Browsing by Subject "psykometrik"

Sort by: Order: Results:

Now showing items 1-1 of 1
  • Ritola, Ville (Helsingfors universitet, 2016)
    Background. Knowing what a psychological test measures and if it works the same way in different contexts, i.e. has measurement invariance (MI), is crucial for its valid and reliable use. The Wechsler Adult Intelligence Scale – Fourth Edition (WAIS-IV) was published in Finland in 2012. However, recent research suggests that the factor model given in the WAIS-IV test manual and the information regarding MI between different age groups and levels of education are lacking. Methods. This study employed the normative sample of the Finnish WAIS-IV. First, the factor model in the manual was examined and improved using confirmatory factor analysis with a mixed data-theory approach. Second, the new model was tested for strict residual MI for different age groups and levels of education, in order to study if the test reaches an acceptable level of MI. Results and conclusions. The results indicated that the normative data is best modeled by an oblique non-g model. The study also replicated a Spatial Visualization factor with loadings from Block Design, Visual Puzzles and Picture Completion, and Quantitative Reasoning factor with Figure Weights and Arithmetic. A previously unmentioned link in factor analytic literature on WAIS-IV was found between Block Design and Processing Speed factors. The results questioned the link between Arithmetic and Verbal Comprehension factor and found the underlying source of shared variance to be links between Information and Arithmetic, which was interpreted as Educational Achievement. WAIS-IV reached strict residual MI for both different age groups and levels of education. The study offers a more accurate factor model of WAIS-IV and gives confidence that psychologists can reliably apply it over different ages and levels of education in the normal population of Finland.