A recent article on the Answers in Genesis website discusses a study on cheating. The researchers went to a rural village in Guatemala before the harvest (when people were most likely to be running out of money) and recruited farmers to participate in a study. The farmers were given a die in a cup with a lid. The participants were instructed to roll the die, peek into the cup, and then roll the die again, and report the results of the first roll to the researchers. Based on the results they reported they would be paid extra money. Because the researchers had no way of knowing what the actual results of the dice roll were, the participants could easily cheat. But by analyzing the numbers reported and comparing them to random chance, the researchers can establish how many people cheated. The test was repeated during the harvest (when the farmers would have more money) to see if scarcity influenced how many people were willing to cheat.

The author of the Answers in Genesis article, Troy Lacey, seems to be trying to use the study to support the idea of original sin. The study concludes that people are equally likely to cheat to benefit themselves whether they are living in times of scarcity or abundance. Mr. Lacey seems to be taking that as evidence that sin is innate and not influenced by your environment. This seems strange to me because there are plenty of actions that Answers in Genesis would attribute to innate sin that they do think are influenced by the environment, for example when they complain that scantily clad women will increase men’s lust. But honestly none of that’s particularly interesting to me.

What caught my attention was a different question – how does Answers in Genesis know that the participants cheated? The researchers assume that the participants cheated because the results the participants reported were much better than would be obtained by random dice rolls. But as far as Answers in Genesis is concerned, isn’t that just the researchers making “naturalistic assumptions”?

If Troy Lacey had been one of the poor farmers in this study, and he had rolled a 5, he undoubtedly would have thanked God for the result. The study was done in rural Guatamala, which is a very Christian area. When the recruiters told the farmers that they would be paid extra money based on their decisions, other’s decisions, and luck, many of them probably went home and prayed. When they came to the study and were rolling the dice, I bet a lot of them said a quick prayer, whether silent or out loud. So how do they know that God didn’t help them out?

Of course, we know that if you did a study in which you assigned two groups of people to roll dice, and asked one group to pray and the other group to not pray, there would be no significant difference in the results of the two groups. Prayer does not affect dice rolls when tested. But the excuse I always hear for this is that God doesn’t like to be tested, so he won’t participate in these studies.

This study was not designed to test God; it was designed to test cheating. The researchers have no way of knowing what the participants actually rolled, so there’s no way they could “catch” God helping the villagers out. Why wouldn’t God answer prayers in this scenario? Isn’t that an awfully big confounding factor for anyone who believes in prayer?

I would be very interested to know what would happen if you did this dice-roll test (called the Fischbacher and Föllmi-Heusi die-roll paradigm) with two groups and asked one group to pray and the other group to not pray. Would people lie for God to make it look like their prayers worked? Of course, you wouldn’t be able to tell whether they were lying or if prayer actually worked, so it wouldn’t necessarily be helpful. The closest example I could find was this study, which uses a different method to detect cheating but has the same problem. If you consider the possibility that prayer actually works, the results are useless.

I don’t have much interest in the results of the Guatemalan study, but while reading it I had a few thoughts so I’ll put them here. It’s unfortunate that they were only able to test two time points.

  • The scarcity test occurred in September, just before the harvest. They confirmed that the participants had been suffering from scarcity more in the months before the September test than before the December test by using a survey. But they had been promised money for participating in the survey, and the harvest would be starting soon. I imagine that would be similar someone who has been out of work for months and then just got a job. Their bank account is still empty because they won’t get a paycheck for a couple weeks, but they feel like celebrating and might splurge or buy a round of drinks. I wonder if there would be different results testing in the middle of the scarcity period.
  • The abundance test was done second, so all of the participants who cheated the first time already knew they could get away with it. This may have influenced the decision to cheat again, even if they normally wouldn’t have risked it when they didn’t need it as badly.
  • The abundance test was completed in early December, which is relatively close to Christmas. I would have expected this to increase generosity in some of their other tests, but generosity toward strangers actually decreased during this time.