Headline: Religious people understand the world less, study suggests, which derives from the peer-reviewed paper “Does Poor Understanding of Physical World Predict Religious and Paranormal Beliefs?” in Applied Cognitive Psychology by Marjaana Lindeman and Annika Svedholm-Häkkinen.
This study, as we’ll see, proves that scientists are the dumbest smart people around. Scientists are intelligent, the best folks to find evidence for a theory, but they’re just as dumb, or dumber, than untrained citizens at seeking evidence against a theory.
Now scientists know the physical world well, in the sense that they understand secondary efficient causes better than, say, grocers. But because of the Cult of Measurement, scientists do not know the supernatural world well, or at all. There is a correlation: the more science (in the customary sense) a man knows, the less he knows of the spiritual.
As a correlation, it is imperfect; it does apply to all. The notion is familiar enough, though, that everybody knows the truth of it. We don’t therefore need to “research” it, unless we’re interested in hyperspecific, and largely useless, percentages among certain populations (say, Minnesotan chemistry professors).
The correlation works both ways, in a sense: the less science a man knows, the more he believes of the supernatural. The word is “knows” and not “believes” because the supernatural is immeasurably larger than the material, and since there is more to it, knowledge comes at a stiffer price and is generally of a different nature. The qualifications are again necessary because of the Cult of Measurement. Proof in the supernatural world is metaphysical, where in the physical world it is measurable.
Everybody in the Western world knows this correlation, too. It is your non-PhD grandmother who is more likely to light a candle and pray for you than your PhD biologist colleague in the next office. We don’t therefore need to “research” this correlation, either. Unless maybe we’re interested in the nature of supernatural beliefs people hold; and much of these natures won’t be measurable, either.
Lastly, no theory about the non-measurable can be proven or disproved using measurement. Or, to put it another, science has nothing to say about the supernatural. Thinking that it does, again, can be laid at the feet of the Cult of Measurement.
Finally to the paper! It was a “study” of the following sort:
Two hundred fifty eight Finnish participants (63.6% women) took part in the online study. Their mean age was 31.81 years (SD = 9.89, range 18–65). Of the participants, 38.1% were working, 44.4% were students, and 17.5% were employed in activities other than those mentioned earlier; 1.2% had grammar school education, 44.2% had vocational or upper secondary school education, and 54.5% had polytechnic or university education. Religious affiliations were none (61%), Christian (37%), or other (2%).
All Finnish mostly female non-Christian students. All were self-selected on-line volunteers. The 258 were those left of 1,537 after initial questions were given 1.5 years earlier. “In addition, 111 participants were excluded because of missing data. The data contained much missing information because both surveys were long…”
A sample not representative of all mankind in all times and places, no?
The handful that remained were given “instruments”; id est, sets of questions given arbitrary quantification (the Cult again) and which purport to plumb the depths of human emotion. For example: “Eight items from the Supernatural Belief Scale…were used to assess religious beliefs” as if a person’s religious beliefs can be summarized by putting ad hoc numbers to eight questions.
Yes, everybody does this. But everybody is nuts.
“Systemizing was assessed with [questions like] ‘If there is a problem with the electrical wiring in my home I would be able to fix it myself'”, et cetera, et cetera, et cetera. As the authors admitted, too many questions for most.
Then came the wee p-values, mostly “correlating” this artificially numerical question with that artificially numerical question. Which might have been okay—who is surprised, as we agreed at the beginning, that “paranormal beliefs” and “religious beliefs” in these mostly young self-selected persistent Finnish women are positively correlated?
But they didn’t stop there. They had to do “factor analysis” and “sequential multiple regression” using “factor scores”, which they self-labeled “physical capability and school grades”. Wee-ish p-values were discovered (some did not make the magic cutoff but were still called interesting).
(I don’t have space here to tell you how much I dislike factor analysis: but take it from me, it’s silly.)
The more the participants believed in religious or other paranormal phenomena, the lower their intuitive physics skills, mechanical and mental rotation abilities, school grades in mathematics and physics, and knowledge about physical and biological phenomena were; the less they reported interests and skills in systemizing; and the more they regarded inanimate targets as mental phenomena.
All of which could have been guessed from hearing the composition of the sample (mostly non-religious student women, etc.) and using our common knowledge as described above. In short: nothing to see here.
Yet newspaper could take this study and write headlines like “Religious people understand the world less, study suggests“. Which could just as easily be turned around into “Scientific People Understand Reality Less”, where Reality includes the supernatural.