The table of communalities shows how much of the variance to be considered for further analysis. The communality value needs to be more than 0.5 for consideration. However, the variables Pers17, Pers20 and Pers38 were below 0.5 and therefore they should be flagged.
The reliability of the items was checked through the reliability statistics table. The Cronbach’s alpha yield a result of 0.503. This suggest that there is a relatively high level of internal consistency of the items. However, from the KMO and Bartletts’ Test table, it can be seen that since the Kaiser-Meyer-Olkin (KMO) is more than 0.6, and the Bartlett’s test is also significant, it can be concluded that the correlation matrix is not an identity matrix.
From the total variance explained table, it is evident that 10 factors will be extracted. The 10 factors are the only ones which meet the cut-off criterion of the extraction method. Evidently, these are the only factors with eigenvalues greater than 1. The 10 factors also accumulatively account for 57.53% of the variance in the items' variance-covariance matrix. On the other hand, all the remaining factors are not significant.
The table named Component matrix shows the loadings of the 44 variables on the ten extracted factors. Clearly, it can be seen that some of the items did cross-load. For example pers13 (1 and 4), pers33 (1 and 9), and pers38 (1 and 10). The cross loading of below .3 will be ignored since they were suppressed. However, items with cross-loading with above 0.3 were eliminated.
After the factor analysis, the next step is assigning a score to each component for each participant. Questions which are strongly loaded to a particular component whill be given scores that reflect the subjects of those questions. The scores can be further used for further analyses such as multiple regression.
Influential cases is the presence of outliers. Influential cases need to be checked since their presence greatly affects the line of regression’s slope and the also the coefficient o determination.
The first assumption for linear regression is that the dependent variable should be measured on a scale that is continuous. The dependent variable, resilience mean, is continuous in nature with a mean of 3.6 and a standard deviation of 0.66. Hence, the first assumption has been met.
The second assumption for the regression is that there should be two or more independent variables which can either be categorical or continuous. The second assumption has been met since in this case, the independent variables mindfulness mean, narcissism mean and conscientiousness mean are continuous in nature.
The third assumption for linear regression is that there should be an independence of observations. From the Durbin-Watson statistic on the model summary table, it can be seen that the statistic is 1.879. Since the statistic is between 1.5 and 2.5, it can be anticipated that there is no first order linear auto-correlation in the multiple linear regression data.
The forth assumption is that the data must not show multicollinearity. From the coefficients table, it can be seen that the VIFs are between 1 and 10. Thus, it can be concluded that the forth assumption has been met since there is no multicollinearity.
The fifth assumption is that the residuals should be approximately normally distributed. From the histogram, it is evident that this assumption has been met since the residuals can be seen to be approximately normally distributed.
From the coefficients table, it can be seen that all the predictors (mindfulness mean, narcissism mean and conscientiousness mean) significantly predict resilience since their p-values are less than 0.05.The relationship being tested when mindfulness predicts resilience via conscientiousness is the interaction effect.
From table 1 above, it can be seen that the relationship between mindfulness and resilience through conscientiousness is statistically significant (p <0.05) based on the total effect of X on Y column.
It can be established that there is a statistically significant interaction between mindfulness and conscientiousness on resilience. Thus, the mindfulness and conscientiousness factors play a combined role in influencing resilience of an individual. Consequently, the impact of mindfulness on resilience depends on the level of conscientiousness.Predicting resilience from narcissism through conscientiousness.Table 2 shows that the relationship between narcissism and resilience through conscientiousness is statistically significant (p <0.05) based on Int_1 column.
It is has been observed that there is a statistically significant interaction between conscientiousness and narcissism that impacts resilience. Thus, the narcissism and conscientiousness factors play a combined role in influencing resilience of an individual. Hence, the impact of narcissism on resilience depends on the level of conscientiousness.
Experimental study design will be used since it allows the researcher to test his hypothesis by attaining a valid conclusion regarding the relationships between the dependent and independent variables.It can be seen from the Mauchly’s Test of Sphericity that Mauchly’s Test shows that the sphericity assumption had been violated, p < 0.05. Thus, a correction was needed to be applied using the Huyn-Feldt correction since ε > .75.
From table 3 above regardless of which row we choose to read from, that is the difference between mindfulness training and control, we can see that there isn’t a statistically significant difference between all three different educational levels (p > 0.05).