Question 1:

- Communality for wQ1=(loading on factor 1)
^{2 }+ (loading on factor 2)^{2}

=0.468^{2}+0.138^{2}

=0.238

- The communality for a wW1 is 0.606 which means the proportion of variation in variable wW1 explained by the two factors is 60.6%.
- Total available variance = number of variables used in the analysis=14
- Total amount of variance accounted for by two factors after rotation=53.005%
- As most of the variability is explained by just first factor so we need not use two factors. That is why researcher decided to use only one factor. Also point of inlexion is second so the number of factors is 1.
- The goodness of fit test is significant indicating that the correlation matrix is different from identity matrix, therefore, there are some relationships between the variables we hope to include in the analysis and hence factor analysis is justified. The three factors explain 58.043% of the variance in variables which is quite moderate and hence factor analysis does not seem to be adequate. Also, the communality matrix shows that the factors explain less than 0.40 variations in wQ1 and wQ2 which is again not enough. Also, looking at patern matrix it is observed that loadings are very small for wl2 for each of the three factors. So factor analysis is not appropriate.
- The person with higher values for questions wl4,wl3,wl5,wl6 , wl1 and wl2 would score high on factor 1 as these questions have higher positive loading by factor 1.
- The person with lower values for questions ww4,ww3,ww1,ww2 would score high on factor 2 as these questions have negative loading by factor 2.
- The correlation between factor 1 and factor 2 is -0.699 which means that these factors are not independent and that this constitute of items which are negatively related to each other, that is items in factor 1 are negatively related to factor 2 items.

Question 2:

Hypotheses are ;

H1: there is a significant difference in the coping assessment for three measure avoidance coping , behavioural coping and cognitive coping which were measured before and after stress condition and one month later.

H2: the three measures differ in first and second coping , second and third coping but do not differ in first and third coping.

Descriptive statistics shows that:

- avoidance coping increases in the second stage after stressful incidence the average coping ability increases while in the third stage lower down
- behavioural coping increases in the second stage after stressful incidence the average coping ability increases and also in the third stage increase is observed
- cognitive coping remains almost same in the second stage after stressful incidence while in the third stage increase very much.

We have repeated measure MANOVA with no between subject variable, so from table of tests of within-subjects Effects we observe that all the four tests lead to rejection of the null hypothesis that there is not a significant difference in the coping assessment for three measure avoidance coping , behavioural coping and cognitive coping which were measured before and after stress condition and one month later as p<0.001 for each test.

Hence there is a significant difference in the coping assessment for three measure avoidance coping , behavioural coping and cognitive coping which were measured before and after stress condition and one month later.

As the results are significant so post hoc repeated measure ANOVA in Univariate test Table were applied and showed that (The Mauchly test is significant for each coping measure avoidance coping (p<0.0001), behavioural coping (p<0.001) and cognitive coping (p<0.001) indicating that Sphericity cannot be assumed.

- there is significant difference in the mean avoidance coping in three point of times ( before stress after stress and a month later) as Greenhouse –Geisser test is significant , p=0.001
- there is significant difference in the mean behavioural coping in three point of times ( before stress after stress and a month later) as Greenhouse –Geisser test is significant , p<0.001
- there is significant difference in the mean cognitive coping in three point of times ( before stress after stress and a month later) as Greenhouse –Geisser test is significant , p<0.001

To test for second hypothesis H2, the contrasts were tested and the results are in table of Tests of within-subject contrasts. The results showed that

- for avoidance coping the mean coping after stress (level 2) is not significantly different than before stress (level 1) as p=0.193>0.05 while the mean the mean coping a month after stress (level 3) is significantly different than before stress (level 1) as p=0.012<.05. this is contrary to hypothesis H2 as we expected the increase in coping after stress and a month later the coping should reduce to first level before stress.
- for behavioural coping the mean coping after stress (level 2) is significantly different than before stress (level 1) as p<0.001 and also the mean coping a month after stress (level 3) is significantly different than before stress (level 1) as p<0.001. this is contrary to hypothesis H2 as we expected the increase in coping after stress and a month later the coping should reduce to first level before stress.
- for cognitive coping the mean coping after stress (level 2) is not significantly different than before stress (level 1) as p=0.738>0.05 while the mean the mean coping a month after stress (level 3) is significantly different than before stress (level 1) as p<0.001. this is contrary to hypothesis H2 as we expected the increase in coping after stress and a month later the coping should reduce to first level before stress.

Question 3:

- partial regression coefficient for gender at stage 1 is 2.152 which means that holding fear constant on an average the productivity of females is 2.152 more than that of males.
- 95% confidence interval for the regression coefficient for gender at stage 1 is (0.278 , 4.025) which means that on an average holding fear constant, the female are 0.278 to 4.025 more productive than males.
- The partial correlation coefficient for FEAR at stage 1 is -0.643 which shows that after controlling for gender there is high negative correlation between fear and productivity.
- The partial correlation coefficient for FEAR at stage 2 is -0.32 which shows that after controlling for gender and job satisfaction there is weak negative correlation between fear and productivity.
- The R square change ar stage 2 of the regression is 0.156 which means that addition of job satisfaction to predictor list leads to increase of 15.6% more variability of the productivity explained by the model.
- The change in the R square at stage 2 is significant as F
_{change}(1, 373)=136.388, p<0.001. - From the analysis of variance table in stage 2 we observe that the regression model is significant after adding job satisfaction to the predictor list in addition to gender and fear as F(3, 373)=166.655, p<0.001.
- The part correlation coefficient for fear at stage 2 is -0.230 which means that if we remove fear from the regression then the R square decrease by (-0.230)
^{2}=0.053 which is very low indicating no use of fear almost. - Tolerance of gender is 0.999 which means that proportion of variability in gender not explained by fear and job satisfaction is 99.9%.

Tolerance of fear is 0.576 which means that proportion of variability in fear not explained by gender and job satisfaction is 57.6%.

Tolerance of job satisfaction is 0.576 which means that proportion of variability in job satisfaction not explained by fear and gender is 57.6%.

As all tolerance levels are greater than 0.1 so there is no problem of multicollinearity.

- Standardized coefficient of job satisfaction is 0.521 which is more than standardized coefficients of gender (0.089) and fear (-0.303). So, job satisfaction has more effect on productivity than gender and fear.
- As fear has high standardized coefficient -0.642 in stage 1 and decreased in stage 2 to -0.303, so it follows that fear has indirect impact on prductivity through job satisfaction. However, the standardized coefficient 0.089 of gnder does not change from stage 1 to stage 2 so gender has no indirect impact on prductivity through job satisfaction.
- Hitogram is symmetric and normal P-P plot points are in a straight line representing normal distribution so the standardized residuals shows normality of residuals. The points in the scatter plot of standardized residuals versus the standardized predicted values are randomly distribtued around the line through zero residual indicating assumtpion of homoscedasticity of variances of errors is satisfied.

Critical value of Mahalanobis distance for three idependent variables is 16.3.

As the calculated values of Mahalanobis distance have maximum value 18.406 which is greater than critical value 16.3, so we cocnldue that there are mutlivariate outliers in the data. Cook’s distance values are not grwater than 1 so there are no influential points in data.

- As the interaction between age and job satsifaction is signficant (b=-0.612, t=-2.513, p=0.012<0.05) and negative so it follows that age moderate the realtinship between job satisfaction and job performance and that the job satisfaction have greater impact on performance in the case if people under 30 years than in case of older people as coefficient of interaction is negative.

Question 4:

- As child adjustment is affected by parenting and which in turn is affected by marital conflict, so in this case linear regression with child adjustment as dependent variable and marital conflict tested first alone and then with parenting, so here hierarchical multiple regression with marital conflict in first stage and in addition the parenting in the second stage will be used as predictors..
- As we have three variables of personality, namley, neuroticism , extroversion and conscientiousess meaures on same children and one factor of four level (of four ultures: Australia, China, Japan, USA), so here single factor independent groups (between-subjects) MANOVA will be used.
- As we have two personal adjustment measures self-esteem and depression scores measured before, after six and after twelve months intervention program, so here we will use single factor repeated measure (witnin-subjects) MANOVA.
- Here psychological well being is dependent variable, and family conflict and family status are independent variable with an interaction of familty conflict and familty status to know the influence on young people of divorced family. So, here linear regression involving interaction term will be appropriate.