So leading on from the previous blog. What I was really interested in was the quality of the feedback given to an individual through the four conditions mentioned in my last blog.
Just to remind you of the four conditions of the test:
1. An online automated product which gets respondents to to fill in a series of 40 questions about the individual and included free text feedback items as well. The individual then gets an aggregated document with the feedback split into sections. They do not know who submitted what feedback.
2. The pen and paper system was operated in two different conditions:
a. The first where the forms were sent direct from the respondent to the individual getting the feedback.
b. In the second condition the forms were sent to a third party (the individuals coach) who then aggregated and anonamised the feedback.
3. The respondents were interviewed face-to-face or over the phone by the individuals coach who then aggregated the feedback and gave it to the individual.
I then had the individual receiving and the respondents giving the feedback all rate (5 point scalar) the feedback in terms of:
In reverse order the results are (drum role)...
2a where the forms were sent direct to the individual
- The Receiver of the feedback: Usefulness average 2.1, Accuracy 2.0, Honesty 4.2
- The Respondents: Usefulness 2.9, Accuracy, 2.9, Honesty 1.8
1. Online automated system
- The Receiver of the feedback: Usefulness average 3.1, Accuracy 2.9, Honesty 3.7
- The Respondents: Usefulness 2.8, Accuracy 3.3, Honesty 1.5
2b Where the forms went to the coach
- The Receiver of the feedback: Usefulness average 3.2, Accuracy 3.5, Honesty 4.0
- The Respondents: Usefulness 4.2, Accuracy 4.1, Honesty 4.1
3. Fact to Face interview with the coach
- The Receiver of the feedback: Usefulness average 4.6, Accuracy 4.7, Honesty 4.6
- The Respondents: Usefulness 4.2, Accuracy 5.0, Honesty 4.9
I have all the tabulated data (sample size, conditions, frequencies, ranges, levels of significance etc) which I will post later.
So just looking at these figures there appears to be a clear difference between the way the feedback is
- Fed back to the individual
In my next post I will discuss these results in greater detail. I'm off for a few days trying to get the next book in some semblance of order. Until then...