High correlations always bad?

Home Forums Methodspace discussion High correlations always bad?

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #1998
    Daand
    Participant

    I’m conducting an evaluation study in higher education on a strategy with underlying strategic subgoals. These subgoals have much in common. These all strive for competency based learning and more flexible learning programs. The sub goals are assessment of competencies, feedback on competencies etc. My correlation matrix shows high correlations, which considering the nature of the subgoals, is to be expected. Are high correlations in this case necessarily bad? 

    #2000

    When you expect that the subgoals have much in common – either because they are measures of the same construct or are causally related – a high correlation reflects what you expect. You could also do a factor analysis in order to check how strongly they load on a single underlying factor.

    #1999
    Daand
    Participant

    Great, thanks a lot, Ingo. That really helps. Factor analysis was fine and PCA too. 

Viewing 3 posts - 1 through 3 (of 3 total)
  • You must be logged in to reply to this topic.