People Analytics Deconstructed
Are you responsible for understanding an employees’ experience? Have you tried to incorporate people analytics in your organization but have struggled? Have you ever wondered what it means to have a data culture? Would you like to make more data-driven decisions? These are the kinds of discussions you can expect to hear on People Analytics Deconstructed. Co-hosts Ron Landis and Jennifer Miller are co-founders of Millan Chicago, a data science consulting company dedicated to helping organizations make the most out of their data. Each week, they will ‘deconstruct’ modern and contemporary topics in the People Analytics space.
People Analytics Deconstructed
Analytics in Practice: Using Factor Analysis to Evaluate Engagement Surveys, Part 1
•
Millan Chicago
•
Season 1
•
Episode 25
In this episode, co-hosts Jennifer Miller and Ron Landis continue their conversation about developing a measure of employee engagement. In this episode they focus on how to use a statistical technique called factor analysis to examine the dimensions of employee engagement.
In this episode, we had conversations around these questions:
- What is factor analysis?
- What is the difference between exploratory factor analysis (EFA) and confirmatory factor analysis (CFA)?
- What is the difference between an EFA and principal components analysis (PCA)?
- What format should the data be in to complete the EFA?
- How does EFA work?
- What is a loading matrix?
Key Takeaways:
- A factor analysis is useful to determine whether you are measuring what you intend to measure with a survey. We continue our example of measuring engagement with four dimensions (satisfaction with manager, satisfaction with co-workers, satisfaction with compensation, satisfaction with working conditions).
- There are three main aspects to conducting an EFA. First, you need to decide on the type of analysis (I.e., PCA, EFA). Second, you need to rotate the solution. Third, you need to interpret the results. In this episode, we cover the first and part of the second steps in EFA.
- We also discussed the concept of a loading matrix. First, each item is correlated with each factor. Each correlation can be squared to get the percentage of variance explained. Second, the sum of all the squared values down a column are computed, which is the eigenvalue. Third, communality is determined by summing across the row for each item. Finally, the uniqueness can be computed by 1-communality.
- The ultimate goal is to associate each item with a factor. The initial solution will almost never allow us to see the underlying structure. The concept of rotation was briefly mentioned and is covered in additional detail in the next episode.