Sample Size in Multiple Regression Models: A simulation study

The aim of the research is to study efficiency following indicators multiple regression model, i.e., value "F", R2, Adj., Standard Beta B, Unstandard and EMS, in light different sample sizes using three methods (standard – gradual hierarchical). descriptive method was used. community consisted a virtual that simulates reality, it obtained through simulation method. It consists 500 items, created Mics. Excel program, which are observations represent one dependent variable denoted by symbol (Y), five independent variables, coded (X1, X2, X3, X4, X5), with evaluating models - progressive hierarchical), number samples, range from (10 ≤ n ≥ 500). noticed conditions were not available samples less than 50, so studied numbers ranging (50 results indicated all increase improve increasing sample, best estimate for when (n = 225), at rate (45) each variable, as R2 (43%), equal modified All good, that, until reaching entire population (n=500), percentage improvement small, (45%). also there no reliance on only indicator judge quality models, well difference according statistical used, especially small samples. recommended large studies, relying more know taking into account used importance researcher's variables.