If you have been even peripherally paying attention to social science over the past decade, you may have noticed that there’s a big problem: “classic” findings from social science continually fail to replicate when a new team of researchers conduct the study. The replication crisis, especially in the field of psychology, has captured mainstream media attention since the first wide-scale replication effort was published in October 2015, with headlines stating, “Over half of psychology studies fail reproducibility test.”
In his latest book, Science Fictions: How Fraud, Bias, Negligence, and Hype Undermine the Search for Truth, Stuart Ritchie lays out how science – primarily the fields of psychology, medicine, and nutrition – has ended up at a place where only a minority of studies replicate, fraud is much too rampant, and metric-based incentives are driving (false) discovery. Ritchie focuses on four areas of concern: fraud, bias, negligence, and hype. Fraud, where a researcher fabricates data and experiments, is potentially the most disturbing of the four given that a too large minority of scientists erode the public’s trust in science, and can actively harm people – such as in the case of the notoriously fraudulent research by Andrew Wakefield allegedly linking vaccines to autism (vaccines don’t cause autism, please vaccinate your children).
The sections on bias and negligence have a strong focus on issues surrounding many replication issues in psychology such as questionable research practices, confirmation bias, and lack of error checking. The final section, hype, demonstrates how overblown, generalized, and sensationalized findings have also contributed to today’s problems. Uncertain, shaky, or preliminary findings are splashed across websites and social media feeds with little regard for nuance, caution, or skepticism. Those working in the field of psychology, in particular, and those who are not living under a rock or in denial of our field’s problems may not find much new information in these sections, but may very well find them, as I did, to be an extraordinary collection of the most pressing issues in modern science with interesting and relevant examples.
In the final section of his book, Ritchie tells us how we can fix science. These solutions, which largely center on ways to shift incentives to reward accuracy and rigor rather than pizzazz and quantity, are well underway in the social sciences: Open science, pre-registrations, registered reports, pre-prints, new priorities in hiring and grant awarding. These practices are actively changing the way new generations of researchers are trained, and shifting the norms of our field in a net positive direction.
Discussion of the replication crisis, particularly in the social sciences, however focuses with near exclusivity on research practices, for obvious good reasons. But, there is another area I believe scientists, and specifically academics and institutions of higher education, should also be focusing on: teaching scientific literacy. There are three specific reasons for a focus on teaching scientific literacy explicitly in the social (and hard) sciences (and with its own dedicated course, I would propose):
- Teaching scientific literacy aligns with the goal of de-incentivizing “hyped” findings and sensationalist claims
- It is a practical skill that adults can use and apply in their daily lives when browsing social media and listening to the news
- It is necessary for informed policy decisions and informed voter decisions
These may sound idealistic, and they probably are to an extent, but active participation by scientists and academics in teaching scientific literacy in the classroom and on social media is a goal that we can’t let slip by. Poor scientific literacy is part of the reason why there is an anti-vaxx movement, “climate change is a hoax”, misinformed policies, and wasted resources implementing ineffective (and maybe harmful) interventions. Poor scientific literacy also leads to people mindlessly click “click-bait” headlines from which they train their social media algorithms to populate their feeds with more and more misinformation.
Better scientific literacy is badly needed, yet is not actively taught. My experience in this domain is largely restricted to the field of psychology and social science more broadly, but I want to share how I integrated scientific literacy training into my courses when I was teaching to provide some examples of how academics can work toward this goal. (P.S. I miss teaching!).
The first is basic but too often overlooked: Teaching students how to read a scientific paper. Delightfully, Ritchie includes an Appendix in Science Fictions on just this topic – what a delight! Why is this important? Because students do not know how to read a science paper when they enter your classroom. Let me repeat: students do not know how to read a science paper when they enter your classroom. If they are not able to read science papers, how can you expect them to fact-check flashy headlines and news articles? They can’t. If you do one thing, teach them how to read a paper. (This is a bare-bones outline of a Scientific Literacy lecture I used in my recent Intro to Psychology course. See also the end of this lecture for demonstration of sharing the replication crisis and debunking some common “facts”).
Once they know how to read a scientific paper, you need to demonstrate and engage them with the fact-checking process and show them how limited and misleading reporting often is. In all my psychology courses I used this Psychology in the News Assignment – it was remarkably successful. Students were shocked by how limited reporting was, they learned a new skill (reading a science paper), and they saw how inaccurate reporting can be. Knowledge they will (hopefully) use when scrolling Facebook and Twitter.
Lastly, teach, reinforce, and engage students in interpreting real tables and figures. Knowledge in psychology changes fast. Teaching students “facts” is far less useful than teaching them scientific literacy skills which they can use to update their knowledge of the field after they leave your classroom. A few ways I have practiced this: My slides show the actual figure or table from a paper I am discussing, rather than a bullet point of the finding. I spend time talking through the labels, error bars, measures, and then ask them to tell me the result. (You can look through my Human Development slides for examples.
Most importantly, I think, is that I frame my class as skill-based. I do this with my Introduction to the Course lecture where I explain my goals, how I am going to achieve them, and how each aspect of the course – lecture styles, active learning, assignments, etc – connect to those goals. The result is more engaged students that walk out of the classroom with useful skills they can apply in life and other courses, rather than only “facts” that may be debunked tomorrow.
These are only examples of ways that I have worked to actively teach scientific literacy in my role as an educator. If it were my department, I would devote a whole course to scientific literacy, and in said course I would assign Ritchie’s Science Fictions as an accessible base reading for the course. Anyone interested in the replication crisis (and those who have their head in the sand about the field’s pervasive issues) should read Science Fictions. The replication crisis has brought about myriad reforms in the field of psychology. We should also ensure that we’re focusing on science education, too, as scientific literacy is much aligned with the reforms we’re already making.