Back in March, I made a presentation to the Fair Vote Coalition on Proportional Representation. The gist of my presentation was that the mirror theory of representation might not be correct. In other words, women think a lot more like the men around them then they do like other women. Education and income explain 19% of voting behaviour, as compared to gender, which explains 3%. As a result, the dreams of many activists for women friendly policies might not come true, even if we elect a lot of women to government. Here are my slides if you’re interested.
My audience was astounded. Both the men and women just didn’t believe that other women disagreed with their political positions. Even after seeing data from the National Election Study. Recently I found a study by Sharad Goel, Winter Mason and Duncan Watts described on a blog called Messy Matters, that explains this reaction. Using Facebook as a data frame, the investigators found that we consistently over-estimate the likelihood that our friends agree with our political positions on certain issues. In fact, participants were unaware of the disagreement 60% of the time.
As a result, we make a lot of assumptions about what’s right for “us”, the social group with which we identify. And what we find when we ask people sometimes surprises us. This causes leaders to make bad decisions and bad policy. We need to stop making assumptions about people’s beliefs. We need to know not only their beliefs, but also their behaviours. Then we can make real, meaningful and lasting change.