Early on in your survey design endeavours, you may find yourself feeling the usual mix of anticipation and confusion that accompanies entering unknown territory. Sooner or later, your research ends with an avalanche of feedback-related buzzwords and terms. Open-ended and closed-ended questions. NPS and CSAT metrics. Cross-sectional and longitudinal study.
How can you find your way around, and which of the terms actually matter to a survey newbie?
The short answer is – all of the above, but at various points of your survey design education.
Let’s take a look.
What is a cross sectional study?
Although the term may not ring a bell, it’s very likely you’ve participated in several cross sectional surveys in your life. This method, widely used across academic and traditional research centers, allows you to collect data relevant to one point in time. But before we discuss theory, let’s jump right to an example to get your imagination working.
Let’s assume you run a chocolate factory and recently introduced your first sugar-free chocolate chip cookies. You’re interested in your customers’ first impressions and basically want to know if they’d be happy if the tooth fairy stuck those under their pillow.
For the sake of this example, let’s assume it’s November 2019, and you won’t be cross-analyzing with any future or pasts surveys.
The results are in and now you can:
- Take all the results, regardless of your respondents’ location, and count the numbers (general assessment).
- Divide the respondents into groups by location, and see if consumers in California found the cookies more mouth-watering than those based in Maryland (group comparison).
By now, you likely already know where the term “cross sectional” came from, right?
In this approach, you can perform a so-called comparative analysis among various segments of respondents, but you won’t be going into any cause-and-effect studies as there’s no time factor involved.
This also basically means you won’t be able to see whether chocolate chip cookies were rated higher in November than right before beach season.
Let’s wrap this up nicely in theoretical terms.
Cross sectional study methods allow you to:
- Discover current attitudes, beliefs, or needs
- Compare responses from two or more subgroups, provided at the same time.
What if analyzing things in the here-and-now won’t cut it, and you’d like to see more of a now-and-then approach?
You’ll likely find more value in the second method.
What is a longitudinal study?
Longitudinal study is a great solution if you’re looking to run a longer examination of audience opinion or behavior, as it allows you to take many more factors into account.
Let’s build upon the initial board election analogy a bit further and see how your analysis options expand.
With longitudinal surveys, you won’t simply study what California and Maryland consumers had to say on your chocolate cookies in November 2019.
You’ll analyze how these results compare to surveys ran with the same audience at any time in the past and future. Let’s also assume that, at some point in time, you also decide to run a survey on a different subject – say, a controversial decision to use palm oil as an ingredient. All this will allow you to:
- Learn how many customers in Cali and Maryland would still evaluate your product the same as they had in the Fall of 2019,
- Check how much influence the controversial ingredient decision had on product popularity, in comparison to previous results,
- Notice the emergence of a new segment of customers – perhaps, a group of consumers who live all over the country, but display a specific opinion or change in behavior unlike anything you’ve seen before.
In survey research theory, these three cases are known as:
- Trend – Spotting a behavioural pattern within an audience
- Cohort – Changes within an audience
- Panel – Tracking the very same group of respondents over time, and – potentially – noticing a new subgroup that represents the same opinions or behaviour.
Can you already tell which of the two survey types sounds like something you could make use of?
The good news is you don’t have to decide on a single survey method for ALL your surveys.
While cross sectional and longitudinal surveys are run and analyzed separately, they can complement each other well in building an accurate audience profile.
So, what’s next, and how do you put theory into play?
Here are a few examples of the most efficient surveys for both cross sectional and longitudinal studies of your audience.
The best part? All of them (and many more!) can be created with Survicate within seconds!
Note: While the examples below are especially helpful if you collect feedback for your product or service, you can also find surveys for other industries in our +100 survey template library.
Without further ado, let’s start with the two most popular customer satisfaction surveys and see where they can take you.
Example #1: NPS
NPS is one of the most popular customer satisfaction metrics worldwide (read more about how to calculate and analyze its results here).
This survey method allows you to ask your audience to evaluate your brand from 0 to 10. In short, the goal is to minimize 0-6, and maximize 9-10 responses.
Here’s an example of a NPS survey for Intercom:
Good for: a longitudinal study of audience satisfaction across a period of time.
With a proper audience segmentation, you can also cross-analyze results from a single NPS survey (for ex. January 2019), and see how differently users evaluated your product/service, depending on age, country, or industry, to name a few.
Here’s a pro tip:
If you decide to run NPS surveys, set follow-up questions to collect additional feedback from users who evaluated your product/service poorly. This advice extends to all sorts of surveys, not only NPS!
Example #2: CSAT
CSAT enables you to ask your audience to rate a specific feature/event on a 5-point scale. It’s very likely you’ve seen at least a dozen before – here’s an example of a Survicate CSAT survey:
Good for: a longitudinal study of how a given department, customer agent, or process (for ex. checkout and payment) are evaluated over time.
Similarly to NPS, you can also cross-analyze results from one CSAT survey. For instance, you can check how well a given customer agent performed among respondents between the ages of 24-49 and 49+ in, say, the third quarter of 2019.
Now, here are some use case-specific surveys:
Example #3: General event feedback
Ask your audience how they evaluate an event you recently organized. For example, compare how satisfaction rates differ among users from various industries.
Good for: Cross sectional study and longitudinal study (the latter, if it’s a recurring event and you want to make sure you keep up the standard).
Example #4: Course evaluation
This may come in handy if you want to see how your audience evaluates your course or webinar.
Good for: cross-sectional study and longitudinal study. As in the case of event feedback, this will do great for a cross sectional study of your students/audience. It will also work for a longitudinal study, if your course takes place each semester.
But that’s not all…
Once you run your surveys and start collecting responses, Survicate can also give you a helping hand in noticing trends, emerging attitudes, or any other crucial insights hidden behind feedback.
Our built-in, user-friendly analysis tool will visualize your survey responses within seconds. You don’t have to take your first data analysis steps in black-and-white spreadsheets!
Give us a try and see how you can turn questions and hypotheses into invaluable insights!