Joy at Work - Ensuring Quality Answers

I’ve set out to discover if Joy at Work is important.

To do so, I’ve cashed in 500 quid of YouGov credit to reach a nationally representative audience.

The results are in and I’m crunching the numbers now. Without wishing to give away too much, Joy is important at work, and we don’t get enough of it. But more of that to follow.

Before I share those results, I need to know that what I have to say is valid.

To be sure of that I’ve had to go through a number of different processes.

It’s a constant frustration to me that survey design isn’t seen as a skill. Survey software makes surveys easy to bang out. And we’ve all taken a million surveys, so we all know what makes a good one, right?

No.

You don’t.

So, allow me to get atop my hobby horse for a bit.

Here are some of the considerations I needed to make when design the survey.

Platform choice

Well, as I say, I had credit with YouGov. I’m not going to deny that wasn’t a strong influence. Free is always good, as long as it doesn’t come at a price.

The price here could have been:

·???????? An unreliable platform i.e. one that couldn’t deliver on promises. With YouGov, I can be sure that if they say they will get me a nationally representative sample, then it will be that. And that means that while I could only afford to speak to 303 people – I can extrapolate that data to speak for more people. I can talk in quite general terms, not just about those 303 people.

·???????? A restrictive platform – i.e. one that doesn’t allow me to format the questions in the way I want to get the answers I’m looking for. It’s a self-serve mass-use platform, so there are some limitations, but none that prevent me from getting to where I need to be. It’s all part of the job – I haven’t yet found a platform that allows me to do everything I want, every time. So sometimes the questions need a little reverse-engineering to fit the platform. It's just understanding when you are tweaking the question to fit the platform versus when you are changing it. It’s easy to iterate once or twice and end up in an entirely different place from where you started

Framing the Survey

Likesay, I wanted to understand about Joy at Work. So, I did a little pre-research with my network – what do you think Joy is, what’s it all about?

Turns out it can mean different things to different people.

It shows the value of this kind of pre-work – it tests your assumptions. Sometimes they’re wrong, but most often they’re missing a bit of detail.

?

So, I next turned to some desk research – my own, and also what AI chucked out. And I combined all the ideas I found to arrive at a definition of Joy at Work being:

·???????? What it is: Joy at work is the satisfaction of doing meaningful tasks in a supportive environment.

·???????? What it does: Joy at work enhances your well-being and productivity, in both good and challenging times.

?Now, here’s the challenge. That isn’t everyone’s definition of Joy at Work. For some, Joy comes with the rarely-achieved highs, those once or twice a year moments where it all comes together, and you truly feel you have properly achieved and contributed. For others, it’s much more day to day, it’s moments of levity and fun.

So, we don’t try to talk to those different meanings. But we do do is ensure that everyone is responding in the same way. It’s deciding which is more important.

In this case, consistency was most important to me. With some qualifying questions and free text at the end to ensure that we can capture all of the different meanings of joy. (And – spoiler – what looks like is emerging is that some people have changed lives, and that’s joyful, and other people have just had days without friction, where they’re in-flow, and that’s joyful too.)

Question Design

As with any survey, there are a million things one might want to ask. But the participant, even if incentivised to complete, has a limited attention span. If I chuck 50 multi-choice questions at them – and then expect good answers to 5 free text questions – then the quality of those free responses is going to be questionable.

Fortunately, my budget limited me to about a dozen ?questions: multi-choice and one free text. Which is enough to get some useful answers, and maintain the quality.

The questions and considerations were:

Q1: How important is it to feel joy at work? We would still like to know your opinion even if you are not currently working.

A five-point importance Likert – nice clear result – the one you see as a header to this!

The qualifier is important – our sample includes plenty of people not in work, for many different reasons. We invite all to respond. Subsequently, we get the opportunity to see if their answers differ.

Q2: How often do you feel joy at work (or did you experience in your last role if not currently working)?

Q3: How often do you think employees should expect to feel joy at work?

Frequency likerts that we can directly compare, which means we can discover: do people feel Joy as often as they feel they should?

Q4: To what extent do you agree that: It is the role of the employer to create joy at work?

Agreement likert. An interesting angle that came out of pre-research

Q5: When we talk about joy at work – what kinds of associations do you make? Choose up to four words or phrases that best represent what you think of as joy at work.

Multi-select. Comes out of the pre-research and all those different types of associations – from simple fun to huge achievement.

Q6: To what extent do you agree that: The things that bring me joy at work are different from the things that bring me joy outside of work?

Agreement likert – another way of exploring those associations

Q7: Which of these interactions is most likely to bring you joy at work? Please choose up to two.

Multi-select. Again, comes out of the pre-research – what kinds of people do you experience joy with; those you work alongside or those you work for?

Q8: To what extent do you agree that: Experiencing joy at work is unprofessional.

Q9: To what extent do you agree that: Experiencing joy at work reduces efficiency.

Q10: To what extent do you agree that: Experiencing joy at work reduces creativity.

Q11: To what extent do you agree that: If I was looking for a new job, the chances of experiencing joy at work would not be a consideration.

Agreement likerts – all derived from the pre-research.

Am I happy about the formulation of the final question? No. Would it have been better to ask a simpler question, but “reverse” the agreement i.e. If I was looking for a new job, the chances of experiencing joy at work would be a consideration. I judged not. I judged that the flow of the questions is that the participant is responding to negatives here and we should maintain that.

Q12: Can you describe a time that you felt joy at work, and how it made you feel?

Free text, and if I only had one questions it would be this. But the preceding questions have now seeded and stimulated ideas, meaning we’ll get a richer answer by asking this at the end.

?

Surveys aren’t easy. Bad surveys produce bad data.

Good surveys produce great data.

Invest in good surveys.