Direct naar inhoud

Accounting For Non Normality In Latent Regression Models Using A Cumulative Normal Selection Function

Door: Molenaar, D. | 01-01-2007 In educational measurement, group differences in ability are investigated using latent regression models. Commonly, these models assume a normal distribution of the latent ability, conditional on group membership.

However, this assumption could be problematic. In the present paper, a new model is discussed in which the latent ability is modelled as if it has arisen from a latent selection process. This will enable the analysis of group differences under a more general family of distributions than the commonly used normal distribution.

Read more
Medewerker aan telefoon

Kunnen we je helpen?

Stel je vraag via onze kanalen of kijk in de veelgestelde vragen.
Voor scholen: Vergeet niet om het brinnummer bij de hand te hebben en/of in de mail te vermelden, zodat we jouw vraag sneller kunnen behandelen!

Bereikbaar Ma t/m vr 08.30 tot 15.00 uur
Bellen (026) 352 11 11
E-mail klantenservice@cito.nl

Zoeken