This guide is intended for people looking to grow in an applied people analytics career. For that reason, we will focus on research done in an applied setting, as opposed to academic research. If you are drawn to academia as a career, I think that is fantastic. There is a growing presence of academic research focused on people analytics and a longer history to draw from in adjacent areas of business, economics, and psychology. However, you may want to seek career guidance more aligned with how to succeed in an academic environment – and that information belongs in a different book than this one. 

While academic research lays the foundation for practitioners by expanding everyone’s understanding of the field, it prioritizes theoretical development over practical application. Here, we will concentrate on applied research – which is research aimed at solving specific problems or developing practical solutions to immediate issues because it better aligns with the definition of people analytics I use in this career guide: the application of data and insights to improve business outcomes through better decision-making regarding people, work, and business objectives. 

Let’s start by discussing what we mean when we say research. Sometimes people use this word a little too casually. People will say “I researched what computer to buy,” but they probably mean they just considered a lot of information before making a decision. They most likely didn’t do true research. Research involves a systematic and structured approach to developing a deeper understanding of a topic, answering complex questions, establishing new knowledge, or reaching entirely new conclusions. Research doesn’t usually involve a simple search engine query or browsing readily available resources. While research is a method for finding something out and does support decision-making, it differs significantly in depth and purpose.

Research can often also get confused with analysis. All research involves analysis, but not all analysis qualifies as research. For example, if you gather information from an employee survey and analyze it to understand employee satisfaction, you are doing analysis, not research. Research requires designing and following a method that ensures a critical evaluation of its credibility, potential biases, and a validation process that scrutinizes the data, findings, and interpretations. Analytics involves assessing and finding meaning in the data. Research uses analysis to find meaning in the data, but additionally attempts to control for other possible answers, and considers the reliability, validity, transferability, and applicability of the findings. Let’s compare an analytics-only approach to that of a research-focused approach in an example.

You might have noticed that the above example didn’t include doing a research study. It simply applied research skills and a research approach to a situation where people analytics insights were needed. Research skills will be critical to your people analytics career even if you never try to “do research.” If you want to find insights you can trust, that measure what you intend to, and are valid for decision-making, you will need research skills; even if you do not ever plan to call yourself a ‘researcher.’

People Analytics Research Studies

Even though this guide isn’t intended for those focused on academic research and even though you don’t have to do actual research studies to apply research skills in people analytics, that doesn’t mean research studies aren’t a fun and important part of many people analytics career paths. There are even job titles like “People Analytics Researcher” and others where research projects are involved even if the word research isn’t in the job title. When people analytics research happens to answer specific business questions, test hypotheses, or uncover causal relationships we call that applied research, and people analytics research is a type of applied research. Applied people analytics research can range from simple hypothesis tests to elaborately designed studies, but you will need to gain knowledge and build certain skills if you want to undertake people analytics research in your career.

To do so, I encourage you to start by understanding research frameworks. A framework simply describes a process or structure, in this case, a framework can help you think about how to conduct research while giving you an approach to follow. Two research frameworks I recommend are the scientific method and design thinking. The Scientific Method is a structured approach in which you form an educated guess (hypothesis) about the problem or expected outcome, design an experiment to test that guess, analyze the results, and draw conclusions based on the evidence. It is a linear process – meaning that you usually finish one step before moving on to the next step. In the scientific method, you are usually aiming to test a hypothesis or set of hypotheses before moving on to further research based on the findings. But, it can also be iterative, especially in an applied setting where your initial findings may lead to additional questions and hypotheses that warrant more or different research. Design Thinking is intentionally structured as an iterative approach. It is a problem-solving framework emphasizing empathy, user research, prototyping, and testing solutions. People analytics research can benefit from design thinking’s focus on user needs – since it provides a natural way to consider employee or organizational needs. Design thinking starts with a problem then encourages creativity to imagine a long list of possible solutions. Those possible solutions are then tested and refined until something is identified as the best and most useable solution. Because design thinking is a human-centered approach, it is particularly well-suited for people analytics research and has the added advantage of allowing more opportunities to invite employees into the process. For example, employees can help brainstorm possible solutions, test prototypes, and provide feedback on potential solutions. It allows the people who will be affected by the results to be more actively involved in creating the data and refining the outcomes.

People analytics research can follow the scientific method, a design thinking approach, or a hybrid of the two. However, the actual design of a study will vary greatly depending on what is feasible for the organization, individuals, or the setting in which the research is occurring. Anytime we work with people, there is usually less opportunity to manipulate and control settings. Typical explanations of scientific or academic research may instruct you to have as much control as possible over your research setting in order to conduct an “experimental research” project. While it is possible to design true experimental research projects in people analytics, they are rare. Experimental research is often considered a “gold standard” in the research world. It is where you can manipulate variables such that one group experiences the thing to be measured while another group (the control group) does not. For example, only one group of employees receives access to a new training program. This allows you to study the impact of the variable (the training) by comparing the outcomes of the control group (those who didn’t get access to training) to the experimental group (those who did get access). When you can control who does and does not get access, you can isolate the impact of the training in your research findings and make stronger assertions about its effectiveness. The problem with most people analytics experimental research designs is that it can often be difficult to do this. It can also be unfair, discriminatory, or just bad for employees to only offer a resource to randomly chosen individuals. Let’s say that a training program is effective and those trained are now more likely to be promoted and earn higher salaries. It might be unfair to deny access to people in the control group. For this reason, you will see more people analytics research following either a non-experimental or quasi-experimental approach. Non-experimental research involves observing and analyzing existing scenarios to identify relationships between variables. For example, if the training was optional and made available to all employees, a non-experimental approach would simply analyze the outcomes of those who chose to take the training compared to those who did not choose to take it. This approach does not allow you to control for important differences between the people who choose to take the training and those who do not. For example, people who choose to take training may differ in important characteristics such as motivation, desire for career advancement, and likelihood to apply the training in their work. This would make it harder for you to draw strong conclusions about the effectiveness of the training because you cannot be sure if the success experienced by those who took the training is simply a result of the training or that success is just more likely for the type of people who choose to take voluntarily training opportunities. You might lose the ability to make stronger assertions from your research findings with this non-experimental approach, but you may still choose it to diminish the ethical issues of denying employees the opportunity to be trained. Quasi-experimental research is an approach somewhere between experimental and non-experimental. It attempts to simulate an experiment by manipulating the thing being measured but works within existing constraints. This approach is often used when true randomization is not feasible. It seeks to simulate experimental conditions as much as possible to assess cause and effect (while acknowledging limitations). In our training example, a quasi-experimental study might compare the efficacy of two new trainings, one given to department A and the other to department B. This way all employees receive training and learning can happen to determine which training may be better for future use. As you improve your people analytics research skills, devote time to learning about the different experimental design methods and practice devising research approaches that are suitable for the topic and situation that you are seeking to understand. 

Just as it is not always possible to conduct a true experimental research study, it is not always possible to gather data on all employees for any given study. Most of the time when conducting research, you will want or need to limit how much data you are gathering. You may not have the time or resources to gather data on every employee – it may be bad for productivity to pull employees away from work to conduct surveys or interviews, or you may want to limit the amount of potentially obtrusive data collection happening in the organization. For this reason, it is important to build knowledge and skills regarding sampling methods so you can select a smaller but still representative subgroup of data from which you can draw conclusions and answer your research questions. If done well, these allow you to make solid, useful inferences about the larger population. Sampling methods are not just for research studies, they are useful for all types of analytics since sometimes you will need to use only a subset of your data.

Based on the type of problem you are trying to solve; different types of research data will be more useful for answering different types of research questions. You’ll want to take a moment to decide if you will conduct a quantitative or qualitative research study, or maybe even a mix of both. As we previously discussed, quantitative data helps us understand the quantity of a thing, while qualitative helps us understand the quality. Quantitative data deals with numbers and statistics, while qualitative data focuses on words and experiences. While it is not always the case, for those starting out, you can generically assume that you will want to use quantitative research if you want to confirm or test a theory or hypothesis and you want an answer that is either numerical or in the form of a “yes” or a “no.” For example, “Does recognition lead to employee satisfaction?” can be answered as either yes or no, so you might want to design a quantitative study. Qualitative research, on the other hand, will be more useful when there is no clear “yes” or a “no” answer and you want to understand the quality of something; especially concepts, people, experiences, and processes. For example, “What workplace experiences do employees feel are most important to their satisfaction?” cannot be answered “yes” or “no.” When you need to understand both a quantitative and qualitative answer (e.g., “Does recognition lead to employee satisfaction and which types of recognition are most meaningful to people?”, you’ll want to design a mixed methods research approach – one where you combine quantitative and qualitative data to answer multiple research questions and get a more complete understanding. I love a mixed methods approach when it is feasible to conduct – I believe that numbers can help us understand a lot and test assumptions and that qualitative will give us the nuance and context necessary for appropriate decision-making. Mixed methods research allows you to answer not only “how much” but also “in what ways.” 

Reliability and Validity

In my opinion, reliability and validity deserve way more attention than they currently get in people analytics.

People analytics is predominately focused on analyzing intangible aspects; things that aren’t physical and therefore aren’t as easily measured. As a result, most of the measures we work with are proxy measures; values used to stand in for something else. That means we are rarely measuring the thing we want to measure. Instead, we are doing our best to make assumptions and get as close to reality as we can. Employee satisfaction, for example, is not something we can look at, touch, or quantify easily. So, we do our best to understand and measure it through surveys, interviews, observations, or a combination of other proxy metrics. This means that we are never quite sure if we are really measuring satisfaction or how well or poorly we are doing in our attempts.

It also means that the people with whom you share your findings don’t know if they can trust your measurements either. They will naturally and understandably question your results – a common response you might hear to your people analytics findings could sound like, “yeah, but the survey isn’t really telling us how satisfied employees are.” Personally, I don’t mind this pushback, the person who says that is someone who is thoughtfully engaging with the difficulty inherent in measuring people. But, it is difficult to put them at ease. The only way to help them trust your analyses is to spend time identifying and then showing them the reliability and validity of your measures, your process, and your data. The more evidence you have to show the reliability, validity, and applicability of your findings, the more others can trust your results.

As you build your people analytics research skills, I recommend spending a little extra time learning about the many different concepts of reliability,  internal, external, and ecological validity and especially the concept of construct validity (content or face validity is the easiest starting point in this category, but criterion is the strongest).