Question about survey question design.

[randomness]

Does this question make any sense?

Which of the following server environments does your organization use today?

Intel/Microsoft Windows Server
Intel/Red Hat Enterprise Linux
Intel/Other
Oracle SPARC/Solaris
IBM mainframe
IBM POWER/AIX
Cisco UCS
Other
None of the above
Don't Know

In particular, does it make sense to have both "Other" and "None of the above" as options, particularly with "Other" as one of the above?

On second reading, I suppose "None of the above" might be intended to mean "Our organization don't use a server environment", but wouldn't it be clearer just to say that? Particularly since one of the previous questions was about whether the respondent's job included interactions with server environments?

Comments

[muffyjo]

I"m not sure where "in the cloud" would work but that might qualify as your "Other" situation. It's certainly the answer to the question in my case.

[karakara98.livejournal.com]

I think Other is fairly standard as a category in surveys as a fail safe in case you didn't ask the question right.

[chhotii.livejournal.com]

In my experience, every survey ever sucks in its design. The problem you point out is typical, and the least of what I see in every survey even. I just won't do surveys any more because they ALWAYS cheese me off.

[mamamoira.livejournal.com]

Uh, what about "AMD/"?

[digitalemur.livejournal.com]

I don't really get why they included both Other and None of the above as choices. I do not think they really meant their "None of the above" to include "Other."

If the tool had the capability of collecting write in answers to go with "Other," and they actually cared about what other things people would write in, "Other" would have been a great choice to include without also including "None of the above." If they didn't care, and only had radio buttons for choices, I'd have put "None of the above" but not "Other" as a choice. I can see reasons to do either of the things i just described, but this question as written is muddled.

*shrug* Whatcha gonna do.

[digitalemur.livejournal.com]

PS: Wait, wait, if they gave you the option to click multiple boxes, then including both Other and None of the above make a _little_ more sense. But still, muddled!

[milktree.livejournal.com]

This is why there are people who actually study how surveys (and people taking surveys) work. It's tricky to write a survey that's not annoying, *and* doesn't assume all sorts of things about the taker.

That use of radio buttons is nearly as absurd as radio buttons for:

When you were in grade school did you...
( ) Walk to school
( ) Take a lunch

[drwex]

That's just bad and lazy survey design. On the other hand, designing good questions is sufficiently hard that a significant portion of a masters in psychology used to be taken up with teaching people that skill.