At least once a day I read or hear journalists make statements, predictions and/or conclusions which are not supported by objective statistics. I’m not suggesting that journalists be prevented from practicing their trade without having the designation of “certified professional journalist (CPJ). But I would feel a lot more confident in judging the accuracy of a journalist’s writing or broadcasting if the CPJ follows his/her name. Most importantly, a journalist who is certified must be competent in basic mathematical statistics, probability theory and sampling methods.
Then a certified journalist would be better able to determine the representativeness and required minimum size of a sample, as an example, to make sound conclusions about the research results. I don’t ever recall one journalist saying, “this sample is not representative of the population being studied. It’s too small to make any scientific judgments about this topic.”
What if a journalist continues to make arguments which are not supported by sound statistical methods? Then I would have to question his political motives.
A recent study released by The University of California at Berkeley, “Higher social class predicts increased unethical behavior,” was widely publicized in the popular press. LATimes.com had its own headline about the study: “Wealthy, motivated by greed, are more likely to cheat, study finds; People of higher status are more prone to cheating, taking candy from children . . . and failing to wait their turn at four-way stops, a UC Berkeley experiment finds.”
But how is wealthy defined? In studies 1 and 2 of the 7 “research vignettes” according to the UC Berkeley paper, “. . .vehicles are reliable indicators of a person’s social rank and wealth; we used observers codes of vehicle status [make, age, appearance] to index drivers’ social class.” Even if every one of the people driving the vehicles with the highest status was indeed wealthy, should one generalize about the unethical behavior of some of these people given the observations made at one intersection in the San Francisco Bay area? Only 27 out of 274 drivers were classified as being in the highest status group. Nineteen of the 27 yielded for other vehicles; 8 did not (29.6%). What about the 70% who did yield therefore not displaying “unethical” behavior? Actually this is a smaller percentage than those who yielded and drove lower status vehicles.
But so what? It’s wrong to judge high status people, call them wealthy, based on the cars they drive [see blog] let alone whether 8 out of 27 drivers of high status vehicles cut others off at an intersection or the other 6 of 13 high status drivers failed to yield to pedestrians. Blanket statements about the so-called unethical behavior of the rich make great headlines. But if one reads carefully beyond the headlines, he will find that the respondents in studies 3 and 4 were not millionaires, not 1 percenters. They were in fact University of California undergraduates who were asked where they perceived themselves to be on the social ladder. What social ladder is that given the fact that these respondents were all students? Does it really matter if those who thought they were of a higher social class took candy from a jar designated for children after permission was given for them to do so?