April 29, 2022

“Everyone talks about building a relationship with your customer. I think you build one with your employees first.”

– Angela Ahrendts (Senior Vice President, Apple)

“Before you become a leader, success is all about growing yourself. After you become a leader, success is about growing others.”

– Jack Welch (CEO, GE)

“It’s not how much money we make that ultimately makes us happy between 9 and 5. It’s whether or not our work fulfills us.”

– Malcolm Gladwell (Bestselling Author)

Brilliant on the Basics

No alt text provided for this image

This quote captures the essence of Atul Gawande’s important work, The Checklist Manifesto.  

For many of us, the complexity in what we do on a day-to-day basis has increased dramatically over many decades. Whether you are firefighter tackling a large multi-story fire or you’re onboarding a new hire, there is simply too much information for us to reliably use our memories. Gawande teaches us about the power of checklists in capturing and simplifying everything we are expected to do, manage, or verify. 

Here are some of the key ideas: 

  1. Checklists protect us against failure; they help us avoid missing a key step or sequence in a process (e.g., whether certain medicine has already been given to a patient).  
  2. Checklists help us define a higher standard of performance and ensure we are delivering the best services or products. 
  3. Checklists are an aid, and they need to be useful. If they aren’t useful, then the checklist is not right – more work is needed in defining the checklist. 

Pro tip: Think about how you can harness the power of checklists in your work. Importantly, think about how you can use them to design and build the right employee experience for your team or organization.  

Read more about how you can implement best practices from The Checklist Manifest here at the Farnam Street blog. We highly recommend Farnam Street as a practical, yet well-researched resource for business leaders at all stages in their career.

Using Statistics to Refine our Understanding: A Series on Good Survey Methodology and Statistics (Part 2) 

Last newsletter, we talked about some best practices in survey design and administration. This week, we will focus on a few standard statistical tests we use to help us evaluate and clean our data.  

  • Correlation. Correlation is used to measure the linear relationship between survey items and/or outcome variables (e.g., engagement, intent to stay, etc.) As most of you know, correlation does not mean the same thing as causation. Correlation simply means there is a relationship of some kind between two variables. For example, there is a known correlation between height and foot size. Often, the taller the person is, the larger their feet. This is what we call a positive correlation. An increase in one variable (height) means an increase in the other variable (feet size).  

Sometimes, however, correlation doesn’t help much. For example, arcade center revenue is highly correlated to increases in the number of computer science doctorates that are awarded in any given year.  

Correlation can be tabulated and the strength categorized as follows: 

No alt text provided for this image

credits: Parvez Ahammad

  • Regression Analysis. Regression is used to determine how one or more variables affect an outcome. 

“Regression analysis is a set of statistical methods used for the estimation of relationships between a dependent variable and one or more independent variables. It can be utilized to assess the strength of the relationship between variables and for modeling the future relationship between them.” 

In our case, regression helps us understand the “drivers” of employee engagement. What aspects of the employee experience make the most difference? Is the relationship between a manager and the employee or is it the size of the benefits package (hint: it’s all about the manager).

  • T-Test. In survey science, this test is used to see if two different groups really scored differently on an item and/or outcome variable, or whether the difference could also be explained as mere chance. In other words, are the differences real?  

Example: Manufacturing 

“A manufacturing engineer wants to know if some new process leads to a significant improvement in mean battery life of some product. To test this, he measures the mean battery life for 50 products created using the new process and performs a one sample t-test to determine if the mean battery life is different from the mean battery life of products made using the current process.” 

The key question is whether the new manufacturing process made an actual improvement to battery life or could the differences be just as likely due to normal variances in the manufacturing process?  

  • ANOVA. This test is used to see if three or more different groups scored differently on an item and/or outcome variable. Similar to the other statistical tests we’ve discussed, there are several ways to use an Anova. For our purposes, Anova and t-tests are siblings in that they perform similar functions. Whereas the t-test identifies significant differences in the means of two groups, Anova’s identify significant differences in the means of three or more groups.  

Organizational surveys often have populations that can be categorized into three or more groups such as tenure bands, management levels, age groups, and departments. Having a statistical test that assesses the significance of differences across all groups at the same time is useful in finding populations that are especially high or low. Organizational leaders are often interested in identifying over and under scoring populations, which is exactly what an Anova test can help us do.     

Of course there are many more statistical tests and tools that we use each day, but these are four of the most helpful. For more information on how we use statistical science to inform our research and products, please visit us at www.decisionwise.com

What’s Happening at DecisionWise


Last week Dave Long and Charles Rogel presented on employee experience trends for 2022 where they presented some compelling insights and data on what to expect in the employee experience space this year. Give it a watch if you missed it live. We also have another webinar coming up this month. Stay tuned for more details!


HR News Roundup 

  • As we embrace work-tech and HR technology at an even more rapid pace, HR professionals need to know how to manage sophisticated technology implementations and digital transformations.  Here are 4 tips to improve your track record.  
  • Once again, we find that non-compensation based elements can be more important than pay. In this article, employee recognition is shown to be more important than salary.   
  • HBR recommends leaving the door wide open for boomerang employees. What do you think? 

Recommended Posts