Assignment 8: t-test - Employee Attrition Analysis

- T-Test number 1

hr$left <- as.factor(hr$left)

t_test_result <- t.test(satisfaction_level ~ left, data = hr) 

print(t_test_result)

## 
##  Welch Two Sample t-test
## 
## data:  satisfaction_level by left
## t = 46.636, df = 5167, p-value < 2.2e-16
## alternative hypothesis: true difference in means between group 0 and group 1 is not equal to 0
## 95 percent confidence interval:
##  0.2171815 0.2362417
## sample estimates:
## mean in group 0 mean in group 1 
##       0.6668096       0.4400980

- P-value Interpretation: The p-value is very small, <.05, so the satifaction level of employees is significant

- T-test interpretation: The range of 0.2171815 to 0.2362417 represents the difference in means with 95% confidnce. The fact that the range does not include zero reinforces the significance of satisfaction levels.

- Non-technical interpretation: Employees with higher satisfaction are more likely to stay.

hr <- hr %>%
  mutate(left = factor(left, levels = c(0, 1), labels = c("Stayed", "Left")))

plot_ly(hr, y = ~satisfaction_level, color = ~left, type = "box", 
        colors = c("Stayed" = "green", "Left" = "red")) %>%
  layout(
    title = "More Satisfied Employees are More likely to stay",
    yaxis = list(title = "Level of Satisfaction"),
    xaxis = list(title = "Employment Status")
  )

- T-Test Number 2

t_test_evaluation <- t.test(last_evaluation ~ left, data = hr)

print(t_test_evaluation)

## 
##  Welch Two Sample t-test
## 
## data:  last_evaluation by left
## t = -0.72534, df = 5154.9, p-value = 0.4683
## alternative hypothesis: true difference in means between group Stayed and group Left is not equal to 0
## 95 percent confidence interval:
##  -0.009772224  0.004493874
## sample estimates:
## mean in group Stayed   mean in group Left 
##            0.7154734            0.7181126

- P-value Interpretation: The p-value is large, >.05, so the last evaluation of employees is not significant

- T-test interpretation: the p value is .4683, so that means there is a 46.83% chance of observing a difference as or more extreme than this one based off random chance

- Non-technical interpretation: The is no significance to the last evaluation the employee recieves

plot_ly(hr, y = ~last_evaluation, color = ~left, type = "box", 
        colors = c("Stayed" = "green", "Left" = "red")) %>%
  layout(
    title = "Last Evaluation doesn't have an effect on staying or leaving ",
    yaxis = list(title = "Last Evaluation Score"),
    xaxis = list(title = "Employment Status")
  )

- T-Test Number 3

t_test_hours <- t.test(average_montly_hours ~ left, data = hr)

print(t_test_hours)

## 
##  Welch Two Sample t-test
## 
## data:  average_montly_hours by left
## t = -7.5323, df = 4875.1, p-value = 5.907e-14
## alternative hypothesis: true difference in means between group Stayed and group Left is not equal to 0
## 95 percent confidence interval:
##  -10.534631  -6.183384
## sample estimates:
## mean in group Stayed   mean in group Left 
##             199.0602             207.4192

- P-value Interpretation: The p-value is very small, <.05, so the average monthly hours of employees is significant

- T-test interpretation: The range of -10.534631 to -6.183384 represents the difference in means with 95% confidnce. The fact that the range does not include zero reinforces the significance of satisfaction levels.

- Non-technical interpretation: Employees with higher average monthly hours are more likely to leave.

plot_ly(hr, y = ~average_montly_hours, color = ~left, type = "box", 
        colors = c("Stayed" = "green", "Left" = "red")) %>%
  layout(
    title = "Employees with higher Average Monthly Hours are more liekly to leave",
    yaxis = list(title = "Average Monthly Hours"),
    xaxis = list(title = "Employment Status")
  )

- T-Test Number 4

t_test_time_spent <- t.test(time_spend_company ~ left, data = hr)

print(t_test_time_spent)

## 
##  Welch Two Sample t-test
## 
## data:  time_spend_company by left
## t = -22.631, df = 9625.6, p-value < 2.2e-16
## alternative hypothesis: true difference in means between group Stayed and group Left is not equal to 0
## 95 percent confidence interval:
##  -0.5394767 -0.4534706
## sample estimates:
## mean in group Stayed   mean in group Left 
##             3.380032             3.876505

- P-value Interpretation: The p-value is very small, <.05, so the time spent at the company is significant

- The difference in mean of staying and leaving based off time spent at the comapny is significant, where the difference in MPG is at least .4534706 years

- Non-technical interpretation: Employees with higher average monthly hours are more likely to leave

plot_ly(hr, y = ~time_spend_company, color = ~left, type = "box", 
        colors = c("Stayed" = "green", "Left" = "red")) %>%
  layout(
    title = "Employees who stayed longer at the company are more likely to leave",
    yaxis = list(title = "Time Spent at Company (Years)"),
    xaxis = list(title = "Employment Status")
  )