This is making the news. A guy we’ve seen before is up to his usual statistical shenanigans in the peer-reviewed paper “Religion as an Exchange System: The Interchangeability of God and Government in a Provider Role” by Miron Zuckerman, Chen Li, and Ed Diener (henceforth “Zuckerman”), in the journal Personality and Social Psychology Bulletin.
Here’s the thesis: “An exchange model of religion implies that if a secular entity such as government provides what people need, they will be less likely to seek help from supernatural entities.”
The “exchange model of religion” fails automatically because the government—yes, even our most beneficent, caring, loving, wise government—cannot provide what people receive from “supernatural entities.”
The model fails if God exists, which He does, because government cannot sustain the universe, let alone balance the budget. The government cannot provide life after death, forgiveness of sins, or a justifiable reason for existence.
The models fails again if per impossible God does not exist, because then people receive nothing from “supernatural entities” and never will or can receive anything. This nothing at least makes it easy for government to provide it.
You would think, then, that Zuckerman’s paper is a short one. It is not. You would think it would have an appreciation for the nature of religion and its history. It does not. You would at least think it would avoid over-certainties and the mistake of equating correlation with causation. Alas, no.
We learn that “researchers” investigated how “the moralizing Big God of the Abrahamic faiths is a mechanism that helped small communities expand into large, cooperative societies of strangers” (my emphasis). Note the assured causal language. We learn about mechanisms but we do not learn that or if Big God is God, a fact which has consequences. The authors do not even consider the point. Science is often purposely ignorant in this fashion. This is no small point. The authors provide no proof of Big God’s non-existence, which we assume they take on faith.
To “researchers”, religion offers benefits, but not transcendence. Benefits like “sense of control”, “self-enhancement”, “self-regulation”. These provide evolutionary benefits, hence religion survives. Of course it could be evolution causes delusion in “researchers”, supposing delusion boosts the chance of breeding, a true statement which cannot be proved false.
Zuckerman thinks religion is akin to a stock market, “an exchange system” in which greater “religious commitment” results in greater access to “commodities”. This ignores the overwhelming sense of duty felt by the adherents of religions, where worship is provided as God’s due. Zuckerman says “Most religions promise the good life to their adherents”, which might apply to certain “prosperity” sects, but it does not apply to the moralizing Big God of the Abrahamic faiths. Christianity promises the sword and the cross.
“[R]eplacing God with government simply means that a moralizing God is no longer necessary for the survival of the group.” Bare physical survival is the sine qua non. To “the extent that the government is responsible for a just order, there is less of a need for God as the arbiter of right and wrong.” This at least is true: absent (an acknowledgement of) God, government must decide between all good and evil, all right and wrong. The State must become God. The State, being led by men, must make man God.
So much for the theory. How about attempts at quantifying the non-quantifiable? Zuckerman obliges.
People in various countries were asked “Is religion an important part of your life?”, with ad hoc necessarily over-certain numeric answers (on a scale of -7 to 42.3, in increments of π/2, how strongly do you agree with this method?). A country was labeled by its predominant’s religion by percent.
Zuckerman then pounced on the World Bank’s World Fact Book and pulled from it “quality of life and government services measures” for each country. Things like education and “number of physicians/1,000 population” and “log transformed” poverty. All these numbers were squashed into one Big “quality of life composite”. Because why not. As long as you in the business of quantifying the unquantifiable, you can do just about anything you want with numbers, no matter how disparate or unrelated.
Gini values per country were taken next. These express, in one single number, the entire complexity of “inequality”, because all you need is one number to express an entire economy. Other “researchers” think so, so Zuckerman thinks so, too.
Then, unquantifiable “life satisfaction, positive emotions, and negative emotions” were quantified and asked of people. Example: “Positive and negative emotions were measured by asking participants whether or not they experienced (an emotion) a lot during the previous day.” How scientific.
Finally, the whole shebang is tossed into a “hierarchical regression” to “to predict religiosity from the quality of life and government services composites and their interaction, adjusting for income inequality and religion.”
Which is nonsense. Great, flaming nonsense. As if entire countries and religions are perfectly homogeneous. Even their wee p-values cannot fix this (common) blunder. Another thing: this was not “quality of life”, it was not “government services”, it was not “income inequality”, and it was not “religiosity”. It was ad hoc single-numbers said to be these things. At best, these were loose and highly uncertain proxies for what could not be measured. At worst, well, you know the worst.
Maybe you don’t. Because after this analysis came “Structural Equation Modeling”, a regression technique used to (falsely) infer cause. Falsely. What it does is measure correlation of all the ad hoc quantified questions to which researchers give fancy names, then it reifies the names into the things, and then forget all the associated uncertainties. I won’t here critique that technique in depth. But see Uncertainty. What we’re left with is a slurry of wee ps and unobservable regression coefficients, from which causality is wrongly inferred (in so many words).
After all this, they did much the same for data from the once-United States. I’ll ignore most of this, except to note a wee p-value led them to say “Higher average quality of life was related to lower baseline religiosity”.
The problem reading that is that everybody forgets the reification. You really do think “quality of life” was perfectly measured on each individual, and that “religiosity” has been unambiguously defined, then perfectly measured.
Let’s move to the conclusion. “[T]he government can provide an extra layer of security…that might help people cope with future needs, both expected and unexpected, and as such, might reduce dependence on God or other supernatural entities.”
This is false in one respect, as noted above. Government cannot provide the most important things. As long as people realize they need these important things, they will look to God (and the entities that bring them closer to God, like the family) and not the State.
But their conclusion is also true in a trivial sense. Indeed, it is so obvious you wonder why Zuckerman bothered doing a study. Everybody already knows that as the State increases, religion must go into hiding. Everybody already knows the State is a jealous god, punishing and aborting the children for the sin of the parents to the third and fourth generation of those who hate the State.