Statistics

# Enjoy Coffee? You Could Be A Psychopath. Or, Time For A Two-Year Moratorium Of Questionnaire Science

A cup of coffee about the lure another victim.

The headline blared “Enjoy coffee or a gin and tonic? You could be a psychopath: People with dark personalities prefer bitter foods and drinks“. And then the bullet points:

* Researchers studied the preferred food and drink choices of 1,000 people
* Participants then completed a series of personality questionnaires
* Study found a preference for bitter foods was linked to dark personalities
* People who liked coffee, radishes and tonic water were more likely to exhibit signs of Machiavellianism, psychopathy and narcissism

The peer-reviewed paper is “Individual differences in bitter taste preferences are associated with antisocial personality traits” in the journal Appetites (yes) by Christina Sagioglou and Tobias Greitemeyer. You might recall Sagioglou as authoress of “Bitter taste causes hostility” in Personality and Social Psychology Bulletin and of “Activating Christian religious concepts increases intolerance of ambiguity and judgment certainty” in the Journal of Experimental Social Psychology. Or you might not.

In her new work, Sagioglou and pal “investigated how bitter taste preferences might be associated with antisocial personality traits” in which a group of people “self-reported their taste preferences using two complementary preference measures and answered a number of personality questionnaires assessing Machiavellianism, psychopathy, narcissism, everyday sadism, trait aggression, and the Big Five factors of personality.”

Did she say everyday sadism? Yes, sir, she did.

Anyway, you know what followed. Pseudo-quantification of emotional states, emotional states said to be perfectly captured and understood by questionnaires, and wee p-values confirming this-causes-that. And so next time you grab a cup of joe you run the risk of snatching up the cleaver, too, and hacking the barrista to pieces.

I hereby call for a two-year moratorium on all “research” and “science” which in any way uses questionnaires. This is on top of my forbidding all uses of hypothesis testing—to something like worldwide acclaim, at least on the Upper East Side. It’s far past the time to clean out the garbage accumulated by all these “studies.”

It’s so easy to make Science&tm;! All you have to do is pay for some “instrument”, such as the one that assesses “Machiavellianism”, and then invent some new question, such as “How much do you like black coffee?”, on a scale of -17.2 to $e^{\pi}$, and then write a paper which shows how preference for espresso-ground beans causes Machiavellianism.

You can do this endlessly. And after you have done it, you’re not finished! You may well have “proved”, with hypothesis testing, that espresso-ground beans causes Machiavellianism, but then you realize you have said nothing about coarse-ground beans! And even if you have said something about coarse-ground beans, you have been silent on how coffee preference in general effects Machiavellianism in women and minorities.

And so on, as I say, endlessly.

It’s all crap, to coin a word. It’s stinkier than stinky tofu, rottener than Unitarian theology, flimsier than Bill Clinton’s excuses. And it’s pervasive. Question-based science is the very foundation of entire fields, like education, sociology, psychology. Disallow questionnaires and hundreds of journals would dry up.

On a continuous scale of -3.2 to 113-1/3, how would rate the idea of quantifying the unquantifiable? Or how about something more scientific: on a scale of 1 to 8.3, in units of 1/r where r is a prime number, how flummoxed does the previous question make you? Do you think that a “flummoxedness” of 8 is twice as flummoxed as an “flummoxedness” of 4? And is a “flummoxedness” of 4 twice as flummoxed as a “flummoxedness” of 2? Have we captured all there is to know about “flummoxedness” in this “scientifically validated instrument”? It’s validated, incidentally, because I am a scientist and say it is.

Add to this the silliness of hypothesis testing and its pretended identification of cause, and you have what we have now. Newspapers and researchers telling us that liking gin and tonics is “linked” or “tied” to psychopathy. How depressing! (Measured on the patented Briggs Depression scale—only \$495.12 per use—I scored a whopping -4, which was statistically significant, which is the way I knew I was depressed.)

The damage done to clear thinking by pretending batteries of questions adequately quantify emotional states cannot scarcely be underestimated. It’s far past the time to take these things seriously.

Categories: Statistics

### 25 replies »

1. James says:

Why is this even a research question? Are interviewers for jobs going to place coffee for their candidates and label them “Light Roast, Medium Roast, Dark Roast, The Prince Roast”?

“Oh Sheryl! He picked The Prince Roast! We shouldn’t hire him.” “I agree, he is barely human and probably not a resource at all!” *vigorous pearl clutching follows*

2. sigh, …one more piece of evidence that the world is going to Hell in a hand basket. Am I a “dark personality”? Very likely, given my preference for coffee and tonic water… Jonathan Swift had it right with his description of Laputa, the satire on the British Royal Society. Modern times exceeds that description.

3. Ye Olde Statistician says:

Your enemies have struck again and again — twice in the same paragraph!
“Anyway, you now what followed. … snatching up the clever, too, and hacking the barrista to pieces.”

4. Briggs says:

YOS,

They snuck by me as I was enjoying a scotch (bitter) neat.

5. Alan McIntire says:

I hate the taste of coffee, but I drink it regularly in late fall and winter to warm myself up and to wake up/be alert for the coming working day.

6. Dear Briggs —

Most enjoyable blog entry. On a scale of -666 to 2014.09, I favored it at the level of 903.156782, with a probability of success of 93.056 %.

7. Questionnaires were never science. What they are a simple, lazy way of pretending to do science. Thus, the popularity.

If this true, aren’t the other scientists trying to turn us into a nation of violent, hating people by getting rid of sugars and making us eat kale?

8. John B() says:

Briggs: They snuck by me as I was enjoying a scotch (bitter) neat.

My neighbor used to love to my reaction drinking a double shot (bitter/neat) of scotch. Now I can tell him the truth about my dark (roasted) personality.

9. DAV says:

People who liked coffee, radishes and tonic water were more likely to exhibit signs of Machiavellianism, psychopathy and narcissism

Well, it’s true. I like coffee and radishes and even had a yen for tonic water. Once my plan to take over the world is complete they will become required staples in the national diet.

10. MattS says:

“Time For A Two-Year Moratorium Of Questionnaire Science”

Why not a permanent moratorium on questionnaire science?

I’ll even allow people to still do surveys/questionnaires and write articles about the results as long as they don’t call it science.

11. John B() says:

MattS

The two year moratorium would be predicated on Briggs receiving a million dollar grand to develop appropriate scales and hypothesis tests.

Otherwise, yes, a permanent ban.

12. Ray says:

But if you banned science by polling most of the epidemiologists would be out of work. Steve Milloy at junk science blog has been criticizing this since he started the blog. He even wrote a book on it which you can download.
http://junkscience.org/sws.html

13. DAV says:

But if you banned science by polling most of the epidemiologists would be out of work.

14. JohnK says:

Banning questionnaires per se?

A two-year ban on asking questions within a standardized format? Far too broad. ‘Qualitative’ research asks questions that are ‘open-ended’ and don’t require any numerical scaling by respondents (what to do with respondent answers is another story).

IQ tests would be banned, though few if any reputable psychometricians would question their general psychometric (or even their predictive) soundness.

And here’s something to really think about. A ban on “Pseudo-quantification of emotional states” is not only undesirable — it’s impossible, because that is impossible for normal human beings.

Normal people in natural language commonly use ‘less’ and ‘more’ to describe their own and others’ emotional states: “Bertha is very happy — much more happy than yesterday. Her beau, whom she thought lost in the jungles of Central Africa, turned up today and proposed.”

“Pseudo-quantification” has also proved — as in proved — to be helpful in medical triage, for instance — pain scales and the like.

I reiterate: though I’m all for banning hypothesis testing, p-values, ‘confidence’ intervals, parameters-as-reality, woolly cause-proposing, and the like, it is unnatural — as in literally impossible — to ban “Pseudo-quantification of emotional states” from human discourse, and hence, from science.

Thus, the topic of ‘questionnaires’ would seem to demand a closer and deeper examination, for ANY ban purely and solely on the basis of “Pseudo-quantification of emotional states” denies human nature and thus must have something wrong with it.

15. MattS says:

JohnK

“Pseudo-quantification” has also proved — as in proved — to be helpful in medical triage, for instance — pain scales and the like.”

Actually, the pseudo-qantified pain scales gets very little use in medical triage.

Real triage starts with some fairly objective questions.

“Which of this set of patients is least likely to survive even with immediate medical intervention?”

“Which patients would survive without medical intervention?”

“Of those patients who we could save and aren’t likely to survive without treatment, which would die first if treatment is delayed?”

They only get to “Who is in the most pain?” if every other criteria is equal across multiple patients.

16. 1) I take two sugars in my coffee – guess I’m pretty mixed up, eh?

2) umm, Doc? did I miss the follow-up to your “Take Burnham’s Test To See If You’re A Liberal: Suicide of the West at 50” on a scale of 1 to 39 🙂 ; from July?

17. Briggs says:

Paul,

Thank the Lord Burnham didn’t try to create a score!

18. hmi says:

Hello.
Espresso; Italian Coffee.
Machiavelli; Italian politics.
What part of this don’t you understand?

19. K.Kilty says:

Ye Old Stat: and in addition to those two, they sneaked “to” out of the figure caption and replaced it with “the”.

Briggs, that darned coffee cup is out of scale with the hero action figure. It looks like an espresso cup.

20. DAV says:

I had a go around with Amazon today trying to get to Customer Service. They make it devilishly hard. The place where I succeeded required several distinct and not immediately obvious steps. But what’s really something is I stumbled upon a page in their forum which purports to answer the question: “i [sic] received my package but is empty, package is not damaged how can I report this and get my items?”

There was answer that takes one on a circular path however the link under the answer asks: “Did this help answer the question?” with Yes being the only available response.

How’s that for a questionnaire?