Exam 1

1. Metainformation

   Tag	Value
   file	Reliability_eur-reliability-210-en_eur-reliability-210-en
   name	eur-reliability-210-en
   section	Reliability/Analysis/Cronbach's alpha
   type	schoice
   solution	FALSE, FALSE, FALSE, TRUE
   Type	Conceptual
   Program
   Language	English
   Level	Statistical Reasoning

   Question

   In the population, shyness is normally distributed with a mean of 12 and a standard deviation of 5. The reliability of a shyness test is .78 (Cronbach’s alpha). In a study on shyness, the 40 shyest children were selected (children with the highest scores on the test). Twenty of these children were assigned to a control groups and 20 to a treatment group. The children in the treatment group received an assertiveness intervention. After the intervention, the shyness test was administered again to all of the children. Below, the means of the control and the treatment groups on both measurements are given.

   Group	Mean_pretest	Mean_posttest
   Control group (n=20)	20	18
   Treatment group (n=20)	20	16

   Evaluate the statements below:

   I. If the reliability of the shyness test were to be 0, then the mean on the pre-test would have been higher for both groups.

   2. Because there is a control group, there is no regression to the mean in this case.
   
   
   1. FALSE: Statement I and statement II are both true
   2. FALSE: Statement I is true and statement II is false
   3. FALSE: Statement I is false and statement II is true
   4. TRUE: Statement I and statement II are both false
   
   

   Solution

   Both statements are false. If the reliability were 0, then you have only measurement error and in that case the scores on the pre-test could have been either higher, lower, or equal. Including a control group does not influence the presence of regression to the mean, but merely concerns the design of the study and whether causal conclusions may be drawn (see WE chapter 7).

   
   
   1. False
   2. False
   3. False
   4. True