(dailyRx News) Scientists have piles of data showing that people simply don't understand or care about statistics. Most people find numbers un-intuitive.
On the other hand, when your neighbor tells you that the doctor caught his cancer in the nick of time, you regard it as the gospel truth.
A study by Hal R. Arkes, PhD, investigated the reactions of people to the Task Force's recommendation against ever using the PSA test to screen for prostate cancer, finding that although the evidence against any benefit was strong, people felt more comfortable knowing there was a test to detect the cancer early.
When the PSA test was officially retired earlier this week, the feedback was less than stellar.
Anecdotes are powerful, and it's a lot harder to point out all the men who had been harmed by aggressive treatment of slow-growing tumors that would never have caused significant problems.
Psychology shows that people tend to place their trust in the people they know, rather than numbers on a page, graphs, or the dry prose of faceless government scientists.
Additionally, mathematical education in the United States tends to focus more on abstract algebra and geometry rather than statistical concepts, resulting in many people being unfamiliar with some of the more unintuitive laws of probability.
Mathematics teachers have frequently speculated that the government-backed state lottery lobby is chiefly responsible for the predominance of pre-calculus mathematics in the high school curriculum rather than statistics.
Lack of rigor and enforcement of proper study design has also led to fuzzy math published in the public realm on occasion, as exhaustively pointed out in the landmark book Lies, Damn Lies, and Statistics.
Similar to the recommendations made earlier last year about the effectiveness of mammograms, many people are upset with the change, as it seems to be a step back in cancer detection and prevention.
"Many folks who had a PSA test and think that it saved their life are infuriated that the Task Force seems to be so negative about the test," said Arkes.
The study was published in the journal Psychological Science on May 3, 2012.
No financial disclosures were made by the study authors.