At Least Half of What You Know About Psychology Is Probably Wrong: Reason Roundup

https://ift.tt/2qUyd82

Psychology's "reproducibility crisis" grows, as Many Labs 2, a global collaboration of scientists attempting to replicate the results of hyped psychology experiments past, continues to fail (or succeed, depending on how you look at it). In attempting to "replicate 28 classic and contemporary published findings," the group was only able to do so around half of the time.

Depending on whether they used conventional or "strict significance criterion," the Many Labs team was able to replicate 15 or 14 of the 28 original studies, respectively.

This is common, notes Ed Yong at The Atlantic. "Whenever psychologists undertake large projects, like Many Labs 2, in which they replicate past experiments en masse, they typically succeed, on average, half of the time," he writes. "Ironically enough, it seems that one of the most reliable findings in psychology is that only half of psychological studies can be successfully repeated."

A few years ago, the Open Science Collaboration's three-year Reproducibility Project looked at 100 previous psychology studies and was able to replicate psychology research results about 40 percent of the time.

"Even famous, long-established phenomena—the stuff of textbooks and TED Talks—might not be real," Yong points out. Here are some of the previous research findings unable to be backed up by repeated experiments:

  • the idea that mimicking happy facial expressions can actually boost people's moods
  • social priming ("the field of research about how thinking about or interacting with something ... can affect later, vaguely related behaviour," as one Psychology Today writer puts it)
  • the idea that willpower is a finite personal resource that can be depleted (the subject of a very well-received 2011 book Willpower by social psychologist Roy Baumeister and journalist John Tierney)
  • the finding that exposure to heat primes people to have more belief in global warming
  • the finding that birth order within a family can predict altruism

Some psychologists have blamed non-reproducable results on isolated bad actors--researchers with sloppy technique or unscrupulous data manipulation. Others insist its the replication scientists who are sloppy, using too-small data sets, or failing to understand how the original experiment was done. To ward off these latter critiques, Many Labs took several steps:

  • Many Lab scientists consulted with researchers behind the original experiments
  • Replication-attempt studies had many more participants than in the originals
  • Replication studies were done repeatedly and with participants from different countries

Different cultures and places ultimately didn't matter much in terms of whether a study was reproducable or not. "Exploratory comparisons revealed little heterogeneity between Western, educated, industrialized, rich, and democratic (WEIRD) cultures and less WEIRD cultures (i.e., cultures with relatively high and low WEIRDness scores, respectively)," reports Many Lab. In addition, "moderation tests indicated that very little heterogeneity was attributable to the order in which the tasks were performed or whether the tasks were administered in lab versus online."

You can read about Many Labs work in more detail here.

FREE MINDS

Capitalized words could scare college students, a U.K. university warned its faculty. The staff memo at Leeds Trinity advised professors to "write in a helpful, warm tone, avoiding officious language and negative instructions" when explaining course requirements and tasks.

Despite our best attempts to explain assessment tasks, any lack of clarity can generate anxiety and even discourage students from attempting the assessment at all. Generally, avoid using capital letters for emphasis and "the overuse of 'do', and, especially, 'don't'.

The memo stated that capitalizing certain words might make students anxious by reminding them of "the difficulty or high-stakes nature of the task."

FREE MARKETS

Free market groups fight tax-break package. Americans for Prosperity and Freedom Partners Chamber of Commerce are urging Congressional Republicans not to renew certain tax extenders, which "provide special interest tax breaks and unfairly pick winners and losers by propping up select industries and companies over others," they say.

"Americans across the country, including lawmakers from both sides of the aisle, have rightfully decried the billions of dollars in corporate welfare given to Amazon," the letter continued. "The billions more that are up for renewal in the tax extender package are no different."

"More than two dozen tax provisions, known as 'tax extenders,' expired at the end of 2017, including tax breaks that benefit the renewable energy, motorsports and horse racing industries," notes The Hill.

Outgoing House Ways and Means Committee Chairman Kevin Brady (R-Texas) told reporters last week that he's developed a "draft package" about which of the expired tax breaks he thinks should be renewed and which should be eliminated following the enactment of Republicans' tax-cut law last year. He also said it's unclear what appetite Congress will have to address the expired provisions in the lame-duck session.

FOLLOWUP

Troops pulled from border before caravan arrives. File under good news, bad motives: After deploying thousands of troops to the U.S.-Mexico border in the days leading up to the midterm election, with a supposed purpose of thwarting a group of Central American migrants seeking refuge here, the Trump administration is now pulling the troops before the migrant caravan even shows up.

In other asylum-seeker news, a federal judge has temporarily blocked enforcement of the Trump administration's new rules limiting who can request to come here.

QUICK HITS

  • Utah Republican Sen. Mike Lee pushes back against his colleague Tom Cotton (R-Arkansas) misrepresenting the FIRST STEP Act.
  • Ivanka pulls a Hillary.
  • "If Iran has a policy of detaining dual nationals as a tool of diplomatic leverage then there will be consequences for Iran," British Secretary of State for Foreign and Commonwealth Affairs Jeremy Hunt said on his recent trip there. "We will not let them get away with it scot-free. They have to understand this is not a sustainable situation."
  • Here's the FDA's new statement on lab-grown meat.
  • Russian police general Alexander Prokopchuk could wind up leading the international law enforcement agency Interpol. "With a Putin-appointed police general at the helm, the Kremlin would no longer need to abuse Interpol to pursue its goals; it would be able to place the organization at its service," warns Washington Post contributor Vladimir Kara-Murza.
  • Women's March founder Teresa Shook is not happy with what it's become:



from Hit & Run : Reason Magazine https://reason.com/blog
via IFTTT