People Analytics Deconstructed

In this episode, co-hosts Ron Landis and Jennifer Miller deconstruct the concept of a flight risk model. They focus on how these types of models can be used to predict the degree to which employees are at risk of leaving an organization.  

In this episode, we had conversations around these questions:  

• What is a flight risk model?  
• Why are flight risk models important?  
• How do you build a flight risk model?  
• What are some clear steps that HR professionals can take to build a flight risk model?  

Key Takeaways:  

• Employee attrition has a variety of consequences. Organizations want to predict who is most likely to leave so that they can better forecast future staffing needs, intervene as necessary to enhance retention, and/or estimate dollar costs associated with predicted attrition.  
• A flight risk model determines the employee characteristics, job characteristics, and organizational characteristics that relate to whether an employee voluntarily leaves an organization.  
• There are different analytic approaches to developing flight risk models (with differing strengths and weaknesses). It is essential to choose the approach that is most appropriate for the given situation and to assess the accuracy of the model for making predictions. 
• At the end of the episode, Jennifer and Ron recommend steps for folks just starting out in this space all the way to the more advanced HR professional.  

Related Links

• Millan Chicago
• Bureau of Labor Statistics 
• Attrition Costs
• SHRM Study

In the third episode of a special three part mini-series about measurement, co-hosts Ron Landis and Jennifer Miller discuss how validity of measurement is critical in People Analytics.  

 In this episode, we had conversations around these questions:  

• What is validity?  
• Why is it is important to consider the validity of measures?  
• What are the different types of validity?  
• How do I determine the validity of a measure?   
• What are some of the steps that HR professionals can take to assess the validity of the assessments used in their organization?  

Key Takeaways:  

• In order to measure what we intend to measure, we must ensure that our measures are valid. One of the most critical questions when selecting or developing a new measure or metric is whether that measure accurately represents what we intend to measure. But how do we make this judgement?  
• There are three approaches to validity including content validity, construct validity and criterion-related validity.  
• Different types of validity require different types of data collection efforts but the central idea is the same. Ron and Jennifer discuss ways in which we can assess the validity of any measure that we use.  
• At the end of the episode, Jennifer and Ron recommend steps for folks just starting out in this space all the way to the more advanced HR professional.  

 Related Links  

• Millan Chicago 

In the second episode of a special three part mini-series about measurement, co-hosts Ron Landis and Jennifer Miller discuss how reliability of measurement is critical in People Analytics.  

In this episode, we had conversations around these questions:  

• What is reliability?  
• Why is it is important to consider the reliability of measures?  
• What are the different types of reliability?  
• How do I determine the reliability of a measure?   
• What are some of the steps that HR professionals can take to assess the reliability of the assessments used in their organization?  

Key Takeaways:  

• In order to measure something accurately, we must ensure that our measures are first reliable. One of the most critical questions when selecting or developing a new measure or metric is whether that measure provides consistent scores. 
• In the People Analytics field, the term reliability is defined as the consistency of a measure. In essence, we want to know whether what we are measuring is consistent across time (I.e., test-retest reliability), content (I.e., internal consistency), and/or raters (I.e., observer reliability).  
• Different types of reliability evidence require different types of data collection efforts, but the central analysis is the same. Ron and Jennifer talked about ways by which we could operationally assess the reliability of any measures that we use. 
• In addition, they also talked about some rules of thumb for what is "reliable enough." 
• At the end of the episode, Jennifer and Ron recommend steps for folks just starting out in this space all the way to the more advanced HR professional.  

 Related Links  

• Millan Chicago 
• Organizational Testing: Assessment Do's & Don'ts

In the first episode of a special three part mini-series about measurement, co-hosts Ron Landis and Jennifer Miller discuss how measures impact People Analytics.  

In this episode, we had conversations around these questions:  

• What does it mean to measure employee behavior in an organizational context?  
• Why is measurement foundational to People Analytics?  
• What is the process to ensuring that measures are good?  
• What are some of the challenges that organizations face when assessing employee behavior?  
• What are some clear steps that HR professionals can take in the field of People Analytics to ensure that their assessments are effective?   

Key Takeaways:  

• In the People Analytics field, the term measurement is used to assess or evaluate employee behavior in organizational contexts. While companies have access to large volumes of data, it is important to consider what is actually being measured. Without a strong foundation in measurement, any decision based on the data analytic process may lead to unintended consequences.  
• Ron and Jennifer discuss the data analytic process in the context of measurement. First, the topic or area needs to be clearly defined. Second, the data collection process should be articulated including who, what, when and where data will be collected. Third, an analysis of the data should be conducted. Finally, we interpret our results and determine next steps in the analytic process.  
• An organizational example related to the measurement of training program impact is presented. How do you measure ROI of training? Jennfier and Ron discuss this example in the context of the data analytic process. 
• At the end of the episode, Jennifer and Ron recommend steps for folks just starting out in this space all the way to the more advanced HR professional.  

 Related Links  

• Millan Chicago 

In another technically-focused episode, co-hosts Ron Landis and Jennifer Miller deconstruct a statistical technique called logistic regression. They focus on how logistic models can be used to predict the likelihood of a particular outcome. Given the numerous organizational outcomes that are binary in nature (for example, turnover, absence, or promotion), logistic models can provide important insights as to the drivers of such variables. 

In this episode, we had conversations around these questions:  

• What is logistic regression?  
• How is logistic regression used in organizational contexts?  
• How can logistic regression be used to drive optimal business decisions? 
• What are some steps an organization can take to more effectively utilize logistic regression models?  

Key Takeaways:  

• Logistic Regression is a technique used to model relations between variables of interest and predict the probability of an outcome. The focus in this episode is on outcomes that take on one of two possibilities. For example, let’s say we’re interested in predicting whether an individual leaves an organization. Our outcome variable is turnover which we can define as either someone leaving or staying with the company. We also have characteristics about those individuals that we can include in the model as predictors to predict the outcome variable. The model will give information on the likelihood of an individual either staying or leaving the organization.  
• At the end of the episode, Jennifer and Ron recommend steps for folks just starting out in this space all the way to the more advanced HR professional.  

 
Related Links  

• Logistic Regression Resource 
• Millan Chicago 

In this episode, co-hosts Ron Landis and Jennifer Miller deconstruct organizational network analysis or sometimes referred to as ONA. Social interactions are becoming increasingly important to understand in the context of organizational success. While many have access to data related to interactions (i.e., communication patterns including email, chat) little has been done to analyze those patterns. Using ONA to understand and quantify such relational data provides organizations with a means for identifying whether individuals (or groups of individuals) have similar or different employee experiences.  

In this episode, we had conversations around these questions:  

• What is organizational network analysis?  
• How can organizational network analysis be used?  
• What kind of data do I need for an organizational network analysis?  

Key Takeaways:  

• Network analysis is a field that studies the relations among a set of actors. Everyday examples include social media platforms (i.e., connections between individuals) and the electric grid. The goal of network analysis is to examine the nature and patterns of those connections.  
• To establish a network, you need certain kinds of data. Networks have components such as actors, nodes, or vertices. The interactions between the components are sometimes referred to as links or edges. Organizational network analysis examines the patterns of interactions between the components.  
• At the end of the episode, Jennifer and Ron recommend steps for folks just starting out in this space all the way to the more advanced HR professional.  

Related Links  

• Millan Chicago 

In this episode, co-hosts Ron Landis and Jennifer Miller deconstruct data literacy. Data is increasingly important to driving important decisions. While many organizations have access to even more data than before, most organizations could gain significant benefits better using their data to its fullest potential. One of the primary reasons for not maximizing the use of data is that many employees do not have key data literacy skills.  

In this episode, we had conversations around these questions:  

• What is data literacy?  
• How can we assess data literacy?  
• How can organizations consider data literacy in the context of so many different roles and positions?  
• What are some examples across users in which aspects of data literacy would be useful?  

Key Takeaways:  

• Organizations must have a robust data culture to make efficient and effective use of their data. Fundamental to data culture is ensuring that everyone is data literate. Gartner defines data literacy as "the ability to read, write and communicate data in context, including an understanding of data sources and constructs, analytical methods and techniques applied, and the ability to describe the use case, application, and result value." Recent studies have demonstrated that data-driven organizations have a higher enterprise value of 3-5%. In our view, data literacy differs based on the role of the employee in the organization from general users to data scientists to executives.  
• At the end of the episode, Jennifer and Ron recommend steps for folks just starting out in this space all the way to the more advanced HR professional.  

Related Links  

• Millan Chicago 
• Data literacy training: What is it and why do you need it?  
• A Data and Analytics Leader's Guide to Data Literacy  
• The Human Impact of Data Literacy  

In this episode, co-hosts Ron Landis and Jennifer Miller deconstruct employee experience. While many organizations have historically focused on employee engagement, there has been a shift to focus on the broader set of experiences employees have within their organization. This increased focus on the experience is in part due to the ongoing pandemic and the significant number of individuals leaving the workforce.  

In this episode, we had conversations around these questions:  

• What is the employee experience?  
• Why is the employee experience important?  
• How do you measure employee experience within an organization?  
• What are some clear steps that HR professionals can take to measure their organization's employee experience?  

Key Takeaways:  

• The employee experience, sometimes abbreviated as EX, refers to all the ways an employee interacts with an organization —including both work-related tasks (“on tasks”) and non-work-related tasks (“off tasks”). The employee experience is a key factor in driving employee well-being and productivity as well as overall successful business performance. A recent survey by Willis Towers Watson found that more than 9 in 10 organizations stated enhancing the employee experience will be a priority in the next three years.  
• The employee experience can be measured and assessed using our four step data analytic process with the Employee Journey Mapping document as a guide. Check out our summary of the process here.  
• At the end of the episode, Jennifer and Ron recommend steps for folks just starting out in this space all the way to the more advanced HR professional.  

Related Links  

• Millan Chicago 
• Willis Towers Watson Employee Experience Survey 
• Employee Journey Mapping 

In our first technically-focused episode, co-hosts Ron Landis and Jennifer Miller deconstruct a common statistical technique called linear regression. They focus on how regression can be used to better understand the relations between key drivers of important outcomes. 

In this episode, they had conversations around these questions:  

• What is linear regression?  
• How are regression analyses used in organizational contexts?  
• How can linear regression be used to drive optimal business decisions? 
• What are some steps an organization can take to more effectively utilize linear regression models?  

Key Takeaways:  

• Linear Regression is a technique used to model relations between variables of interest and to use these relations to forecast future states. For example, in a simple linear regression, we might be interested in predicting a key outcome variable such as sales from other predictor variables such as number of customers. This kind of statistical technique can be used when the underlying relation between the predictor and outcome is linear (I.e., when the predictor and outcome is plotted, it follows a relatively straight line).  
• At the end of the episode, Jennifer and Ron recommend steps for folks just starting out in this space all the way to the more advanced HR professional.  

Related Links  

• Millan Chicago 

In this episode, co-hosts Ron Landis and Jennifer Miller discuss the importance of utilizing data to make data-driven decisions. While many organizations have turned to analytics to better understand customers, employees, and processes, many are struggling to effectively get the most out of their data. Often, companies fail to use data to its fullest potential because many often lack a strong data culture. 

 In this episode, they had conversations around these questions:  

• What is data culture?  
• Why is data culture important? 
• How do you assess an organization's data culture?  
• What are some steps a company can take to improve their data culture?  

 Key Takeaways:  

• Data Culture refers to an organization's ability to utilize data to make decisions. Companies with a strong data culture consistently reinforce and facilitate the informed use of data in decision making. Data culture is driven by four dimensions: human resource capabilities, human resource processes, technological capabilities, and technological processes. 
• To understand how your organization is currently doing with data culture, take the Data Culture Readiness Assessment for an initial snapshot. 
• At the end of the episode, Jennifer and Ron recommend steps for folks just starting out in this space all the way to the more advanced HR professional.  

Related Links  

• Millan Chicago 
• Data Culture Readiness Assessment