Usability: Why You Only Need to Test with 5 Users

Author: Jakob Neilson & Tom Landauer

Problems found(i)= N(1-(1-L)ⁱ)

where

ProblemsFound(i) indicates the number of different usability problems found by aggregating reports from i independent evaluators,

N indicates the total number of usability problems in the interface

L indicates the proportion of all usability problems found by a single evaluator.

In six case studies (Nielsen and Landauer 1993),

The values of L ranged from 19 percent to 51 percent with a mean of 34 percent.

The values of N ranged from 16 to 50 with a mean of 33.

Using this formula results in curves very much like that shown in Figure above, though the exact shape of the curve will vary with the values of the parameters N and L, which again will vary with the characteristics of the projet