Analyze your survey data like a pro, by following a few simple principles and utilizing Intersect visual blocks.

How to analyze Typeform survey results with Intersect

Anita Kirkovska
Anita Kirkovska
October 15, 2021

You wrote a great Typeform survey and distributed it to the right audience. Perhaps it was a survey sent to customers, the general public, partners, or even employees...

Finally, your survey results are in, and you are thrilled with the number of responses!  

Now, you are ready to analyze their answers.

The question is, where to begin?

Even after writing the perfect survey, and receiving amazing response volume, failing to analyze your responses in a coherent way, may lead to poor insights.

Don't be intimidated by data analysis because it sounds technical and complicated.

By using no-code platforms like Intersect and following some simple principles, you'd be surprised how much you can accomplish on your own.

Contents

First, Get total responses

As a first step, before you do any analysis, it’s important to know your total responses.

Let’s imagine that you’ve organized an event, where 200 people participate. You would like feedback on the speakers, logistics, food, etc.

You’ve sent your Typeform to every attendant, but only 5 respond.

Obviously, the feedback of 5 survey participants isn’t necessarily representative of the experience of the other 195 attendants.

While this is an extreme example, it is important to recognize the statistical significance of your total responses in relation to the total population.

Next, Calculate response rate

Let’s continue with the previous example. Since you received only 5 responses from 200 total participants, this gives you a response rate of 2.5%.

Calculating your response rate is very simple:

Response rate = Number of responses / Number of participants you asked to fill out the survey

According to PeoplePulse, customer satisfaction surveys and market research surveys often have response rates in the 10% – 30% range. Employee surveys typically have a response rate of 25% – 60%.

While different types of surveys tend to yield varying number of responses, nevertheless, the higher your total responses and response rate are, the more confident you can be that your sample is representative of the entire population.

You can calculate this in a calculator or Excel sheet, as Typeform doesn’t display this insight for you.

It’s important not to confuse the “Completion rate” with the "Conversion rate”, as the completion rate is the percentage of people who completed the Typeform, vs those who opened the Typeform. Therefore this value fails to account for the population as a whole.

Aggregate your survey responses

Next, you need to aggregate your individual responses.

Aggregating is simply the process of summarizing the responses by totaling how many participants answered each answer option.

You can find all your individual responses under the Responses tab in your Typeform.

Luckily Typeform automatically performs the aggregation of responses in their reports. This can be found under the “Summary” tab in Typeform.

The example above shows the breakdown for the “What is your age” question:

  • 18 to 24: 3 responses, or 50% response rate
  • 35 to 44: 2 responses, or 33.3% response rate
  • 65 to 74: 1 response, or 16.7% response rate

There are 6 total responses to this question.

Compare Typeform results with cross-tabulation

Cross tabulation means that we quantitatively analyze the relationship between multiple variables. Simply put, we are interested in learning how different groups of people behave or how a single outcome may be affected by different factors.

When you sent out the survey to your participants, you had some ideas in mind about the types of comparisons you hoped to make. Maybe you want to know if younger people find your oldest speaker interesting or not.

But unfortunately, there is no option for you to cross-compare different variables within Typeform reporting. You can do this by using a tool that is natively integrated with Typeform.

Intersect, is an official analytics integration, and it gives you more power over what you can calculate, analyze and visualize from your Typeform data.

Using the simple interface, you can connect your Typeform survey, and automatically pull your data into a data app. You can learn how to connect your Typeform in Intersect’s official documentation.

Then by using visual blocks you can aggregate and merge different variables so that you can start to identify correlations. To illustrate this, let's look at data from a random event survey.

Using the simple interface of Intersect, we aggregated the survey data to examine how age groups rated the speakers and how likely they are to recommend us to friends and colleagues.

The data shows that 18 to 34-year-olds enjoyed the event the most. Based on these results, the event organizers may want to promote their next event to a younger audience for similar topics.

Benchmarks - Comparing survey data

In our example, we sent out a feedback survey to our event participants. Analyzing this data might reveal that people aged 25-34 rated the staff very high, or with a rating of 60%. But, how do we know if this is good?

That’s why we use benchmarks.

Benchmarks allow us to compare our data to other previous results. An example of how to do this would be to compare this event's survey responses to our previous events. Did this rating improve over time? If yes, 60% is an improvement, and we can cross-tabulate to understand what changes led to this improvement.

However, if this is our first event, then we don’t have any previous data to form a benchmark from. If this is the case, you can calculate the overall split between the responses for each question, and use this as your benchmark.

Then as you cross-tabulate and create sub-categories of your data, you can compare those results with the calculated benchmark to add context to your insights.

Survey analysis - Why VS What

This qualitative data can easily show us what happened. However, the real challenge is to understand why a specific thing has happened.

Understanding the “why” is the first step to gathering real insights, and making improvements.

This "why" can sometimes be answered with one question, sometimes with several questions and multiple choices. But sometimes it is up to the analyst to determine causation (if any).

So all this is easy to say, but this is where things can get complicated.

When interpreting survey results, it is easy to mistake correlation for causation.

Survey analysis - Correlation VS Causation

Let’s understand the difference between correlation and causation.

Correlation describes associations between variables; in other words, when one variable changes, the other one changes as well.

Let’s use a random example to illustrate this. For example, as ice cream sales increase, sunglasses sales increase as well. This correlation only tells us that these two variables move together, but it doesn’t tell us why they move like this, or what is the cause behind this connection.

Causation, on the other hand, means that changes in one variable directly cause another variable to change, thus there is a cause-and-effect relationship between variables.

For example, when the temperature rises in the summer, ice cream, and sunglasses sales rise as well.

Survey reports

Once your data has been analyzed, it's time to share your findings with your team or upper management.

Reporting is the final phase of your survey analysis, and it will help you share your insights across the board.

At this stage, however, you need to remember that the story matters more than any number, and you need to have a narrative to share all the insights you've gathered.

When creating this report, remember to think of your audience and ask yourself how to best engage them effectively.

If you want to show improvement, mention the benchmarks and illustrate the positive improvement.

Also, if you are reporting to different teams, build different reports that illustrate the top 3 things that are most relevant to each audience. This will increase the effectiveness of your campaign well beyond a single, one-size-fits-all report.

Sadly, Typeform reporting options are very limited, so you’ll definitely need a better reporting tool to showcase your findings.

With Intersect, you can document your story, visualize interactive graphs and then share and invite interaction from the rest of your team.

You did your due diligence in creating a well-planned survey, identifying an appropriate target audience, and gathering plenty of feedback. So don’t sell yourself (and your response data) short by settling for weak, generic analyses.

FAQs

No items found.

George Holt, Growth Analytics @Legacy

Before Intersect Labs, cleaning our customers' data was slow and manual. Now, it is done in couple of minutes.

Book a Free Demo
Book a Call
Book a Demo
This is some text inside of a div block.

How to analyze Typeform survey results with Intersect

Analyze your survey data like a pro, by following a few simple principles and utilizing Intersect visual blocks.
Anita Kirkovska
Anita Kirkovska
October 15, 2021
October 15, 2021
Updated on:
Contents

Automate Your Data Work

Intersect enables you to build data workflows and turn them into easy-to-use interfaces.
Get Free Demo

You wrote a great Typeform survey and distributed it to the right audience. Perhaps it was a survey sent to customers, the general public, partners, or even employees...

Finally, your survey results are in, and you are thrilled with the number of responses!  

Now, you are ready to analyze their answers.

The question is, where to begin?

Even after writing the perfect survey, and receiving amazing response volume, failing to analyze your responses in a coherent way, may lead to poor insights.

Don't be intimidated by data analysis because it sounds technical and complicated.

By using no-code platforms like Intersect and following some simple principles, you'd be surprised how much you can accomplish on your own.

Contents

First, Get total responses

As a first step, before you do any analysis, it’s important to know your total responses.

Let’s imagine that you’ve organized an event, where 200 people participate. You would like feedback on the speakers, logistics, food, etc.

You’ve sent your Typeform to every attendant, but only 5 respond.

Obviously, the feedback of 5 survey participants isn’t necessarily representative of the experience of the other 195 attendants.

While this is an extreme example, it is important to recognize the statistical significance of your total responses in relation to the total population.

Next, Calculate response rate

Let’s continue with the previous example. Since you received only 5 responses from 200 total participants, this gives you a response rate of 2.5%.

Calculating your response rate is very simple:

Response rate = Number of responses / Number of participants you asked to fill out the survey

According to PeoplePulse, customer satisfaction surveys and market research surveys often have response rates in the 10% – 30% range. Employee surveys typically have a response rate of 25% – 60%.

While different types of surveys tend to yield varying number of responses, nevertheless, the higher your total responses and response rate are, the more confident you can be that your sample is representative of the entire population.

You can calculate this in a calculator or Excel sheet, as Typeform doesn’t display this insight for you.

It’s important not to confuse the “Completion rate” with the "Conversion rate”, as the completion rate is the percentage of people who completed the Typeform, vs those who opened the Typeform. Therefore this value fails to account for the population as a whole.

Aggregate your survey responses

Next, you need to aggregate your individual responses.

Aggregating is simply the process of summarizing the responses by totaling how many participants answered each answer option.

You can find all your individual responses under the Responses tab in your Typeform.

Luckily Typeform automatically performs the aggregation of responses in their reports. This can be found under the “Summary” tab in Typeform.

The example above shows the breakdown for the “What is your age” question:

  • 18 to 24: 3 responses, or 50% response rate
  • 35 to 44: 2 responses, or 33.3% response rate
  • 65 to 74: 1 response, or 16.7% response rate

There are 6 total responses to this question.

Compare Typeform results with cross-tabulation

Cross tabulation means that we quantitatively analyze the relationship between multiple variables. Simply put, we are interested in learning how different groups of people behave or how a single outcome may be affected by different factors.

When you sent out the survey to your participants, you had some ideas in mind about the types of comparisons you hoped to make. Maybe you want to know if younger people find your oldest speaker interesting or not.

But unfortunately, there is no option for you to cross-compare different variables within Typeform reporting. You can do this by using a tool that is natively integrated with Typeform.

Intersect, is an official analytics integration, and it gives you more power over what you can calculate, analyze and visualize from your Typeform data.

Using the simple interface, you can connect your Typeform survey, and automatically pull your data into a data app. You can learn how to connect your Typeform in Intersect’s official documentation.

Then by using visual blocks you can aggregate and merge different variables so that you can start to identify correlations. To illustrate this, let's look at data from a random event survey.

Using the simple interface of Intersect, we aggregated the survey data to examine how age groups rated the speakers and how likely they are to recommend us to friends and colleagues.

The data shows that 18 to 34-year-olds enjoyed the event the most. Based on these results, the event organizers may want to promote their next event to a younger audience for similar topics.

Benchmarks - Comparing survey data

In our example, we sent out a feedback survey to our event participants. Analyzing this data might reveal that people aged 25-34 rated the staff very high, or with a rating of 60%. But, how do we know if this is good?

That’s why we use benchmarks.

Benchmarks allow us to compare our data to other previous results. An example of how to do this would be to compare this event's survey responses to our previous events. Did this rating improve over time? If yes, 60% is an improvement, and we can cross-tabulate to understand what changes led to this improvement.

However, if this is our first event, then we don’t have any previous data to form a benchmark from. If this is the case, you can calculate the overall split between the responses for each question, and use this as your benchmark.

Then as you cross-tabulate and create sub-categories of your data, you can compare those results with the calculated benchmark to add context to your insights.

Survey analysis - Why VS What

This qualitative data can easily show us what happened. However, the real challenge is to understand why a specific thing has happened.

Understanding the “why” is the first step to gathering real insights, and making improvements.

This "why" can sometimes be answered with one question, sometimes with several questions and multiple choices. But sometimes it is up to the analyst to determine causation (if any).

So all this is easy to say, but this is where things can get complicated.

When interpreting survey results, it is easy to mistake correlation for causation.

Survey analysis - Correlation VS Causation

Let’s understand the difference between correlation and causation.

Correlation describes associations between variables; in other words, when one variable changes, the other one changes as well.

Let’s use a random example to illustrate this. For example, as ice cream sales increase, sunglasses sales increase as well. This correlation only tells us that these two variables move together, but it doesn’t tell us why they move like this, or what is the cause behind this connection.

Causation, on the other hand, means that changes in one variable directly cause another variable to change, thus there is a cause-and-effect relationship between variables.

For example, when the temperature rises in the summer, ice cream, and sunglasses sales rise as well.

Survey reports

Once your data has been analyzed, it's time to share your findings with your team or upper management.

Reporting is the final phase of your survey analysis, and it will help you share your insights across the board.

At this stage, however, you need to remember that the story matters more than any number, and you need to have a narrative to share all the insights you've gathered.

When creating this report, remember to think of your audience and ask yourself how to best engage them effectively.

If you want to show improvement, mention the benchmarks and illustrate the positive improvement.

Also, if you are reporting to different teams, build different reports that illustrate the top 3 things that are most relevant to each audience. This will increase the effectiveness of your campaign well beyond a single, one-size-fits-all report.

Sadly, Typeform reporting options are very limited, so you’ll definitely need a better reporting tool to showcase your findings.

With Intersect, you can document your story, visualize interactive graphs and then share and invite interaction from the rest of your team.

You did your due diligence in creating a well-planned survey, identifying an appropriate target audience, and gathering plenty of feedback. So don’t sell yourself (and your response data) short by settling for weak, generic analyses.

Enjoyed The Read?

Don’t miss our next article. Get updated on current data trends and powerful automations that can solve your data challenges.

No items found.

Get your time back.

Build your first interactive data app today and start saving hours every month.

Book a Demo