The first Analysis of Variance I was asked to perform was by hand, before even calculators were easily available.
Not exact matches
Principal Component
Analysis is a translation
of the data so the new origin is the mean and then a rotation so that the
first axis is in the direction which explains the largest amount
of variance, the second axis is the axis orthogonal to the
first that explains the largest amount
of the remaining
variance, the third is the axis orthogonal to the
first two that explains the largest amount
of what remains and so on.
The
analysis highlighted 6 factors (the
first six eigenvalues were 11.4, 4.2, 2.4, 2.2, 1.7, 1.6) accounting for 41.2 %
of the total
variance.
In hierarchical regression
analyses with the various ENRICH factor scores as dependent variables and job satisfaction as the independent variable in the
first block, the two SSQ factors in the second block, and the WOC factors in the third block, between 24 and 38 %
of the
variance in seven
of the nine ENRICH factors (not significant model for «Family & Friends» and «Marriage & Children») could be explained by the variation in all the independent variables with varying weight
of the several independent variables (Table 3).
For the descriptive
analysis of the individuals with a different relationship status and to answer the
first hypothesis, χ2 - tests for independent samples and one - way
analyses of variance were used.
Three nested models with increased degrees
of constraint were compared in multigroup
analyses (fathers versus mothers): We specified a
first model
of configural invariance, in which the parameters (factor loadings, item intercepts, residual
variances, factor
variances, and covariance) were freely estimated in each group, whereas the factor means were constrained to zero in both groups.