Riposte on the Spirit Level

Kate Pickett, one of the co-authors of Spirit Level, has emailed me to say I misrepresented some of the book’s statistical arguments in my 5 March post. I did point out that the post makes clear it is based on a limited range of sources – I have not read the book yet. However, in the interests of a fair and balanced debate, I have pasted her statistical rebuttal below. I hope to review the book properly before too long:

“First, our Index of Health and Social Problems does not contain happiness, it is based on hard, factual data from reputable sources like the World Bank, OECD, UN, etc. It contains life expectancy, infant mortality, obesity, mental illness, imprisonment, homicide, teenage births, educational scores, social mobility (correlation between father’s and son’s incomes over 30 years) and trust. I think the only one that is arguably a “soft” outcome is trust. This is based on official surveys of random samples of the population who say that other people can be trusted or not. But really, if someone says they don’t trust other people, they probably don’t. And indeed the causal impact of inequality on trust has been demonstrated by others. We show relationships with income inequality for all of the outcomes in our index separately as well as when combined, and we show the same for the 50 US states, as well as rich market economies.

“Second, although definitions of mental illness do indeed change over time, we use data from the World Health Organization’s Consortium on Mental Illness, which used the same psychiatric diagnostic interviews in population samples of several different countries at the same time, so the data are certainly comparable and are not simply measures of how people are feeling.

“Third, we do not, of course, only use evidence from primate studies to help us understand how status insecurities and anxieties can affect our behaviour and biology, but they can be very useful. It would be silly to think that we do NOT have an evolved response to social status and social interactions and anxieties. I imagine while we all sat in the Moral Maze green room, our adrenaline and cortisol were running pretty high, all due to our feelings about whether or not we were going to make a good showing on the programme and how we were going come across and be judged!”