Alas, it’s not rocket science

Boris Johnson writes in The Telegraph that since we could get to the Moon, we should be able to get out of the EU: “They went to the Moon 50 years ago. Surely today we can solve the logistical issues of the Irish border”.

I sympathise with the sentiment in the argument. A lot of smart people – including a lot of social scientists – are working on solving complicated and difficult social issues, and it should be possible to solve the issue at hand. And why not expect this when scientists can solve complicated issues in the natural sciences? After all, it’s not rocket science.

However, there are very good reasons to why we cannot simply solve social issues that – intuitively – should be easy to solve. In brief, social science is much more complicated than a lot of the issues that we deal with in the natural sciences. We simply believe that we understand complex social phenomena when the truth is that we are not good at understanding or/and predicting such phenomena. Accordingly, when social scientists say that social science is not rocket science, we envy the simplicity of rocket science.

Duncan J. Watts describes this clearly in his book, Everything Is Obvious: How Common Sense Fails Us: “Well, I’m no rocket scientist, and I have immense respect for the people who can land a machine the size of a small car on another planet. But the sad fact is that we’re actually much better at planning the flight path of an interplanetary rocket than we are at managing the economy, merging two corporations, or even predicting how many copies of a book will sell.”

The problem is that Boris Johnson is assuming that the ontological parsimony in the natural sciences easily applies to the social sciences. There are specific reasons to which this perspective only works for the natural sciences and not the social sciences. Seva Gunitsky (2019), for example, describes why ontological parsimony works in the natural sciences: “The scientific version of ontological parsimony, most often associated with theoretical physics and mathematics (but sometimes imported into social science), argues that reality itself is governed by parsimonious physical laws. The fundamental physical nature of matter itself, at least at the subatomic level, possesses a symmetry that abets and even demands parsimonious explanations. Parsimonious theories that take advantage of this symmetry are appealing not just because they are elegant, but because they are more likely to be true.”

However, we cannot draw on the simplicity of the natural sciences to infer the potential to identify and suggest solutions to social issues. In the social sciences, we do not have the luxury of studying parsimonious physical laws. On the contrary, the social world is much more complicated. Acknowledging this is important if we are to actually understand and solve social issues – including the logistical issues of the Irish border.

Big Five personality traits in non-WEIRD settings

A new study, published in Science Advances, questions the validity of the Big Five personality traits outside of Western, educated, industrialized, rich and democratic (WEIRD) populations.

I was interviewed by New Scientist in order to give my take on the implications of the study. The article is available online.

New article in European Political Science Review: Bailout or bust?

Robert Klemmensen, Michael Baggesen Klitgaard and I have a new article in the May issue of the European Political Science Review. The article is titled ‘Bailout or bust? Government evaluations in the wake of a bailout‘. Here is the abstract:

Governments are often punished for negative events such as economic downturns and financial shocks. However, governments can address such shocks with salient policy responses that might mitigate public punishment. We use three high-quality nationally representative surveys collected around a key event in the history of the Dutch economy, namely the outbreak of the financial crisis in 2008, to examine how voters responded to a salient government bailout. The results illustrate that governments can get substantial credit for pursuing a bailout in the midst of a financial crisis. Future research should take salient policy responses into account to fully understand the public response to the outbreak of financial and economic crises.

You can find the article here. Replication material is available at GitHub and the Harvard Dataverse.

Potpourri: Statistics #56

Hands-on Machine Learning with R
The Truth About Linear Regression
Data Viz Book Reviews
Make Multi-point “dumbbell” Plots in ggplot2
City Intelligence Data Design Guidelines
shinyApp(), runApp(), shinyAppDir(), and a fourth option
Reordering and facetting for ggplot2
R Docker tutorial
S4: a short guide for the perplexed
Introducing trendyy
R you ready to make charts?
Ten more random useful things in R you may not know about
Testing statistical software
Tidy Data Science Workshop
Using tidyverse tools with Pew Research Center survey data in R
A Gentle Introduction to tidymodels
Changing Glyph in legend in ggplot2
Practical Data Science: an introduction to the PeerJ collection