Bilimsel Araştırmalarda Varyans Kontrolü için MaxMinCon İlkesi


Tan Ş.

Eğitim Araştırmaları, sa.6, ss.147-156, 2002 (Diğer Kurumların Hakemli Dergileri)

  • Yayın Türü: Makale / Tam Makale
  • Basım Tarihi: 2002
  • Dergi Adı: Eğitim Araştırmaları
  • Sayfa Sayıları: ss.147-156

Özet

In experimental research, experimental and control groups must be equalized in term of variables that are not in the experimental design but may effect  the dependent measures. All the groups must be equall in non-study variables so that the differences between the groups could be explained by the effects of the independent variables that are used in the design. In experimental studies, controlling of the extranoucous varibles could be provided by; randum assignment of subjects, using matched subjects, using covariance analysis, repeated measure designs, and, including more independent variables in the analysis. All this methods could be summurized by MAXMINCON principle in scientific researches: MAX; component stants for  maximizing the between group variance, MIN; component stants for minimizing the within group variances, and, CON; stants for controling the independent variables that are not in the experimental design and may effect the dependent measures.