A random sample of welfare recipients was assigned to a new job-training program. Another random sample was assigned to a conventional job-training program. (Note: The type of jobtraining program they were assigned to is one of the ways in which the participants were classified.) Participants were also classified according to whether or not they had a high school diploma. All of the participants in each group found employment in the private sector at the end of their training. Their mean hourly wages are shown in this table:1
First, consider the column means of $6.72 (for the conventional program) and $8.78 (for the new program). These suggest that, overall, the new program is superior to the conventional one. In other words, if we temporarily ignore whether participants have a high school diploma, the new program seems superior to the conventional one. This difference ($8.78 – $6.72 = $2.06) suggests that there is what is called a main effect. A main effect is the result of comparing one of the ways in which the participants were classified while temporarily ignoring the other way in which they were classified.