MS and red-facedness poll

[nancylebov]

In a recent post, I suggested that people take a poll at an MS blog to increase the range of data about a possible connection between problematic brain veins and MS. However, you need to register to reply there, and this is probably more trouble than it's worth to answer a single question, so I've replicated their poll here.

[Poll #1489798]

Comments

[thnidu.livejournal.com]

Gee, isn't LJ's poll software interesting? Specifically, the bar graph, which I just checked my impressions of by putting a ruler to my screen:
* 0 = 1cm
* 2 = 3cm
* 15 = 16cm

Therefore, 2 / 3 = 0.

To be a little bit more constructive, why don't they just use a little red dot or "|"?

I think I'll post this on the LJ blog as well.

[nancylebov.livejournal.com]

I'd noticed the lack of scaling on the poll results.

Thanks, but could you hold off on the repost for a day or so in case there's a "redder/darker" or somesuch rephrasing?

Update: Could you modify it to "redder/darker" or "redder or darker" if you prefer? See above for the reasoning.

[thnidu.livejournal.com]

I've already posted it as a problem report at http://www.livejournal.com/support/see_request.bml?id=1023355

I don't understand what you mean by "redder or darker". Color has nothing to do with it. They represent any value of n, including 0, by a bar of length = n+1. If all values on the graph are much larger than 1 the effect is negligible, but with small values like these it looks quite anomalous: the bar for 0 is 1/3 as long as a bar for 2, and half as long as a bar for 1. That tells your perception that 3x0 = 2, and 2x0 =1. That's WRONG.

[nancylebov.livejournal.com]

Oh. I had two different topics there.

'Redder or darker' is about what color a person's face might turn. Their skin color is going to affect how much difference increased blood flow makes.

[thnidu.livejournal.com]

That's what I thought you meant, but the bars are also red. My complaint to LJ has nothing to do with the specifics of your poll -- that's why I didn't give a pointer to it. The only reason I mentioned the values is that they're small enough for the bug to make a difference in perception.