Consumer Behavior Research Methods
Lernen
« zurück zur Kartenübersicht « Liste
137/157 [137]

Factor Analysis

2. How do I interpret the factors? - 
2.2 Interpret rotated solutions

Check “Rotated Component Matrix”

  • Matrix presents factor loadings of each item = correlation between factor and each original item

  • Items with high loadings on a given factor are used for interpreting the factor

  • Factor loadings can be positive or negative (depends on item scaling) --> consider absolute values of factor loadings for interpretation

  • Look for simple structure: each variable (hopefully) loads high on 1 factor and low on other factors

  • If the mathematical factor produced by the analysis represents some real-world construct then common themes among highly loading questions can help us identify this construct
 
 
Highlight the highest factor loadings per row, and then look at highlighted items for each factor

Sometimes an item loads equally high on 2 factors

  • Decide on content to which factor it belongs

  • Include it in the subsequent item analysis of both factors to see where it fits best statistically (one at a time!)

  • This may mean that there are correlated factors --> use oblique rotation (note: not part of this course, do not use in the assignment)

 
Final Intepretation

  • Once found the optimal solution, it is possible to consider the factors as new “macro-variables” to use instead of the original variables in further analyses

  • This can also serve as the input for the cluster analysis 

Diskussion