Appliication conditions of sum of squares types(Type 1, type 2 and type 3) acording to data types (Balanced, unbalanced) and exami̇nati̇on of type 1 error

Analysis of variance, hypothesis tests I. type, II. Type error and the sum of squares types (Type I, Type II and Type III) in our thesis, we also provide information about the methods used in calculating the sum of squares analysis of variance (Type I, Type II and Type III) according to data types (balanced, unbalanced) application cases 1 In terms of type error, we made a simulation study in fortran program. The results obtained in our 100 000 simulation studies using balanced and unbalanced data sets for 4 different distributions (z, t, beta and chi-square) are listed in tables. In the experimental designs where the data showed a balanced distribution as a result of 100 000 simulations, 4 different distributions (z, t, beta and chi-square) were calculated using the sum of Type I, Type II and Type III squares. It was found that there was no difference and the results obtained were at the level (α = 0.05) of the first type error. In the 100 000 simulation studies using 8 different unbalanced trial design scenarios and Type I, Type II and Type III squares for 4 distributions (z, t, beta and chi-square), type 1 error was initially determined at 0.005.