The distribution of data was considered sufficiently normal for t-test (i.e., skewness < |2.0| and kurtosis< |9.0|.17 In addition, the homogeneity of variance
was tested with Levene's F-test, and since metaworry lacked homogeneity and created a bias in the parameters, therefore, statistical corrections were applied.
The results of histogram of data of current study depicted the approximate normal distribution (figure 1), whereas the P-P plot displayed a linear relationship (figure 2) and the residual plot presented no set patterns and, therefore, the assumption of homoscedasticity or homogeneity of variance
of the effect of instructional leadership was not violated.
Leven1s test for homogeneity of variance
has been carried out to assess the homogeneity of variance
In addition to the aforementioned tests, we conduct the Breusch--Pagan test to check the homogeneity of variance
in a linear regression.
Table 10 enlists values of means and standard deviations, test of homogeneity of variance
and ANOVA for PsyCap.
Levene's test was used to investigate the homogeneity of variance
. Dunnett's T3 and Bonferroni's methods for heterogeneity of variance and post hoc Scheffe's method for homogeneity of variance
In order to examine the effect of non-normality on F-test robustness, a one-way design with 3 groups and homogeneity of variance
Repeatability model assuming homogeneity of variance
and the equal genetic correlation between lactations.
Levene's test checked the homogeneity of variance
. The assumption of homogeneity of variance
was not violated where the P-value was >0.05.
It is also seen that the residuals roughly form a horizontal track around the 0 line, implying that there is homogeneity of variance
Because the number of plants is classified as a discrete quantitative variable, resulting from count data, additivity was tested using the GLM (General Linear Models) procedure, by means of analysis of covariance of the squared predicted values, resulting in p = 0.0596, normality by univariate procedure, with W statistics (ShapiroWilk test), with p = 0.0584 and the homogeneity of variance
by Bartlett's test (p = 0.0624).
Primary analysis was performed to make sure there were no violations of the assumptions of normality, linearity, multicollinearity, and homogeneity of variance
. Because of the high value of multicollinearity between age and years of practice (0.86), the age was excluded from the model.