Unscientific method: Study finds all those big data studies are mostly big mistakes


Researchers say studying individuals, not large groups, is the only way to accurate conclusions


BERKELEY, Calif. — People like to read studies. This, we at StudyFinds, know to be true. But we also know that people like to debate — and often debunk — the veracity and viability of studies, too. Now a study that actually studied studies seems to side with the naysayers, finding that research which evaluates large groups of people leads to skewed results. In order to get a better and more accurate grasp of mankind, the authors say, scientists need to study people individually.

Researchers from the University of California at Berkeley believe that big data can be a big mess, especially for health practitioners who depend on medical research to guide them in their practices. That’s because human beings can be so markedly different from one another, often-studied subjects like mental health and physiology can yield unreliable conclusions when coming from massive segments of a population.

Books, research
In order to get a better and more accurate grasp of mankind, scientists need to conduct studies on people individually, not in large groups, a new study finds.

“Diseases, mental disorders, emotions, and behaviors are expressed within individual people, over time. A snapshot of many people at one moment in time can’t capture these phenomena,” argues study lead author Aaron Fisher, an assistant professor of psychology at the university, in a statement.

Fisher collaborated with scientists from Drexel University and the University of Groningen in the Netherlands to analyze data on hundreds of adults — some mentally or physically sound, others suffering from various conditions such as depression, anxiety, or post-traumatic stress disorder. Participants had completed surveys about their mental health and had their heart rates monitored via electrocardiogram.

Researchers used the data to conduct six different experiments. They sought to find out whether the conclusions of each study would successfully apply to participants individually.

One study that focused on how frequently depression sufferers reported feeling worried. Results tallied from the pool of participants showed that depressed people worry a significant amount. But when the analysis was applied individually, the results were all over the map. Some participants worried hardly at all, while others were notably beyond the group average.

Another experiment that centered around the link between fear and avoidance showed a strong correlation when measured as a group. Yet a significant number of participants who experienced fear had no issues with avoiding various activities.

Across all six experiments, the authors could not show that what was concluded for the group applied to most individuals.

“If you want to know what individuals feel or how they become sick, you have to conduct research on individuals, not on groups,” says Fisher, who argues that studies should simply be modified instead of completely ignored. “People shouldn’t necessarily lose faith in medical or social science. Instead, they should see the potential to conduct scientific studies as a part of routine care. This is how we can truly personalize medicine.”

The full study was published June 18, 2018 in the Proceedings of the National Academy of Sciences journal.

Comments

    1. Excellent ! Others have pointed out that science never leads to truth, only to the testing of hypotheses within parameters of time, space, society, demographics, perception, accuracy…
      The best science remains “grounded” in measurables, and its results tested in actual applications.
      Almost all science earlier than the mid-20th century has proven to show little more than the philosophical beliefs of its practitioners, and big-data studies discover little more that the effects of researchers’ beliefs upon their data sets.
      What I believe may be a contemporary outcome of big-data studies can be seen in the leftist propaganda manufactured by social-media companies from interpretations of collected comments, lies, distortions, bragging, frustration, dreaming and scamming.

  1. A long study studying studies has recently concluded that most studies should be restudied in light of new studies showing studies to be inconclusive.
    Studies on the subject of health seem to be the most inconclusive of all because of major factors not studied within the context of most studies. The most significant factors not studied appears to be common sugars though studies of studies are not in agreement on that point.
    Many students of studies have given up studying studies as there does not really seem to be any point in studying studies anymore.
    However, some students of HEALTH studies HAVE concluded that INDEED sugar is implicated in ALL major metabolic disorders and ALL minor disorders including mental diseases.

    Of course, we must assume that some of the students of studies, as well as the study authors, were probably high on sugar, or sugar containing substances, when they conducted their studies and therefor the study results were probably skewed.
    Of this we are sure, that there will be more studies to study soon.

    1. It concerns me that you can be so confident the studies are skewed or the students have studied the wrong study. A panel should be convened and study the question.

      1. No, no, I have studied it and studied and studied it to death. It cannot be studied anymore cuz it’s dead.

  2. And in other news today, the sun rose from the east.
    It just shows how idiotic many “researchers” are that we need a study to tell us that the mean of a population tells us nothing without knowing the variance/standard deviation. 99% of the crap we get from these studies are from politically motivated academics.

  3. Science is supposed to be about induction, reasoning from observation, experimentation, etc. to conclusions. Supporting data is expected to be available to all for verification. Debate is supposed to be about whether the methodology is applied correctly. If it is, then the conclusions are true. If not, conclusions are rejected.

    Leftists have tried to make science deductive, as in, conclusion oriented. Championed by Stalin’s pet scientist Trofim Denisovich Lysenko, while going through some motions of data collection, the actual decision turns on whether proposed conclusions are acceptable to a specific ideological outlook. If politically correct, conclusions are accepted and the data adjusted. If not, then conclusions are rejected and their proponents silenced — permanently.

    The former Soviet Union would send dissenting scientists to the Gulag. In the U.S., leftists have to settle for character assassination and/or legal harassment.

  4. I’ve always believed that anecdotal evidence was usually more accurate than the influenced financed studies by the professional ‘studiers.’

  5. It seems Mark Twain was right on the money when he said:
    “There are lies, damned lies and statistics.”

  6. Here’s a sentence taken from the article:
    “One study that focused on how frequently depression sufferers reported feeling worried.”

    I just made a “little data” study. The result is, “Learn how to write a sentence!”

  7. and they tell us that “all”of the data on medical cannabis is “anecdotal” ie individualized therefore very second class per the gov’t and then of course the routine chant of “more study is needed ” well all kind of interesting as regaards this article

  8. “One study that focused on how frequently depression sufferers reported feeling worried.”

    There is no definition of either depression or being worried. It is totally subjective. The idea that anyone puts stock in such a study in the first place is hilarious.

    The more notable junk science right now comes from the global warming people, who make claims that this year was 0.01 degrees warmer than last year.

  9. This is only true for the studies of studies. Any study of climate change, global warming or whatever the term is use today cannot be argued.

  10. The Left has made the use of BAD SCIENCE and junk data a norm now and if you question them on how their facts don’t fit suddenly you are a flat earther or a luddite.

    1. Good news is people are catching on. Polls show (for what they’re worth) 72% of people believe the media lies by omission and selectively chooses and edits which stories to cover based on their agendas, including 90% of republicans, 70 something percent of independents and 48% of democrats. Numbers might vary a couple percentage points, my memory is far from eidetic, though I believe closer to it than most.

  11. At the end of every tax payer study you can typically find the same last sentence. “More study is needed” aka Give us more grant tax money to whiz away.

  12. It has pretty much been proven that 95% of scientific studies are not reproducible. Bad data, bad statistics, bad science is behind most studies.

  13. If find it ironic that one of the universities that politicized and helped discredit all future scientific studies is NOW saying that we should not listen to studies but instead change the way the studies are preformed.

    I would like to study what would happen to all the liberal “scientist” when they all lose their comfy jobs, are homeless, AND finally exposed as the hacks they are.

  14. The real truth is: all studies are, is just an excuse for researchers to get funded by taxpayers through government grants. It’s a win-win for researchers. They aren’t held accountable, so they profit no matter what-no matter how ridiculous the project is. What a scam.

  15. So the conclusion of the study on studies is that the standard deviation is greater than zero?
    .
    Next, they should study whether water is wet.

  16. It seems like the author of this article doesn’t understand the point of these studies. In the case of the cited mental health study, the point is to try to establish a baseline which individuals can be compared to. Outliers do not necessarily somehow render a study incorrect.

Leave a Reply

Your email address will not be published.