OPINION27 May 2010
Radio buttons get a bad press. For a reason. Let’s ban them.
Like PowerPoint, the ‘radio buttons’ used so heavily in online surveys get a bad press. Research would like to contribute to this: radio buttons are horrid.
There’s nothing wrong in principle with a little round button that you click to choose an answer (just as there’s nothing wrong with a piece of presentation software complete with templates and visual effects). But in practice there’s quite a lot wrong.
We’ve been doing some online surveys in the office recently, and our hatred for the radio button has grown strong. First of all, they’re ugly. They’re devoid of style, making surveys look like the complicated settings on your computer that you only touch when something’s gone wrong. But more importantly, they’re fiddly. The gaps between them are generally bigger than the buttons themselves, so you have to be quite deft with a mouse to work through them, especially when the text used to label them has been put at the other side of the screen. From a usability point of view, they’re a liability. And surveys can’t afford that because, unlike most things on the internet, the benefit of a survey is very much skewed away from the user and towards the website.
So should we ban radio buttons from surveys? Would we miss them?
Defenders might say that radio buttons don’t kill respondents, researchers do. What’s more, they have become a scapegoat, a shorthand that people use when they want to complain about poor design, lack of care, lack of imagination.
It’s true that the real problem is the misuse of radio buttons, not the buttons themselves. But when they are so often misused, and so painfully (we’ve gone through grids of more than 150 of them and emerged gasping for breath), that point seems academic. If they weren’t available, maybe there’d be less crime.
So could a ban work? Let’s look at an example where it’s been done: London-based research agency Cobalt-Sky has developed its own survey software and banned the old radio buttons and check boxes. They haven’t looked back. Managing director Raz Khan told Research: “We’ve watched a lot of conference presentations where people have talked about why surveys still look like they did at the end of the 1990s, when the rest of the web has become much more graphically rich and interesting. Then I kept seeing companies claiming they had all these Flash toolkits, but actually were still using the old check boxes and buttons. They only used the Flash tools on high days and holidays or when the client was willing to pay extra.”
When the firm developed its own tools – which are offered at the same prices as before – staff liked the new capabilities but still found the transition hard work, Khan said. Eventually, though, he decided the firm had to bite the bullet. “I thought, we’re not going to do this unless we just stop messing about with the old system. So I told them: ‘You’re not going to do any more surveys using radio buttons and check boxes. I want you to produce good-looking surveys every time.’ There’s no excuse. No one’s going to see a Cobalt-Sky survey with the old-fashioned stuff unless they specifically want it.”
The firm has now been using its new interface on every project for several months. So there you are: it can be done. Let us know whether you support a ban on radio buttons in surveys using the poll on the right-hand side. The buttons in the poll are, of course, exempt from the rule.
The world's leading job site for research and insight
Spalding Goobey Associates
Commercial Research Director
Director level base plus outstanding bonus
Spalding Goobey Associates
Associate Director/Director – Quant – Insight Agency – Life Changing Move to New Zealand
NZ$ 125,000 to 180,000 + Benefits
Resources Group
Senior Research Manager – Government & Social clients
£40–50,000 + great benefits
-
Brought to you by:
©2024 The Market Research Society,
15 Northburgh Street, London EC1V 0JR
Tel: +44 (0)20 7490 4911
info@mrs.org.uk -
© Copyright 2024 Research Live
12 Comments
Joyce Rachelson
14 years ago
Ocucom has been doing online surveys for years without radio buttons. Demo is on our web page. www.ocucom.com
Like Reply Report
Karen Schofield
14 years ago
Are you seriously suggesting we ban radio buttons and in the same breath asking us to vote on this using radio buttons?!!
Like Reply Report
Dan Culshaw
14 years ago
I'm in the camp that thinks that radio buttons are not the issue. Whether you have a 10 x 10 grid of radio buttons or some alternative, it is always going to be very off-putting for a respondent. Best thing to do is to limit the number of rows per page to 5/6 max... then start thinking about how it looks. Also, most systems will allow you to make an area around a button active too so you don't have to hit the button dead on.
Like Reply Report
Reg Baker
14 years ago
I'm with the folks who see radio buttons as a scapegoat for a host of other researcher sins. Radio buttons are fundamental to interacting with the Web. Their use is universal in virtually all Web applications. They are a fundamental convention and it makes complete sense that surveys would use them. If surveys are to be easy for respondents to complete then they should follow the same conventions as other Web applications. We want people to put their cognitive energy into thinking about how to answer the question not how to record that answer.
Like Reply Report
Anon
14 years ago
Was this article sponsored by Colbalt Sky?
Like Reply Report
Steve Cohn
14 years ago
With our iSURVEY product (www.isurveysoft.com) we provide a slider on the iDevice as an alternative to radio buttons for the Likert scale. We are working on a version for the iPad which will allow you to place a series of sliders one under the other to provide the same functionality as a grid of radio buttons (which incidentally, I agree, looks intimidating and ugly on the page). Anyone who looks at the site will note that there are no set "notches" on our slider - our next release provides those - this release gives you a scale of 0 - 100 in the answer behind the scenes. Anyone doing this on the web rather than smart devices could easily build a flash or javascript slider to do the same as what we are doing
Like Reply Report
Steve Taylor
14 years ago
There are some really interesting comments against this story. Good data collection software, as Dan Culshaw comments, doesn't require a "bulls-eye hit" in order to record the response and as Reg Baker says, they are a standard control that any Internet user will be accustomed to. I think the debate is more about the industry's addiction to rating scales. Do people really walk around scoring everything out of ten? I think most think in more simple terms of "I like this" or "I don't like this". The problem is that banks of rating scales fill up questionnaires as part of survey design and in terms of results they can be presented as many charts of mean scores. However, if we accept that people don't score every life experience "out of ten", what insights are these devices really delivering to clients? In a world of declining response, "death by a thousand rating scales" isn't very respondent-friendly in my opinion. We need to both engage and respect respondents.
Like Reply Report
John O'G
14 years ago
I would have to agree with the researchers who suggest radio buttons may be used as a scapegoat for underlying issues. I feel that this issue is relatively trivial and whilst the irony of commenting on a discussion that I do not consider to be worthwhile is not lost on me, I fear that by discussing such matters, it may appear to respondents and people outside the industry that we are missing the point. Furthermore, the idea that radio buttons are problematic because users take longer to select their response due to the size of the button almost suggests that the time taken to respond is shorter than that of the time taken to consider the question, and therefore underestimates and patronises the participants.
Like Reply Report
Michael Hollon
14 years ago
There are many other ways to improve the way market research is conducted that rank far ahead of how radio buttons are used. For example, anyone who has had the experience of going through a list of attribute statements read to them over the phone by an interviewer and has had to repeatedly ask "Now tell me again, what were the options for answering" knows that radio buttons would be a welcome relief from that experience. I applaud the effort to become more respondent centric. But if we really want to make things better for respondents I say ban any survey that takes so long to complete that it leaves somebody saying 'I'm never going to give those market research people my time again' (radio buttons or not.)
Like Reply Report
Maurice Delaney
14 years ago
As a survey scriptwirter with 14 years experience in setting up online surveys, I've seen my fair share of radio button horrors. But, as others have said, with the right scripting (so that the area around it responds to the mouse), radio buttons can still be quicker & easier to use than flash sliders (one-click vs click & drag to right about there, no, there). Our latest Flash offerings using Askia (see demo at http://www.infocorp.co.uk/Inweb.htm) can be very engaging for the consumer, but for B2B research, a more sober, clean finish can still be easy to use with radio buttons.
Like Reply Report