How to run a longitudinal study for employees and customers

People’s opinions and preferences change over time.

In the business landscape, where customers and employees are key to the success of every organization, this idea is critical to remember.

To stay on top of people’s changing opinions, it’s best to check in with them consistently and proactively. Survey scientists often refer to this kind of survey technique as a longitudinal study.

In this article, we’ll review best practices for designing a longitudinal study with surveys for your customers and your employees. But first, let’s review the concept in more detail.

Longitudinal study definition: A longitudinal study is the consistent collection of feedback from the same set of individuals over a period of time. By gathering longitudinal data, you can track patterns and changes in target variables over time. For example, since the beginning of the Covid-19 pandemic, a number of different researchers have embarked on longitudinal studies of the impact of lockdowns and other policies on things like emotional responses and stress. When you gather data from the same set of individuals over time, you reduce the chances that observational changes are attributable to some factor that varies according to the research sample. 

Running a longitudinal study allows you to accomplish the following:

So how, in practice, does a longitudinal study work? When you embark on a new longitudinal study, you should consider yourself as committing to following the same subjects over a specific period of time. However, the nature of the study can vary enormously in size and complexity, as will the number of points of data collection that you’ll have to administer. At one extreme, a very large sample could be repeatedly studied over decades. For example, starting in 1958, the Baltimore Longitudinal Study of Aging (BLSA) is the world's longest-running study of human aging, and has played a crucial role in advancing societal understanding of the impact of getting older on a range of outcomes like housing quality, cancer diagnoses and more. However, longitudinal studies need not be so long or enormous in scope. It's common in business and marketing research to follow smaller samples of consumers over a shorter period of time, like a few weeks, or even just a few days.

Regardless of scope, there are two main rules of thumb when it comes to longitudinal studies. First, you need to collect data from the same set of participants over time. Most longitudinal studies do have some churn, and it’s important to remember that executing a strict longitudinal study is near impossible. People change jobs, customers leave, and workers retire. But if you’re able to keep surveying a sample of people with the same characteristics over time, your study will continue to provide valuable insights.

In addition, you’ll usually need to gather the same information every time you survey or collect information from your participants. For instance, if you’re interested in the impact of economic changes on employment characteristics, you might repeatedly survey your sample on their employment status, the kind of job they have, and how much they are paid. If you gather the same data regularly, you can track changes in that data over time, and link those changes to other changing factors, like age or health.

The term longitudinal study actually describes a number of different methodologies for gathering and analyzing data over a certain period of time. Let’s take a look at some of the different types of longitudinal studies you might consider using. 

A panel study is a study of the same individuals over a period of time. For example, opinion panels are a type of study in which members of the public are invited to complete regular surveys of their attitudes, beliefs, and opinions. Panel studies can even be conducted of entire households, families, or other groups of people. If you’re thinking of creating a panel study and don’t know where to start - SurveyMonkey Audience can help.

A cohort study is designed to track the lives of individuals who share similar characteristics or expect to experience some of the same events over the course of their lives. Probably the best examples are birth cohort studies, which track a random sample of the same children born in the same week. Universities and colleges also use cohort studies to track the employability of their graduates over time. 

Using retrospective studies, a target sample is asked questions about their personal history, such as important events that have occurred in the past. They may also be asked to supply historical administrative data. For example, knowledge about Covid-19 is being advanced through studies that ask people about their travel histories, lifestyle choices, and medical histories. 

There are many benefits of longitudinal studies:

They may be more accurate than other kinds of studies. Although no research methodology is perfect, longitudinal studies make observations more accurate, because the same sample is regularly followed and the same data is repeatedly collected. This reduces the likelihood of data observations being contaminated by extraneous variables relating to sample or data changes.

Here are a few items to keep top of mind before running your longitudinal study:

Think about the specific areas that you hope to influence. Once you’ve picked them out, brainstorm the goals for improving each area and when you should improve them (we’ll use an example to illustrate this point later on).

You want to run enough surveys throughout the year to identify issues on time and to effectively capture trends in your data. However, over-surveying recipients can sour your relationship with them and lead to a lower response rate.

Unfortunately, there isn’t a simple one-size-fits-all cadence for sending out your surveys. Each survey should be handled on a case-by-case basis while taking the following rules of thumb into account:

Pro tip: Use the Data Trends feature in SurveyMonkey Analyze to see how your survey responses change over time.

Once you send your first survey, hold off on editing your questions. Even small edits can significantly influence your respondent’s answer to a given question, making comparisons across survey touchpoints unreliable.

Learn how to automate your sending surveys using our recurring surveys feature.

With this foundation for collecting longitudinal data in mind, let’s review best practices in running a longitudinal study with key audiences for your organization—beginning with your customers.

To collect valuable longitudinal data from your customer base, ask yourself the following questions:

You’ll notice a theme in these questions: segmenting your customers, or grouping them based on shared characteristics.

Segmenting your customers is fundamental to collecting valuable longitudinal data. It allows you to better personalize your questions and gather more detailed, actionable feedback.

To segment your customers, identify the factors that influence their engagement with you.

Here are a few commonly chosen segments:

Once you’ve chosen your segments, work towards defining each one’s success at different points in time. This allows your team to better prioritize its efforts and understand whether—and to what degree—each customer is successful.

To review these steps, let’s walk through a scenario:

Say you work at a company that hosts reviews for all types of small businesses, similar to Yelp. Your goal is to make sure that each business sees value from using your platform.

You notice that restaurants are, in general, focused on using their listing to increase the number of people who map out directions to their business. Meanwhile, heating, ventilating, and air conditioning (HVAC) businesses want to use theirs for receiving more calls from potential clients.

Given the different goals of these two business types, you create a segment for each.

You’d then survey each segment on a quarterly basis to understand their level of progress. The responses will help to prioritize your team’s efforts as well as align with the type of consultation that each business needs.

The questions for your HVAC businesses can look as follows:

1. How strongly do you agree or disagree with this statement: “I am satisfied with the number of calls I’ve received from your platform during the past 90 days.”

2. In a sentence or two, please describe why you selected that particular level of agreement.

The questions for restaurants would be the same, except the prompt for the first question would use mapped directions instead of calls.

Boost Employee Engagement With Longitudinal Data

The costs associated with employing disengaged workers is significant. Unfortunately, it’s a reality for many organizations.

According to Gallup, companies with high employee engagement see a 20% increase in profitability. However, only 13% of employees worldwide consider themselves engaged in their job.

To track and address employee engagement, collect longitudinal data from your entire team.  Longitudinal data allows you to discover areas that need to be addressed, understand changes in employee engagement over time, and measure the impact of any and all of your efforts.

Before running your longitudinal study, consider all of the factors that can influence your employees’ experience. These include:

Remember that employee engagement needs to be reviewed both closely and holistically. Each of these areas offers an opportunity to improve your employees’ engagement, while looking at your longitudinal data as a whole tells you whether your employees’ experience is going in the direction you want it to.

Given the complexity of measuring employee engagement, your employee survey can be comprehensive and look like this.

Interested in developing a more robust process for tracking and improving on employee engagement? Explore SurveyMonkey Engage. →

It’s easy to forget that the people who matter to your organization are human. Their opinions, preferences, and expectations change in ways that aren’t always easy to predict. Longitudinal data keeps you informed and enables you to act in ways that are ultimately in the best interest of your clients, employees, and your organization.

So, now you know how to perform a longitudinal study, and why you might want to consider. Let’s look at some examples of the types of surveys that use a longitudinal approach.

Longitudinal market research is common when you need to continually monitor a fixed sample of a population. For example, if you want to look at how consumer preferences in a target audience change over time, you should consider a longitudinal approach, because gathering data from a different sample could cause sample bias in your results. Types of market research studies that are a good fit for a longitudinal study include:

Market research professionals often take a longitudinal approach when they want to gain feedback on a specific product, especially one that is newly launched. The typical process is like this: first, a target sample will be sent a survey that gathers data on things like their demographic characteristics, their purchasing and consumption habits, and their preferences. Depending on the results, certain members of the sample will be sent a new product to test in the mail. After a period of testing, the same sample will be sent a follow-up sample that looks at whether their preferences and purchasing decisions have changed as a result of the product test, as well as feedback on the product itself.  In this way, product feedback surveys can provide you with highly valuable feedback that you can use to tweak your marketing strategies and drive sales among the target market for your product. 

Customer satisfaction surveys also lend themselves very well to the longitudinal approach. Longitudinal customer satisfaction studies are valuable at any time, but especially if you anticipate the occurrence of some event that might impact customer satisfaction. For example, if you’re introducing a feature change to a new product, or a seasonal rush in orders that slows down delivery, consider gathering feedback from customers before and after the change to make sure that customer satisfaction doesn’t diminish - and to give you the heads up if it does.

Employee engagement surveys are best administered repeatedly and over time. Why? Although you’re likely to have some comings and goings, your employees are a relatively stable sample, making them highly suitable for longitudinal research. And, in any case, employee engagement - the level of motivation and personal investment that your employees put into their work - isn’t a one-shot variable. It will change over time, as a result of policy changes, changes in workload, strategic and operational moves, and more. So, make employee engagement surveys a core and regular part of your human resources strategy - and we can help. 

We hope you’ve enjoyed this comprehensive guide to running a longitudinal survey for employees and customers. We can help you get started with a range of different longitudinal studies - like this brand tracking study. Or, if you have any other ideas for longitudinal studies - feel free to get in touch with one of our experts for help.