Dear Colleagues,
I have had the most unfortunate experience with my DClinPsy Research. I was conducting a simple online questionnaire project with adolescents. However due to lack of access to computers a number of schools opted to print the questionnaires. Unfortunately something to do with the formatting from Bristol Survey resulted in the last question Q26 of the Self-Compassion Measure (attached) not being printed.
I now have 392 questionnaires returned missing this item which is a central tenant of one of the subscales. Due to the anonymous nature of the data I cannot match participants if I desired to seek this response. I would really appreciate any advice people have on how best to manage this missing data issue?
Possible options
- Using a mean score of the items of the same subscale
- Using the item with nearest correlation as a guide
- I have sought data from a similar project to explore the role of this item
I do just now have the slight option of time and I could recollect next year, however this is a large amount of data and I am uncomfortable not using it if there is a viable way.
Any advice would be greatly appreciated
Ciara
Date: Tue, 20 May 2014 08:43:18 -0700
From:
[log in to unmask]Subject: Re: To report or not to report...
To:
[log in to unmask]I wouldn't bother to report things like the K-S test. (I never do a K-S test.)
I might report skew and kurtosis, and might say "three outliers were removed".
In a journal article you're usually short on space, so that kind of thing gets cut.
Jeremy