No, look at the effect of the intervention on time 2, controlling for time 1.
(What was the answer to your previous problem?)
Jeremy
On 2 August 2011 16:30, Amy Green <[log in to unmask]> wrote:
> Hi Jeremey,
>
> Thanks for your response. I have found the answer to my previous questions.
>
> I now have another problem. I have conducted research with a control group
> and intervention group. Each with a time 1 and time 2. To find the effect of
> the intervention would it be correct to analyse the data from time 2 between
> the groups and leave time 1 out of the picture.
>
> Thanks
> Amy
>
> On Mon, Aug 1, 2011 at 11:37 PM, Jeremy Miles <[log in to unmask]>
> wrote:
>>
>> On 1 August 2011 15:15, Amy Green <[log in to unmask]> wrote:
>> > I tried to do a MANOVA with 3 IV's and 7 DV's which in spss returned the
>> > message that there were too many factors for it to do the command. In this
>> > instance should I do ANOVAs?
>> >
>>
>> I think we need more details, but if there are two many factors for
>> manova, there are likely to be too many for anova. What were the
>> factors, and how many levels did each have?
>>
>> > Also, I did a Cronbach's Alpha on a scale with two items but the 'if
>> > deleted column' in SPSS cannot be calculated. I assume this is because there
>> > are only two items as the rest of the test is fine. Am I correct in this
>> > assumption? And is this ok or should I use another reliability test in SPSS?
>> >
>>
>> I'm confused. Are there two items? Or what is the rest of the test?
>> Does alpha really make sense in a 2 item test? (It will just be [a
>> function of] the correlation between them.
>>
>> Jeremy
>
>
|