Using the Bell Curve to Make Accurate Generalizations

While understanding that every individual deserves to be treated as an individual

I love the bell curve. I appreciate how it allows me to understand the characteristics of a population without making untrue blanket generalizations. For example, I would never say to be true, “Men are taller than women,” because there are so many individual women that are taller than individual men. I could look at two bell curves, though, that plot height distribution of adult men and women in the U.S., and see that the one for men centers around 5'10", whereas the one for women centers around 5'5". So I could say something like, “Over 50% of adult men in the U.S. are taller than 5'7”, and 50% of adult women in the U.S. are shorter than 5'7",” and it would be true. If I had better math skills, I might even be able to calculate the odds that any individual man picked at random would be taller than any woman picked at random.


When I hear people make generalizations like, “Rich people are greedy,” or “Mexican immigrants are criminals,” I immediately reject it. To my way of thinking, as long as some of the people in the group don’t fit the description, it’s an invalid statement. I hate it when people see an example of certain individuals behaving a certain way, then say their whole group acts that way. Sure, some men are stoic and have difficulty expressing their emotions, but it doesn’t mean all men are like that. Sure, some women get debilitating menstrual cramps every month, but it doesn’t mean all women are like that. I like to interact with individuals as individuals, treating each one as a mystery waiting to be explored with curiosity and open-mindedness.

I think this is a great way to lead life, yet it does hamstring me a bit when it comes to debating public policy. One thing I’ve noticed in life is that every group has people that are kind, every group has people that are bigots, every group has people that are compassionate, every group has people who lack empathy. One way I think of it shorthand is that every group has “good” people and “bad” people, even as I understand that the qualities I think of as “good” — kindness, friendliness, compassion, empathy, consideration of others — and those I think of as “bad” — selfishness, narcissism, cruelty, close-mindedness, lashing out without thinking — are present in all people some of the time, and different situations will bring out different sides of people. In my experience, nobody is wholly good or wholly bad, so that phrase “every group has good and bad people” comes with a huge disclaimer — it’s a simple phrase that points to a complex truth without being the truth itself.

Given that every group has people acting anywhere along the spectrum from selfish cruelty to loving giving kindness, it can be difficult to assess the efficacy of that group’s culture. I have a friend who shared with me that she doesn’t feel as comfortable around Muslims as around Christians, because her understanding of their religion is that it promotes violence in the form of jihad, so Muslims are more likely to be violent than non-Muslims. I shared with her my perspective that the Christian religion has also been used to promote violence, from the Crusades to the Inquisition to killing doctors that perform abortion, so I fear Muslims as little as I fear Christians, because I think the vast majority of both groups are regular people leading regular lives, and it’s not fair to judge the whole group by the visible, vocal minority.

Although that is true for me, it did make me wonder if it would be possible to compare Islam and Christianity to see if one is more likely to produce violent behavior, separate from other various influences such as what country a person is born into, how poor they are, whether they grew up in a war-torn country or a peaceful one, or how much education they have. The idea I’m playing around with is that maybe some cultures are better than others at producing “good” behavior. Even though I know every group will have a full range of “good” and “bad” people, so I don’t want to make a blanket judgement like “Christians are good,” or “Christians are bad,” perhaps it’s possible to look at a bell curve distribution of behavior among Christians and behavior among non-Christians and see how the bell curves compare. With the example of men and women, although there are many women who are taller than men, it is possible to see that the male population averages a taller height than the female population.

It could be fruitful to do a similar analysis with groups. Which groups have the highest percentage of loving kindness? Which groups have the lowest? Is there a correlation between loving kindness and wealth? Loving kindness and a particular spiritual belief? Loving kindness and a particular country? Loving kindness and a college education? If we were able to pinpoint the factors that lead to more loving kindness, we could work to increase those factors and improve our culture.

Now, I can’t do those studies. I don’t have the funding, and I’m not sure how loving kindness could be measured objectively to create a statistical base to analyze. So I will continue to assess the loving kindness of people on an individual basis, and seek to maximize it in myself and in my companions, while striving to notice and negate any prejudice I have that an individual is more or less likely to be loving and kind based on the group to which they belong.

Show your support

Clapping shows how much you appreciated Kris Williams’s story.