George Wolford

Who they are

This page records the public professional record only. Nothing adverse is recorded and no motive is attributed.

He is described in press coverage of the report as a Dartmouth cognitive psychologist, and the report lists Dartmouth College as his affiliation. His wider research was not surveyed for this page, and nothing about it is asserted here.

Key papers

Shah, A. M., & Wolford, G. (2007), 'Buying Behavior as a Function of Parametric Variation of Number of Choices', Psychological Science, 18(5), 369-370, doi:10.1111/j.1467-9280.2007.01906.x.

The report itself was not opened for this page, which rests on press accounts and on the citation in Barry Schwartz's 2016 edition of The Paradox of Choice.

Key ideas

The report varied the number of kinds of pen offered to passers-by, between two and twenty, and measured whether they bought. As Wolford is quoted in coverage, purchases peaked at ten and fell on either side. Schwartz, citing the work, gives the peak as eight to ten (p. 109).

Two cautions apply and the sources on this site do not remove either. The study is short, and the accounts do not give a sample size or an effect size. And it measured a stall of pens, so it shows that an optimum can exist, not where one sits for a different product or a different shop.

The value of the finding is its shape. Where the choice overload literature debates whether more options reduce choosing, this report describes a rise and then a fall, which is the pattern a number-of-options optimum would produce. See choice overload.

The quote and the paper

The figure of ten comes to most readers through a quotation. Wolford is quoted in press coverage as saying that the peak in pen purchases was at ten, and Schwartz's 2016 edition gives it as eight to ten (p. 109). The two are close but not the same, and the difference is a reminder that the number has been restated by people who were not reporting the data themselves.

The report's own text was not read for this page. That matters because a quoted peak and a measured peak are different things. A quotation from a researcher in a news story is a summary of the study, and a summary leaves out the sample, the spread and the cases that did not fit.

What the study can support

It can support the claim that purchases rose and then fell as the number of pens grew from two to twenty. It cannot support a rule such as keep ten lines. The setting was a stall, the product a cheap pen, and the people were passers-by, none of which resembles a shop with thousands of lines. The grocery studies on the choice overload page have the opposite strengths, a real shop and real households, and no clean count of the best number of options.

Taken together, the two kinds of evidence do not give an owner a number. They give an owner two reasons to test.

Using the finding in a shop

A shopkeeper who wants to use the pen result has one defensible move, which is to treat it as a prompt for a small test. Take a display or a category, vary the number of options across comparable periods and watch purchases. The design needs nothing beyond a till record, and it avoids borrowing a number from a stall of pens.

The grocery evidence points the same way from a different direction. Borle's team advise comparing selected categories to limit the damage from a cut, which is a testing approach in all but name. Both pieces of work, taken at their own limits, say that the right number of options is found in the shop and not in a paper.

The sources on this site do not say how many shops have tried the pen design or what they found. This page reports only that the study exists, what it is said to have found, and why the figure should be held lightly.

---