ANOVA Case Study: Assignment
Case Study: ANOVA (Taken from Predictive HR Analytics, p. 216)
Learning Question: Do engagement levels differ among countries?
The one-way ANOVA test is a powerful method for testing the significance of the difference between sample means where more than two categorical conditions/groups are compared. It effectively tests the null hypothesis that all the samples are drawn from the same population and that there is no difference between them. The data set used is ‘Chapter 6 Turnover team DATA.x’ and the data variables include:
- Team identifier (unique team number).
- Team size (number of people in the team).
- Team turnover 2014 (separation rate between 0 and 1).
- Country (1 = UK; 2 = United States; 3 = Canada; 4 = Spain).
- SURVEY: ENGAGEMENT items COMBO (composite engagement percentage across the team).
- SURVEY: TeamLeader Rating (composite team leader percentage across the team).
- SURVEY: CSR rating (composite corporate social responsibility percentage across the team).
- SURVEY: Drive for Performance (composite percentage team score on perceived ‘Drive for performance’ percentage).
- SURVEY: Performance, Development and Reward (composite percentage team score on perceived fairness of performance, development and reward across the team).
- SURVEY: Work–Life Balance (composite percentage team score on perceived work–life balance across the team).
Here we are going to explore the engagement differences between the four countries in the data set (the UK, United States, Canada and Spain). The data set includes 2,910 employees employed across four countries organized in 212 teams. We have team-level engagement survey data provided by our external survey company; for each team we have an average positive response score from a 2013 engagement survey that measured scales of engagement.
Following the Excel guide starting on p. 385 (Levene et al., 2021), run a One-Way ANOVA and the Tukey-Kramer procedure to answer, are there country differences in team engagement?
Deliverables:
- Using the excel worksheet, perform an ANOVA with the Tukey Kramer analysis. I have given the formula for conducting an ANOVA on work life balance. You will need to adjust the formula to analyze engagement. You will also need to fill in the critical range of the Q statistic using Table E.7.
- Then, write up a sample paragraph (5-10 sentences) to the senior management stating the null and alternate hypotheses, explaining your method, providing your findings, and suggesting 2-3 organizational policies or practices based on the results and relevant literature.