Advertisement
Advertisement
Advertisement
Nine out of ten dogmas
20 November 2008
Frank Furedi on the assumptions, agendas and distinctly iffy data behind those ubiquitous words, 'research shows'
As someone devoted to academic research, I feel increasingly embarrassed when I encounter the words "research shows" in a newspaper article. The status of research is not only exploited to prove the obvious, but also to validate the researcher's political beliefs, lifestyle and prejudice.
So last month a study by John Alford of Rice University proved that right-wing Americans are likely to be far more nervous than left-wing counterparts. Liberal readers will be delighted to learn that they are typically relaxed.
We know this is cutting-edge research because he interviewed as many as 46 people. There was also good news last month for people of faith. University of Oxford researchers have discovered that belief in God works as a wonderful form of pain relief. After testing 12 Roman Catholics and 12 atheists, they concluded that believers can draw on reservoirs of spirituality to endure suffering with greater fortitude than unbelievers.
This was not news to the Anglican Bishop of Durham, who observed that the "practice of faith should, and in many cases does, alter the person you are". It also turns people into honest, generous, trusting citizens, according to a study published in Science in October.
If you are offended that your lifestyle and belief have not been validated by gold-standard research, you will be delighted to know that there must be a study out there that proves your moral worth. Liberals may be more chilled out, but right-wing folk are nicer. There is now an important corpus of research that demonstrates that a right-wing outlook disposes people to be happy and to act philanthropically.
They are even less materialistic than those shopping addicts on the left. Peter Schweizer, in his book Makers And Takers, provides incontrovertible data to show that liberals are twice as likely to resent others' success. Conservatives also do well in the nurturing stakes; they hug children more often than frigid lefties.
Thankfully, there is robust research out there that challenges the association of nurturing with a right-wing mindset. The Democratic Party's favourite academic, George Lakoff, is certain that liberals are inspired by the values of "empathy and responsibility".
The politicisation of research is not a new development. Throughout the 20th century, advocacy research served to promote moral crusades and political agendas. Cold War-era psychologist Hans Eysenck reassured his refined readers that "middle-class Conservatives are more tender-minded than working-class Conservatives; middle-class Liberals more tender-minded than working-class Liberals; middle-class Socialists more tender-minded than working-class Socialists and even middle-class Communists are more tender-minded than working-class Communists".
In current times the kind of prejudice promoted by Eysenck is recycled through the dark art of brain research. Scientists can now prove that contrasting political outlooks are related to differences in how the brain processes information. Scientists at NYU and UCLA believe that their experiments show that liberals tolerate ambiguity and conflict better than conservatives because of the way their brains function. They too have numbers - liberals are 4.9 times as likely as conservatives to indicate activity in the brain circuits that engage with conflict.
Despite its links with the past, advocacy research has now acquired an unprecedented significance in Western culture. One important driver of its expansion is the growing significance that people attach to their lifestyles. The very subjects that advocacy research addresses suggest that lifestyle issues such as emotional orientation, parenting styles and the management of relations have become increasingly politicised.
In a world where lifestyle has unprecedented significance, people seek to endow it with moral worth. So it matters when a study concludes that children of gay parents "do just fine" or that single mothers' sons can succeed at school, or that marriage protects elderly adults from mental illness.
Naturally, academics also take their lifestyles very seriously. But it is important that we resist the temptation to discover the moral worth of our lifestyle through our research. And maybe we should take the lead in informing the public that when they see the words "research shows", they should assume the role of a sceptic.
Frank Furedi is professor of sociology at the University of Kent.
Readers' comments
If you do not mind, I would like to be free to quote this statement freely: "But it is important that we resist the temptation to discover the moral worth of our lifestyle through our research." I will use it to respond to all statements of moral virtue being supported by research. We have, in our obsession with scientism and measurement, abandoned the core struggles of philosophy and ethics. We cannot know, we can only know ourselves trying to know.
Wasn’t it Newton's older brother who postulated 'for every study there is an equal and opposite study'? The information age does not discriminate against the impossible any more than the improbable. Right and wrong, truth and fiction, smart and stupid have all become matters of personal choice. If you don't like what you see Google it again with different criteria.
Great article. Before I retired from teaching I had begun to refer to my Director of Education as "Mr. Research-has-shown". "The dark art of brain research" is a marvelous phrase.
Dear Professor Furedi, <p>Do you think that some of the so-called research that you are describing might have been influenced by advertising? There they have two kinds of research to be doubly sure - quantitative, perhaps featuring quantities such as you cited, say 12 or 46 subjects - and the more revealing, deeply probing and thoroughly unreliable form - qualitative, which presumably produces higher quality, more insightful speculation. <p>Much money is spent on advertising based on qualitative research. Apparently it often takes the form of focus groups of 10 or 12 people asked to answer questions. How was the group selected? And what if you asked a different group to answer the same questions in depth? You'd still get telling answers, just completely different ones from the first group's. If you want to massage clients and get them the answers they want, you could conduct enough focus groups so that you could select the ones that tell you what you want to hear. <p>This research produces what is now called "knowledge" in corporate circles. It produces "truths," even "compelling truths," for a fraction of the price of real, reliable, statistically valid research. Who's got time for it? We have Toyotas to sell ... <p>Only a professor would fail to understand that. No wonder you don't want to live in the real world ... <p>Keep up the good work. Cheers, Jack
I am a professor of Psychology and share Frank Furedi's objection to research conducted to confirm lifestyle biases. However, the author appears to have mixed biased research with legitimate investigations.
I would warn against the danger of presuming that quantitative research is always superior to qualitative. It isn't. Positivistic quantitative research is frequently done badly by researchers at all levels who don't really understand it and have scant knowledge of methodology. <p>As an undergraduate I took part in numerous psychology experiments where no-one involved had troubled themselves much with the question of validity (does this proxy measure really capture the phenomenon under investigation?) and the researchers had only a rudimentary understanding of what consitutes a control group. <p>I was asked to make sense of a data set where a psychology graduate had been asked to conduct a survey concerning the factors responsible for children with asthma being admitted to hospital in a 12-month period. She had obtained comprehensive socio-economic and lifestyle data on the hospital admission cases. It was a nice descriptive, quantitiative data set, but what could be concluded from it was limited by the fact that there was no control group (i.e. children with asthma who were not admitted to hospital over the same 12-month period). <p>In another study, a professor interested in knowledge transfer wanted to quantify the business of a nascent industry, and obtained quantitative measures of output from a small, unrepresentative and heterogeneous sample. He did the statistical analysis anyway and published the results. <p>An alternative, qualitative approach is to spend time on the premises of the businesses, observing, talking to staff at all levels and finding out what they do and how knowledge transfer occurs, but that would have taken longer and cost more to do. <p>Quantitative and qualitative studies tell us different things, but that doesn't mean either one is superior to the other. The danger, though, is in researchers having too much faith in their favourite paradigm and not understanding methodology well enough to appreciate the limits of what can be concluded from their data.
Harry Efz: "the author appears to have mixed biased research with legitimate investigations" <p>The difference is easy. Legitimate investigation confirms your beliefs. Biased research comes up with other conclusions. <p>This is not totally true, but is accurate FAR more than many believers would like to admit. <p>It's not just a matter of 'advertising' vs 'academic' research. Both can be wrong whether for the search for money (or publication credits) but also because much of that type of research has inherent problems: small sample size, limited sample cross section, and most significantly, the testing is done on extremely oversimplified questions, which are then generalized to apply to large swaths of people. The selection of those questions, both consciously and unconsciously by the researcher can alter everything about the result.
Harry Efz said: "However, the author appears to have mixed biased research with legitimate investigations." <p>That would be my objection, as well. Just because some research reflecting on lifestyles may be questionably conducted or invalidly interpreted doesn't mean it all is. <p>For example, it DOES matter "when a study concludes that children of gay parents do just fine". In that instance, conservatives argued that gay marriage shouldn't be allowed because it supposedly was bad for the kids. They had no data to back that assertion, just the prejudicial opinion that it must be true because they think gays are bad people. <p>So, a number of studies have now tracked the progress of children of same-sex and different-sex couples and have found by comparing objective data that there is no discernible difference in the outcome of being raised by one kind of couple or the other. Which in turn means that particular objection to gay marriage is untrue and invalid. Which is a genuinely important datum in the public debate on the subject. <p>I would be the first to caution people that all "research" is not trustworthy, but this kind of scattershot snarking at all of it as an undifferentiated mass is no more helpful than slack-jawed credulity would be. If Furedi wanted to help, why didn't he offer tips on distinguishing between good and bad research?
modestly -- if belatedly -- suggests learning about Affirming the Consequent.
I am exasperated by the increasingly popular "let's look at which bits of the brain light up". Since we are essentially clueless on the details of brain function it's a bit like using television adverts as an experimental tool for understanding how electronics work. <p>Combine this with the use of small, biased sample sizes and you have anecdote, not evidence
This piece worries me. If taken to heart, it leads to the conclusion that we might as well relax and dismiss all studies as probably bogus. <p>It strikes me that this dismissal of “studies show” coincides more or less with the leftists’ affection for “truth is subjective” and “all is political”. I’ve come to believe that what is so appealing about those two philosophies is that they give leftist journalists/polemicists the rationale to peddle half-truths without guilt. <p>Furedi writes, “Scientists and NYU and UCLA believe that their experiments show that liberals tolerate ambiguity and conflict better than conservatives”. No kidding – the ambiguity and conflict between what is true and what one wishes to be true. Liberals are quite gifted at that kind of toleration. <p>We are just getting over -- finally -- the 3 decades-long feminist willfully-blind insistence that all behavioral differences between males and females are socially constructed. Or how about the decades-long pipe-dreams that the “self-esteem”, “busing”, and double-standards movements would soon remedy black educational failure? Or, going back a few years, look what a spectacular flop the “War On Poverty” was. It never did reduce poverty but it succeeded wonderfully in undermining the black family. <p>These hulking “Titanics” all sank soon after striking the same big iceberg – the realities of human nature. At least that’s what studies show.
My research shows your bias against research has the potential of costing me grant money. All those surveyed, including a few people not even related to me by blood, agree that only a Neanderthal authoritarian anarchist would come to the conclusions you cite as sacred. Where's your proof that research is bad? <p>I agree, though, that saying, "Marriage protects adults against mental illness" is like saying, "Cutting off your legs protects against ingrown toenails".
Remember too that: <p>a) even the most rigorous studies have a chance of statistical error. For that reason, the more studies, the higher the likelihood that one or more of them is simply wrong. Nobody's mistake: just wrong. <p>b) negative studies seldom get published. No journal can afford space for an article which finds that religion, wealth, education, gender or whatever has NO effect on the variable being tested - even though there may be twenty of these studies for every one that finds the opposite.
Opinion polls show that most members of the general public have never been opinion-polled.
The problem is more one of reporting than research, as can be seen by the way news services also report "break-through findings" in medicine and environmental science, for example, on slow news days. Any decent scientist knows that one study proves nothing. However, reporters don't care, and laymen (and Furedi) don't know any better.
yes, people are all to willing to draw dubious conclusions about different 'lifestyles' from research, and we probably don't have the understanding at present to do so well, but that doesn't mean it is or will forever remain impossible to do so.
the US and Canadian educational systems nowadays force students to take enough statistics to understand sample sizes, and the difference between a hypothesis and a theory. <p>The only ones taken in by shoddy 'research shows' are those whose age denied them this education, or the recalcitrant liberal arts majors who refused to listen in stats class.
Furedi's article insinuates in passing that SOME of the studies cited are probably too small and sketchy to be valuable. Beyond that, he does not really debunk these sorts of studies. Studies of how political affiliations correspond to other sociological and psychological traits are interesting and, when performed correctly, a lot more reliable than "folk wisdom." If there's a big problem with pointing to what "studies show" about politics, faith, hugging, or anything else, Furedi hasn't shown me what it is.
It's amusing to see people respond to this article with indignant cries of "How can Furedi say this particular study is unscientific when it clearly supports my own pet cause or opinion? Sure, some of those other studies are bogus, but certainly not one that concludes something I've made up my mind to believe!" <p>Bogus is bogus, folks, even when the "studies show" something you have an emotional attachment to. We live in an age in which faddish pseudo-science is used to promote social causes and beliefs in the same way advertising is used to promote products. But advertising isn't science, hype isn't fact, and counting how many people hold a particular opinion doesn't make that opinion valid.
Where is the news? We know we need to read research carefully and skeptically; we know we tend to want to confirm what we think -- I mean, the humans among us do; we know what gets published depends on dollars and name recognition; we know it's really difficult to develop our research question, let alone our interview questions; we know the qualitative/quantitative prejudice; we know we are slaves to trends; we know our mom's told us in high school that just because 48 people were doing it didn't mean that everyone was. I do, however, appreciate Robert Speirs' joke about marriage--even if it's point, too, is an old joke written with different words. Slow news day in my office, too.
Isn't this author, by his use of terms without inverted commas (I mean "Liberal" and "Right Wing") displaying (naively) total prejudice as to the moral worth and otherwise of certain people? "Liberal" is just a label, and people who use it of themselves (to promote themselves to the moral high ground) are rarely liberal in any real sense; rather they are closed-minded and quite intolerant of others who may not be "right thinking". So let's see those inverted commas, please, unless you want to naively display your moral arrogance. Yes, research has shown that "Liberals" ...
there is nothing good or bad but thinking makes it so-thus declared Shakespear.our thinking is moulded from our infancy when children are inadvertantly brain washed by parents , and knowingly by their teachers at school and by the time one reaches college it is extremely difficult to overcome prejudices, superstitions which have buried ones rational faculties . this is mostly true of eastern nations where even the highly educated harbour superstions despite the evidence to the contrary. And english essayist Addison said through a charachter " much can be said on both sides" where research is done by people who haave not read much of people like Bertrand Russel , G.B.shaw, Aldous Huxley and such intelelctual gaints, the result will be dubious.
THIS IS AN ONLY UNPROVED THEARY. HOW CAN WE NEGLECT SCIENTIFIC RESARCHES AND SOCIAL AS WELL. COMMENTS MIGHT BE POLITICAL BUT RESULTS CAN NOT BE... SINCERELY DR. V. GANI