Your work over the past four months at the company has been quite beneficial for the organization. The executive team appreciates the insights that you have provided around employee engagement, commitment, burnout, and turnover. They have been so impressed that they have come to you with another request.
Turnover is costly for the organization. Hiring alone costs over $2,000,000 annually. This is a conservative estimate for the cost of turnover as this does not even account for lost sales or productivity. The executive team calls you into a meeting and says: we would like an analysis of employee turnover drivers/predictors and a set of recommendations to reduce our annual turnover cost. You think to yourself about the interesting predictors you’ve already explored and also consider some archival data, such as compa-ratio, tenure, and job title.
Your manager asks to prepare a plan for this project that addresses all of the following:
· What is the research question?
· What is your theoretical explanation/model?
· What are your null and alternative hypotheses?
· What sample would you like to use to test your hypotheses, and how many observations do you need?
· How will you operationalize the drivers from your model? (explain survey item construction for any survey variables)
· How will you determine the reliability of your variables? (what kind of reliability and which statistical test?)
· How will you determine the validity of your variables? (what kind of validity and which statistical test?)
· Which statistical test(s) will you use to test your hypotheses? In which software program will you run these tests and how?
· Assume that you get significant results. What is your interpretation and how will you explain it to the non-technical exec team?
· Given your significant findings, what recommendations would you provide to address the turnover costs?
