1
$\begingroup$

If you are not fortunate enough to have a control group and have a single condition only, what meaningful questions can you ask about the change in pre to post intervention scores?

If ANOVA or ANCOVA are not feasible here, what methods--other than t-tests--would you use to pursue those meaningful questions about changes that occur within a single condition design?

Do you recommend looking at something like a reliable change index (RCI)?https://link.springer.com/referenceworkentry/10.1007%2F978-94-007-0753-5_2465

$\endgroup$
  • $\begingroup$ The first task is to understand what the 'scores' are measuring. How much does a typical change in score have to say about patient comfort? quality of life? freedom from recurring disease? What is important in assessing change: difference? ratio? sign? $\endgroup$ – BruceET Mar 16 at 3:53
  • $\begingroup$ @BruceET In my case, scores measure symptom severity that the intervention is designed to directly target. Change in score could indicate clinically meaningful symptom relief that may be attributable to the intervention itself. Although we have a separate, dichotomous index of symptom remission, the continuous outcome measure I am comparing at pre and post will not yield this information (remission) in and of itself. $\endgroup$ – user81715 Mar 16 at 16:19
  • $\begingroup$ What is important about change: I would say difference--can we use some fom of pre-to post difference index to say that the change was meaningful and not simply a reflection of the regression to mean or other similar artifcats. Btw, we don't have a cut-off score established for this outcome measure $\endgroup$ – user81715 Mar 16 at 16:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.