Comparison of the performances of parametric k-sample test procedures as an alternative to one-way analysis of variance

Objectives: The performances of the Welch test, Alexander-Govern Brown-Forsythe test and James Second-Order which are among parametric alternatives one-way analysis variance included in literature, to protect Type-I error probability determined at beginning trial a nominal level, were compared with F test. Methods: Performance tests error; cases where variances homogeneous heterogeneous, sample sizes balanced unbalanced, distribution data is accordance normal log-normal distribution, how it affected by change number groups be has been examined on simulation scenarios. Results: not performed well situations heterogeneous. was when size equal. Conclusions: that can recommended as an alternative