1

To carry out this test firstly we will check the Equality of variances assumption.

For this let us carry out equality of variance test by using Excel's data analysis tool pack.

F= 1.826 and P value=0.1122 which is GREATER THAN 0.05 THEREFORE WE CAN ASSUME THAT POPULATION VARIANCES ARE EQUAL.

Now

Null Hypothesis H0:

Alternative Hypothesis Ha:  

Let ​​​​​​ be the population mean for PLACEBO

and be the Population mean for Treatment.

Level of significance =0.05

t stat= 2.081

degrees of freedom= 34

P value=0.0450

Since P value Smaller than the level of significance 0.05 therefore SIGNIFICANT

Decision: REJECT H0.

Conclusion: We have sufficient evidence to conclude that Population mean for placebo is not equal to the Population mean for treatment.