What types of personal information should be collected during a usability study?

Question

I'm curious if usability analysts collect personal information about their participants when performing a usability study, and if so what kind of questions do you ask? Does age, gender, nationality, etc matter when analyzing the results, and if so when is it most important?

Answer 1

Its best, when you can, to get all the information possible. But when you want to cut down this a good guide to follow:

1. Who is the user in relation to the product. If its an email program 'company size' might effect the use, but if its a social app for liking food it wouldnt.

2. Sometimes you can have too much data when you don't know what to do with it. If that is the case its better to create "buckets" (or user types, or verticals) that you want all participants to fall into. Then you can look at what those groups chose and trended as a group)

3. Ask questions that reveal bias. In a poll I did on 'What people look for in portfolio sites' I asked if they had interviewed before. This helped me look at what the average person is looking for and what the person looking at my resume is looking for.

4. Ask questions that test previous questions to validate data easily. A question like: "Which product name is the most memorable" might give you Option A as the most popular. However, if on the last page you ask "Which product name do you remember?" Another Option might appear on top. In this fashion you can validate data by simply re-asking another question.

5. As a quick reference: Age, gender, marital status and geographical location have the biggest impact. In addition, a persons Job can be an indication of their personality type which can be a big help as well.

Answer 2

Consider your participants' privacy and don't ask for more detailed information than you will need for recruitment, selection or for a thorough analysis (could be different questions at each stage).

General demographic questions include age, gender, race/ethnicity, nationality, location, employment status, household income, family size, level of education, computer literacy, native language. For many variables you don't need exact values, so you can let them choose between categories (age <20; 20-30; etc.).

Answer 3

Our company will rarely ask participants for personal information (rightly or wrongly). We make software products exclusively for professionals in a particular industry, so our target market is not very broad.

However, we do ask questions like:

• How long have you been in this industry? List certifications, if any.
• Have you used this product before? If so, how long have you been using it?
• Which features of this product have you used?

If you have a diverse target audience, then personal information may play a more important role while analyzing the results. I have found age and cultural background to be important factors to consider during analysis.