Pregledni rad
https://doi.org/10.21857/moxpjhz4dm
Factorial designs Analysis of variance (ANOVA) in types sum of squares 1, 2, 3 (type one SS, type two SS, type three SS) in unbalanced designs – principle of marginality
Siniša Opić
orcid.org/0000-0001-9800-0145
; Učiteljski fakultet Sveučilišta u Zagrebu, Zagreb, Hrvatska
*
* Dopisni autor.
Sažetak
The aim of this paper is to show the specifics of use, but also the limitations of factorial designs in quantitative differential designs in the field of social sciences. The specifics of two-way and multi-way (n-way) factorial designs in the Analysis of Variance (ANOVA) are presented with regard to the specifics of the types of sum of squares (type one SS, type two SS, type three sum of squares). An unbalanced 2*4 two-way ANOVA model was tested using Type three sum of squares (type three SS) in which the main effect was absent. Due to the principle of marginality, the main effects cannot be interpreted using the Type 3 sum of squares (type three SS), but for the main effects the need for a particular approach, one-way ANOVA and two-way analysis of variance (factorial design) is imposed, but not with using Type 3 sum of squares.
Ključne riječi
ANOVA; interaction effects; marginality; sum of squares of type 1, type 2, type 3
Hrčak ID:
313603
URI
Datum izdavanja:
22.1.2024.
Posjeta: 734 *