2

I mesured two constructs,  intention to book a room (4 items) and intention to book from a web site (3 items), (Pearson correlation = 0.344**) I've been asked by a reviwer to merge the two scales. Please tell me what's the right way to do that! (both come from the same questionnaire) Factorial analysis of the 7 items at time showed One factor with 60.6 cumulative var thanks 

ZAck_Sorbonne
  • 63
  • 6
  • Try pca. Use first component. If FA succesful then it should works. – zlon Feb 14 '17 at 20:58
  • This answer is of general type and discusses politics of merging scales/items https://stats.stackexchange.com/a/260682/3277 – ttnphns Nov 12 '21 at 18:29

1 Answers1

0

You can either:

  1. Manually merge them. E.g., NewVariable = average(Variable 1, Variable 2, ...)
  2. Save the factor/component from a PCA/Factor analysis.
  3. If estimating a SEM of some kind, change the setup of the model so that it assumes a single latent variable.
Tim
  • 3,255
  • 14
  • 24