Subjects were informed that the purpose of the experiment was to examine the
effect, if any, of differences in graph structure on the perceptions of the data
portrayed. They were also told that all the graphs to be shown represented time
series data of earnings per share, which is one of four key financial variables
frequently reported graphically by companies. An example of the graphic
displays was shown at the start of the experiment. The instructions to subjects
were given in writing and were worded as neutrally as possible to reduce the
likelihood of subjects guessing the specific experimental hypothesis and
answering in line with the researchers' expectations.