Statistics employee retention case,
the questions are the following
1. Assume that the population proportion of employees that had left the company prior to the time this dataset was established was 20%. Suppose that you surveyed 117 employees, what is the probability of observing that 40% or more left company if the true population proportion is 20%? Is it reasonable to assume that this could be likely, or does it appear that a 40% exodus would be unusual?
2. Given the survey data that was collected, calculate the 95% confidence interval for the proportion of employees that left the company. Interpret this interval. If you believed that the true population proportion of employees that left the company is 20%, is there any evidence to suggest that you were wrong?
3. Create a classification tree with “Left the Company” as your Y. Split the tree several times so that you can clearly see if there are any continuous variables that might be a surrogate measure for whether a person left the company or not. Cut and paste your tree and report on what that continuous variable is.
4. Calculate the 95% confidence interval for the satisfaction rating. Interpret this interval. If you believed that the satisfaction rating was .80, is there any reason to believe that you are wrong?
5. With Satisfaction Rating as your Y, test to see if the number of projects might be related to the mean Satisfaction Rating. In order to do this, place all of the employees that had 6 or more projects in one group and all those that had 5 or less in another group. Then see if there is a significant difference in Satisfaction Rating when comparing the two groups.
6. With the Satisfaction Rating as your Y, test to see if whether a person that had an accident results in a significant lesser Satisfaction Rating than an employee that did not have an accident on the job.
7. Split the data set into two groups, one with the number of years at 5 or more, and the other group with 4 or less. Using these two groups, test to see if the mean Satisfaction Rating of those with a longer tenure is different than the mean Satisfaction rating of the lesser tenure.
8. Perform a Graphical Summary (Stat>Basic Stats>Graphical Summary) of the Monthly hours data. Report on the mean, standard deviation and shape of the data.
9. If you were to split the data set based on 195 hours, is there any evidence to suggest that the number of hours worked has an influence on satisfaction rating? Test to see if the mean Satisfaction Rating of those working more hours is significantly less than those working less than 195 hours.
10. For the scenario in Problem 9, perform a hypothesis test to determine if there is any evidence to suggest that the variability in Satisfaction Rating for the employees that worked less than 195 hours is different than the variability in Satisfaction Rating of the folks that worked more than 195 hours.