🏢 Drivers of employee engagement
Research question: Which factors predict employee engagement when considered together? Answer: A multiple linear regression showed that the predictors together explained 46.1% of the variance in engagement (1–5), F(4, 175) = 37.42, p < .001. Significant predictors were Satisfaction with pay (1–5) (β = .27, p < .001), Manager support (1–5) (β = .47, p < .001), Workload (1–5) (β = −.20, p < .001), Training hours last year (β = .22, p < .001).
Open this example in StatLab Dataset and codebook (CSV, Excel, .sav)
What is the situation?
An HR survey (fictional company, 180 staff) measured engagement and four possible drivers: satisfaction with pay, manager support, workload (all 1–5 scale scores) and training hours last year. The data are synthetic — produced by a seeded random process for teaching; they describe no real people or organisations.
What is in the dataset?
180 cases.
| Variable | Label | Measure | Value labels |
|---|---|---|---|
id | Employee ID | scale | |
pay_sat | Satisfaction with pay (1–5) | scale | |
mgr_support | Manager support (1–5) | scale | |
workload | Workload (1–5) | scale | |
training_hrs | Training hours last year | scale | |
engagement | Engagement (1–5) | scale |
How is it analysed, step by step?
- Descriptives — Describe every variable in the model. (guide)
Engagement (1–5): M = 2.74, SD = 0.58, range 1.45–4.93, skewness 0.32
Satisfaction with pay (1–5): M = 2.97, SD = 0.81, range 1–5, skewness 0.17
Manager support (1–5): M = 3.30, SD = 0.81, range 1.32–5, skewness −0.07
Workload (1–5): M = 3.33, SD = 0.73, range 1.61–5, skewness 0.14
Training hours last year: M = 19.44, SD = 10.47, range 0–51, skewness 0.18 - Correlations — Bivariate relationships, and a first look at collinearity between predictors. (guide)
Engagement (1–5) – Satisfaction with pay (1–5): r = .43, p < .001
Engagement (1–5) – Manager support (1–5): r = .55, p < .001
Engagement (1–5) – Workload (1–5): r = −.14, p = .055
Engagement (1–5) – Training hours last year: r = .27, p < .001
Satisfaction with pay (1–5) – Manager support (1–5): r = .36, p < .001
Satisfaction with pay (1–5) – Workload (1–5): r = .06, p = .419
Satisfaction with pay (1–5) – Training hours last year: r = −.00, p = .950
Manager support (1–5) – Workload (1–5): r = .14, p = .062
Manager support (1–5) – Training hours last year: r = .04, p = .570
Workload (1–5) – Training hours last year: r = −.14, p = .068 - Linear regression — Each driver’s unique contribution, holding the others constant. (guide)
R² = .461, adjusted R² = .449, F(4, 175) = 37.42, p < .001, Durbin-Watson = 2.05.
(Constant): B = 1.324, SE = .216, t = 6.13, p < .001
Satisfaction with pay (1–5): B = .195, SE = .043, β = .27, t = 4.58, p < .001, VIF = 1.15
Manager support (1–5): B = .339, SE = .043, β = .47, t = 7.81, p < .001, VIF = 1.17
Workload (1–5): B = −.156, SE = .045, β = −.20, t = −3.45, p < .001, VIF = 1.04
Training hours last year: B = .012, SE = .003, β = .22, t = 3.93, p < .001, VIF = 1.02
How would I write it up (APA style)?
A multiple linear regression showed that the predictors together explained 46.1% of the variance in engagement (1–5), F(4, 175) = 37.42, p < .001. Significant predictors were Satisfaction with pay (1–5) (β = .27, p < .001), Manager support (1–5) (β = .47, p < .001), Workload (1–5) (β = −.20, p < .001), Training hours last year (β = .22, p < .001).
Every number on this page was computed by StatLab's statistics engine from this dataset when the page was built, and the same analyses are part of the reference validation against scipy/statsmodels (433 reference test cases, 36,554 values checked against scipy/statsmodels/numpy, 0 failures). “Open this example” loads the data and runs these steps live in your browser.