Stephen,
Assuming that the items are all rated on the same metric. why not perform the analyses on the average rating, rather than the sum of ratings?
Regards,
Stephen Brand
---- "Salbod wrote:
> Dear Friends,
> A colleague has presented me with an interesting problem. She used a measure with nine subscales. Three Parenting Attributes (PA): Bonding, Discipline and Education are crossed with three Age Levels: Infant, Adolescent, and Adult. Thus, the nine scales were (1)Bonding-Infant, (2)Bonding-adolescent, (3)Bonding-Adult, (4)Discipline-Infant, (5)Discipline-adolescent, (6)Discipline-Adult, (7)Education-Infant, (8)Education -adolescent, and (9)Education -Adult.
> The number of statements (behaviors) that respondents are asked to endorse (5 point scale) within each PA are different: 14, 9, and 4, respectively.
>
> The problem is that the sum of the PA scales (Bonding, Discipline, and Education) are based on different number of items.
>
> My question is: how can I transform the data so that it can be treated with a 3X3 Repeated Measure ANOVA?
>
> I want to thank you in advance for any suggestions you will send me about a possible solution to this problem.
>
> Warm regards,
>
> Stephen Salbod
> Psychology Department
> Pace University, NYC
>
--
For personalized and experienced consulting in statistics and research design, visit www.statisticsdoc.com
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
