My apologies if this question is best suited for another SE site; I looked around and this seemed to be the best fit.

The answer to the question will hopefully be a link or cite research.

I am being asked to create a form. On this form are two fields for which the correct value is an item in a short list of possible values.

I am being asked that, rather than select the value from a list, the user be required to enter the value manually. The user's input will then be validated against the known correct values.

What I would like to know is whether this approach is demonstrability better at reducing data entry errors. I haven't been able to find an answer using Google.


It turns out the customer did have a compelling argument/requirement for this method of entry: fraud prevention. Functioning as a context-sensitive captcha, the value for this field would be obvious in normal day-to-day activities.


Logic would dictate that, noo, it would not be better to allow manual entry. A pre-written/defined set of options will always be more accurate than a person transcribing those set of options manually.

But that's purely in terms of data ENTRY.

In terms of data INTEGRITY, that's entirely up to the validation system. And a good validation system would make sure either scenario is equally valid.

Then there's the overall usability of each option, which can vary based on the context and the users.

| improve this answer | |
  • I've corrected the title to remove the discrepancy between data integrity and data entry errors. Although I know that logic would seem to dictate selection of a value would be the best way to go (and that's certainly true for usability), I'm wondering if the underlying purpose of this entry method is to prevent "laziness", and reduce entry errors by way of keeping users mindful/purposeful/attentive. This is just a guess though. If this is the case, I'd really like to see some evidence before creating what I perceive to be a usability problem. – overslacked Jan 18 '12 at 23:36
  • 1
    I think usability should trump in this situation. It seems like a difficult task to even test for. If people get 'lazy' it's likely due to repetitive data entry and I can see either model becoming repetitive and people relying more on muscle memory than anything. An example: the system password you have to type in to login to your computer that IT forces you to change every 3 months. Every time that happens, I'm still typing my old password for at least a week or two due to the muscle memory that's developed. – DA01 Jan 18 '12 at 23:56
  • I agree completely; I'm just looking for some data/evidence to present since it's currently my feeling vs. the customer's feeling. – overslacked Jan 19 '12 at 0:06
  • You've encountered the #1 challenge of our profession! It's always nice to find real data, but I think we (as well as our clients) sometimes use it as too much of a crutch. Often, our own personal experience, knowledge and research is enough. Granted, you have to sell that to the client, and that's usually the tough part. At this point, if your client wants real data, propose it to them. Set up an a/b testing scenario and run the site for a few months and take a look at the data. Or perhaps simpler, just ask the users what they want. – DA01 Jan 19 '12 at 1:50

Like DA01 said, there isn't much hard research on such a fundamental question. Or if there is, it's probably from the '80s and doesn't really apply anymore.

I think the real answer to your question is "it depends." How short is the list? Here's a good blog post that discusses when lists are too short or too long to be useful.

Nielsen, one of the usability demigods, advocates using drop-downs sparingly, but they do have their place.

On the other hand, the iOS design guidelines make the good point that selecting from a list is easier for a mobile device than typing.

Finally, this report on what form elements to use when may provide you with some convincing arguments.

| improve this answer | |
  • Thanks for the links! I strongly agree with regards to the use of drop-down controls in most cases. It turns out the customer was able to articulate the underlying requirement; I've updated the question. – overslacked Jan 21 '12 at 8:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.