There are no guidelines as such, but the paper reporting the
standardisation of the questionnaire said they had 15% missing data, but
didn't mention a pattern.
However would I be right in saying that a pattern to the missing data is
more important than the sheer amount e.g. if say 20% is missing completely
at random this is less important than 10% missing but showing a pattern?
On Apr 23 2010, Joe Butler wrote:
>Are you able to access the publishers guidelines for the scales? As in
>my experience the authors usually state the number of questions that
>can be ommited before they lose validity.
>
>Cheers,
>J
>Sent from my iPhone
>
>On 23 Apr 2010, at 16:04, "[log in to unmask]"
><[log in to unmask]> wrote:
>
>>
>
|