Keeping You

Lifestyle: Do you generalise from small samples? Everyone else does

14 Sep 2018

I am as guilty as anyone when it comes to generalising from small samples. I observe something several times and I make it the rule. I am quickly convinced, especially when it occurs again, exactly as I predicted. My generalisation quickly becomes a rule and then a law.

I set myself up for nasty surprises when I do this sort of thing and it is particularly ignominious when I have just announced the obvious (to me) in public only to discover that what I thought was a ‘law’, was just a big, fat generalisation.

Generalisation from small samples is a widespread human bias which means a lot of our understanding of the world is based on incomplete and often erroneous information.

Polls are a good example of making generalisations from small samples and how often are the polls wrong? They are always run off small samples because running polls is expensive. They are also dependent upon the method used to find people to question and on the type of people that get to take part.

I am the one in our office with the most experience and I was reflecting on my situation. Am I right more often than my colleagues? Do I have deeper and better insights? Jared would say, “Definitely not.” But what do I think about this? I think the main benefit I bring as the experienced one is the understanding that none of us really knows the definitive answer to anything. We never have enough information and we don’t have the time to analyse it fully. We must act, of course, often based on about 30% of the necessary information, and we must give our actions time to take effect. But then, on review, we often must modify what we did, or do it again.

Never more so than in bringing up children. I have had a fair amount of practice in this department. In my latest venture, I am bringing up a second family, a 10-year old daughter Frances in Dunedin, with Amanda. I am often proved wrong when, using my experience from my first family, I have suggested rules learned long ago with my three grown-up boys. Different sample. Different city. Different wife. Different everything. Down here I just had to start again, from scratch and learn new ways of doing things, the hard way.

In bringing up Frances, no matter what experience I bought to this new baby game, I had the advantage of working with someone who had a lot of skin in the game. It wasn’t just me running the show and Amanda has spent a lot more time with Frances than I have. But we must be doing something right. In nearly every report Frances has ever had, her teachers mention that she is popular with students and teachers alike, because she is kind.

Still, I have to remember, it is a small sample.

“Me mahi tahi tatou mo te oranga o te katoa” – “We should work together for the wellbeing of all people

Keep asking great questions …

Read more articles in this fortnight's edition of 'News Farmers Can Use':