Water and wine

Marketers have always argued that price signals quality; this leads to the startling idea that one should just set a high price. 

If you don't believe it, note how Coca Cola and Pepsi turned tap water into a premium-priced $1.7 billion market.  As we now know, Dasani and Aquafina are just bottled tap water.

Even if one can turn water to wine, now researchers discovered the same rule applies.  Unlike most scholarly articles, they actually published a well-made chart to illustrate their experiment.

Testers were given the same wine but told either it cost $10 or $90.  Their brain activity is measured.  The chart showed that those thinking it cost $90 (green line) had much better sensation about the wine than those thinking it cost $10 (blue).

A standard way to display this information is a data table that spells out every estimate and its standard error, plus some asterisk or bolding scheme to indicate statistical significance.  Visualization is far superior.

For more examples, see Gelman's paper or Kastellec and Leoni's paper.

Reference: "Study: $90 wine tastes better than the same wine at $10", News.com, Jan 14, 2008.

Comments

[derek]

The basic concept is good, but it's junky and clunky in all sorts of ways in its execution.

* the y axis has too many tick labels

* the y axis description label crowds the tick labels

* the x axis is a category axis instead of an interval axis, and the tick labels are between the ticks instead of at them

* the x scale labels are at the zero line, mixing in with the data. They should be out of the way, with a separate zero line provided if necessary

* inappropriate x axis spacing of three ticks per label, starting at -5, so the zero point "degustation onset time" gets missed out

* a clunky arrow tries to make good the lack, but the legend gets in the way, so the arrow looks like it's pointing to a space between "$10" and "wine"

An ugly case of Excel Default graphing there.

[Xan Gregg]

Took me a while to realize "-$10 wine -$90 wine" was the legend.

Besides derek's issues, I think the error bars add lots of noise and little value. Since they show little information (about as much variation as a unadorned curve suggests), they can be omitted altogether here.

Is there a better way to show that info when it is useful (or when the audience is more technical)? I usually like shaded regions around the line (a translucent fill between the lower confidence curve and the upper confidence curve), but that sometimes doesn't look good when you have two curves (as here) and their confidence regions overlap into an unexpected color combination.

[derek]

Xan, perhaps a thinner or lighter line on either side?

[Xan Gregg]

Thanks derek. I'll experiment with thinner/lighter/dashed confidence curves. My other idea it to look for fill colors that overlay well, maybe very light colors or grayish shades.