Posted by: salamandercandy | December 26, 2005

Statistics

Modern science is grounded in statistics. Scientists can’t conclude much from their observational or experimental results unless they can show with statistical significance that a competing possibility is unlikely to be true. This entire concept is largely foreign to non-scientists. I learned plenty of math in high school, but statistics is quite different from deductive math, and most people never really learn anything about this discipline which is the basis of science.

When most people think “statistics,” they think percentages. 37% of children prefer hot dogs, 27% prefer hamburgers, 78% of people believe in aliens, 36% of Americans approve of the Vice President’s performance, etc. But figures like these are not the fundamentals of statistics. Statistics is about testing whether figures like these fit the observed data significantly better than some other value. How likely is it that children actually enjoy hot dogs and hamburgers equally, but the pollsters happened to ask a larger number of frankfurter-lovers by chance? If you can’t answer that, you aren’t really doing statistics yet.

Usually, there is one explanation that is the simplest, called the null hypothesis. You can’t really show that the null hypothesis is true, but in science you kind of have to act like it is until you can show that it is extremely unlikely to be true (usually there needs to be less than a 5% chance that you would observe the data you have if the null hypothesis were true). The simplest explanation is the one that supposes the existence of fewer things: “dinosaurs never roamed the planet;” “global warming isn’t happening;” “there is no God.” We can only reject these statements scientifically if we have overwhelming evidence to the contrary. It wouldn’t be hard to teach some basic statistical concepts in public schools, and this would greatly help people understand how scientists decide what is true, what is false, and what can’t be known with the available evidence.

Advertisements

Responses

  1. I couldn’t agree more! The concept of the null hypothesis is not widely known. I don’t think I even learned it till college. The better understanding of this would contribute to a more rational society, I think!


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Categories

%d bloggers like this: