This might be a stupid question but I have tried everything I can come up with and searched for answers everywhere, so I hope that maybe I can get some help here.

I'm at the end of my BA in forensic psychology and have contstructed a pilot study to check if my survey and measuring instruments will work for my final study.

The important questions are supposed to measure how trustworthy people think that people with downs syndrome is when giving a statement after being a victim of a crime. 

In this pilot study I have two questions to measure trustworthiness, and to me it seems obvious that they are measuring the same thing.

Roughly translated to english they are:

How trustworthy do you feel Anna was when she claimed her cash that was stolen?

Not trustworthy at all                                                                                                                                                                        Completely trustworthy
            1                               2                             3                              4                                     5                                  6                                   7


How trustworthy do you feel Anna was when she claimed her necklace was stolen?

Not trustworthy at all                                                                                                                                                                        Completely trustworthy
            1                               2                             3                              4                                     5                                  6                                   7


However a reliability test shows an Cronbach's alfa of ,289

There's 20 respondats in the pilot study, half was given the information that Anna has downs syndrome and the other half didn't. 





------------------------------
zanno Jacklin
------------------------------

#SPSSStatistics

Hi. If you look at a page like this, you will see that alpha is based on the number of test items, the average covariance, and the average variance. With a mere 20 subjects, you are going to have a lot of variability from sample to sample.

It may seem obvious to you that they are measuring the same thing (did you look at the correlation between the variables?), but your 20 cases tells a different story.

------------------------------
Rick Marcantonio
Quality Assurance
IBM
------------------------------