6 <The date

6

Human Resource Management Project

This summary aims to summarize the three main steps of the analysis of the data sample of Wolfpack Widgets, collected in the engagement survey. The survey was used to determine if the engagement characteristics might work as determinants of employee behaviors. The analysis started from the evaluation of the inter-correlations among engagement measures, such as affective commitment, engagement, social integration, and burnout. The analysis revealed that affirmative commitment positively and strongly associated with engagement (r = .70) and social integration (r = .75), and negatively and weakly associated with burnout (r = -.36). There is a moderate positive association between social integration and engagement (r = .54), and negative moderate association between social integration and burnout (r = -.42). Finally, social integration is negatively and moderately associated with burnout (r = -.40). These findings show that employees, who have higher affirmative commitment, usually are more engaged, and more socially integrated. Such employees have lower burnout rates. On the other hand, employees with high level of burnout are less affirmatively committed, less engaged, and less socially integrated.

The next step of the analysis was to explore how predictors (engagement measures) and outcomes (employee behaviors) are related to each other. The correlations can be interpreted as in the step above – they are provided in the appendix. In addition, eight scatter plots were created for four outcome variables. Among the relationships between outcome and predictor variables, retention is mostly correlated with engage (r = .46) and the weakest correlation was with social integration (r = .28). Affirmative commitment after intervention was mostly correlated with engagement (r = .69) and the weakest correlation was with initial burnout (r = -.39). Burnout after intervention was mostly correlated with initial burnout (r = .67) and least correlated with initial affirmative commitment (r = -.33). Finally, the highest correlation with turnover was observed in initial affirmative commitment (r = -.16), and least correlated with initial social integration (r = -.0003).

The final part of the analysis was conducting a multiple linear regression analysis to see whether the set of four independent variables can significantly predict outcomes. Four regression models were created – one per each outcome variable. The model created for retention showed that the coefficients of the equation are jointly significant, F = 9.14, p < .001. The linear combination of four predictors explain about 20.16% of the variability in retention. However, only initial engagement appeared to be an important determinant of retention (t = 3.495, p < .001). The second model was aimed to predict affirmative commitment after intervention. The model was significant overall, F = 48.21, p < .001. The model explains about 52.33% of the variance in affirmative commitment after intervention. Such factors as initial engagement (t = 7.00, p < .001) and social integration (t = 3.537, p < .001) were important determinants of affirmative commitment. The burnout regression was significant, F = 8.379, p < .001. The model explains only 14.65% of the variance in burnout after intervention. At the 5% level of significance, only engagement at baseline was an individually significant predictor (t = -2.241, p = .027). Finally, a binary logistic regression analysis was performed in R to estimate the factors that are important in predicting turnover. The analysis showed that only initial affirmative commitment is an important determinant of turnover (z = -2.142, p = .032), with higher commitment associated with lower chance of turnover.

Appendix

Step 1.

 

affcom.t1

engage.t1

soc.int.t1

burnout.t1

affcom.t1

1.00

engage.t1

0.70

1.00

soc.int.t1

0.75

0.54

1.00

burnout.t1

-0.36

-0.42

-0.40

1.00

Step 2.

 

affcom.t1

engage.t1

soc.int.t1

burnout.t1

retention.t2

affcom.t2

burnout.t2

turnover

affcom.t1

1.00

engage.t1

0.70

1.00

soc.int.t1

0.75

0.54

1.00

burnout.t1

-0.36

-0.42

-0.40

1.00

retention.t2

0.35

0.46

0.28

-0.28

1.00

affcom.t2

0.67

0.69

0.57

-0.39

0.50

1.00

burnout.t2

-0.33

-0.36

-0.35

0.67

-0.15

-0.37

1.00

turnover

-0.16

-0.11

0.00

0.04

-0.36

-0.16

-0.05

1.00

Step 3

Retention

Regression Statistic

Multiple R

0.475785

R-squared

0.226372

Adjusted R-squared

0.201615

Standard error

0.772421

Observations

130

ANOVA

 

df

SS

MS

F

Significance F

Regression

4

21.82265

5.455663

9.144068

1.64E-06

Residual

125

74.57927

0.596634

Total

129

96.40192

 

 

 

 

Coefficients

Standard error

t-statistic

P-Value

Lower 95%

Upper 95%

Y-intercept

2.240345

0.734919

3.048423

0.002808

0.785848

3.694841

affcom.t1

0.037339

0.154613

0.2415

0.809564

-0.26866

0.343337

engage.t1

0.489201

0.139961

3.495256

0.000656

0.2122

0.766202

soc.int.t1

-0.00787

0.187364

-0.04202

0.966548

-0.37869

0.362943

burnout.t1

-0.10028

0.083589

-1.19966

0.232541

-0.26571

0.065155

Affirmative Commitment

Regression Statistic

Multiple R

0.731045

R-squared

0.534427

Adjusted R-squared

0.523342

Standard error

0.546977

Observations

130

ANOVA

 

df

SS

MS

F

Significance F

Regression

3

43.27213

14.42404

48.21139

8.03E-21

Residual

126

37.6971

0.299183

Total

129

80.96923

 

 

 

 

Coefficients

Standard error

t-statistic

P-Value

Lower 95%

Upper 95%

Y-intercept

0.225866

0.505497

0.446819

0.655773

-0.7745

1.226229

engage.t1

0.588308

0.08409

6.996189

1.38E-10

0.421897

0.754719

soc.int.t1

0.367822

0.104001

3.536712

0.000568

0.162007

0.573637

burnout.t1

-0.05197

0.059114

-0.87915

0.380994

-0.16895

0.065014

Burnout

Regression Statistic

Multiple R

0.407828

R-squared

0.166324

Adjusted R-squared

0.146475

Standard error

0.889081

Observations

130

ANOVA

 

df

SS

MS

F

Significance F

Regression

3

19.87061

6.623536

8.379288

4.02E-05

Residual

126

99.59862

0.790465

Total

129

119.4692

 

 

 

 

Coefficients

Standard error

t-statistic

P-Value

Lower 95%

Upper 95%

Y-intercept

5.468385

0.631395

8.660796

1.91E-14

4.218872

6.717898

affcom.t1

0.035581

0.177729

0.200196

0.84165

-0.31614

0.3873

engage.t1

-0.34971

0.156074

-2.24069

0.026798

-0.65858

-0.04085

soc.int.t1

-0.40107

0.210673

-1.90378

0.059219

-0.81799

0.01584