I think the test from Barefoot is not the best test to use as an example. The version on the right has just one radio button, so at first glance there seems to be no 'way out', which can scare some users away. The version on the left has multiple radio buttons, so the user will feel more in control and will probably be more likely to stay long enough to read the "(Optional)" text. The test is not only about the Hobson's +1 effect because of users' expectations about radio buttons.
Thanks for taking the time to write Sjoerd. I actually ran exactly the same test as on barefoot (btw it's not what they do, I added it) on another website and it worked better when there were 2 options. And there was no difference in conversions, but huge increase in signups. The big difference is that the checkbox was not selected by default in my test. In this my example was a bit misleading. That said your hypothesis could be true. But It's worth running a test for your case and making sure it's a checkbox instead of radio button.
I think the test from Barefoot is not the best test to use as an example. The version on the right has just one radio button, so at first glance there seems to be no 'way out', which can scare some users away. The version on the left has multiple radio buttons, so the user will feel more in control and will probably be more likely to stay long enough to read the "(Optional)" text. The test is not only about the Hobson's +1 effect because of users' expectations about radio buttons.
Thanks for taking the time to write Sjoerd.
I actually ran exactly the same test as on barefoot (btw it's not what they do, I added it) on another website and it worked better when there were 2 options. And there was no difference in conversions, but huge increase in signups.
The big difference is that the checkbox was not selected by default in my test. In this my example was a bit misleading.
That said your hypothesis could be true. But It's worth running a test for your case and making sure it's a checkbox instead of radio button.
@@clicksdontlie Nice! That should be more reliable. Thanks for your response!