By Sharon Begley
Reuters

A high-profile 2013 study that concluded that different varieties of happiness are connected with dramatically different patterns of gene activity is fatally flawed, according to an analysis published on Monday which tore into its target with language rarely welcomed in science journals.

The new paper, published such as first in Proceedings of the National Academy of Sciences, slams the analysis for “dubious analyses” and “erroneous methodology” and says it “conjured nonexistent effects out of thin air.”

In the 2013 study, researchers had adults answer a 14-item questionnaire intended to sort them into two groups: excited about hedonic well-being (fun and selfish pleasure) or eudaimonic well-being (leading a meaningful life).

The two groups, researchers led by psychologist Barbara Frederickson with the University of reported, had different patterns of activity in 53 genes.

Hedonists had DNA activity akin to people experiencing chronic, illness-inducing stress. Stress-related genes including those needed for inflammation were overactive; genes needed for making antibodies that fight infection were underactive.

Hedonists, it seemed, were going to a disease-ridden existence and an early grave, as media reports warned in stories with headlines like “Meaning is healthier than happiness.”

The claim caught the eye of Nick Brown, a British i . t . worker that has turn into persistent amateur critic with the he sees as shoddy statistical analysis in psychology research.

As he and colleagues scrutinized the 2013 paper, they saw numerous problems, he was quoted saying. In whose sale benefits, the authors didn\’t reject that people with specific gene-activity patterns during the defense mechanisms may very well be underneath the weather when tested.

More crucially, Brown said, the enjoyment questionnaire was flawed. Those that scored high on three items directed at identify hedonists scored equally highly on 11 items designed to identify people who seek eudaimonic well-being.

“Each constructs are essentially measuring the same principal,” Brown said, so putting individuals one category as an alternative to another was “meaningless.”

Most devastating was what happened when Brown grouped those things randomly, calling those that scored on top of questions 1, 7 and 8 (or all of 8,191 other combinations) one types of person and those that scored on top of others an alternate type.

Even with your meaningless groupings, there was clearly patterns of gene activity seemingly manifestation of each group.

Statistics professor Andrew Gelman of Columbia University, who had been not needed for either study, called Brown’s critique “reasonable.”

Flawed statistics are getting to be this sort of major problem for journals that many of the world’s top titles are adding extra levels of statistical checks from the peer-review process. Deputy executive editor Daniel Salsbury said PNAS isn\’t changing its longstanding practice, and that is “to your workplace in your review tactic to be sure the attempts are sound in most aspects.”

In a reply to Brown, Frederickson and coauthor Steven Cole with the University of California, L . a ., reject the criticism and say they have got replicated their 2013 findings in a very new sample of 122 people.