Are striped tables, aka zebra tables, not worth the effort? I don’t think this is the conclusion drawn by Jessica Enders in her recent article for ALA, but she certainly puts a big mark question on the subject. In my opinion the approach she took was wrong, and her article shouldn’t have seen daylight, certainly not in this form.
Only two table designs were tested and there were unfortunate choices too. Why testing on tables with borders and stripes? Things that they should have taken into consideration is the color of the stripes, the cell padding, the vertical or horizontal display of stripes, stripes on hover. Just look at how many styles tables are there, they have got to have different performances.
Also as it has been mentioned in the comments following the article, there was a visible timer in the test. When I see a timer my first instinct is to try to do my task as fast as I can. Did I win? Nope. Guess what: I made more mistakes! I am sure that influenced the result.
And maybe they should make it an endurance test. How many errors one makes after an hour of continuous testing. I am sure this influences the result.
When 46 percent of people prefer the stripes and 33 percent don’t have a preference at all doesn’t mean that the rest of 21 percent won. If this was a black or white thing the stripes would have won with almost 70%.
Invoking cost/benefit is silly as one can stripe any table in the world in less that 5 minutes. If the table is for a web app you will save the user tons of frustration and time. Mark my words: anything that takes less that 5 minutes to alleviate users pain is worth it!
This is not the first time wrong conclusions were drawn from incomplete statistic measurements. It is our jobs as designers to solve these problems. We just want well thought test results — we’ll draw the conclusions.
Comments on “Zebra tables”