• Question: Would you recommend becoming a statistical scientist, like yourself? Why? What are the benefits?

    Asked by samuel007 to Andy on 15 Jun 2011.
    • Photo: Andy MacLeod

      Andy MacLeod answered on 15 Jun 2011:

      Yes, I would. All scientists should know some statistics.

      Hi samuel007. My job title includes “statistical” because I do use stats a lot in my work. I’m looking for genes that have an effect on intelligence, but some of these genes have such small effect that we need to look at lots of people’s DNA to detect them. The study I’m working on at the moment has over 3,000 volunteers, but still that’s not enough to detect some of the genes. This is why we collaborate with other groups around the world, so we can pool our resources, increase the total sample size, and hopefully find more of the genes. With such large numbers, we need to use statistical tools and analysis to detect things.

      But I think all scientists need to know statistics. Even the most simple of experiments needs a statistical analysis to prove (or disprove) what you’re trying to test. Some scientists (even in this zone!) say they don’t like statistics, but that may be because they think of “statistics” as something seperate from “science”, whereas it should really be an integral part of it.