By Alex Fradera
Part of my role at the Digest involves sifting through journals looking for research worth covering, and I’ve sensed that modern social psychology generates plenty of studies based on questionnaire data, but far fewer that investigate the kind of tangible behavioural outcomes illuminated by the field’s classics, from Asch’s conformity experiments to Milgram’s research on obedience to authority. A new paper in Social Psychological Bulletin examines this apparent change systematically. Based on his findings, Dariusz Doliński at the SWPS University of Social Sciences and Humanities in Poland asks the bleak question: is psychology still a science of behaviour?
Doliński is following in the footsteps of a paper published over a decade ago, which looked at the January 2006 contents of what Doliński describes as social psychology’s “flagship” publication, the Journal of Personality and Social Psychology, published by the American Psychological Association. Roy Baumeister and colleagues sought research that involved behavioural outcomes more active than “finger movements” – typing on a computer, clicking a mouse or filling in a survey by hand. They found active behavioural outcomes were measured by only 12 per cent of studies published in the journal that month, compared to an estimated 80 per cent in the same journal 30 years earlier.
Was this a blip, or a trend? Doliński looked at six issues of the Journal of Personality and Social Psychology published in the second half of 2017. He found that from 45 empirical articles, only four went beyond participants answering questions or filling in surveys; overall, only 6 per cent of the analysed studies measured actual behaviour.
Moreover, when behaviour was studied, it was frequently nothing more than another case of finger movement. One study measured whether people would over-report their achievements on an experimental task; another looked at endurance in solving cognitive tasks; a third involved the “prisoner’s dilemma”, a financial game that doesn’t require any overt behaviour as such. Only one study looked at a truly active behaviour – the social interactions of pre-school children. Doliński likely had his tongue only partly in cheek when wondering whether this approach would still have been used if pre-school children were comfortable filling in questionnaires.
What’s behind this anti-behavioural turn? Some of it is connected to the cognitive revolution: the recognition in psychological science that internal processes like memory, learning and attention are crucial to understanding behaviour. This has been immensely useful and worth celebrating, but it doesn’t exclude studying a diversity of behaviours as well.
Doliński raises a few other obvious pressures that are harder to celebrate. Studying behaviours is more difficult, and often more costly and expensive, than studying verbal declarations. Behaviours are also harder to verify (was that spontaneous gesture friendly or unfriendly?) and to record. It may also be harder to convince ethics boards to approve a behavioural study than another survey study. In addition, observed behaviour is often binary and one-shot per participant. You offer help to cross a street; you return the found item; you go to vote; you sign the petition,… or you don’t. Not only does this often call for larger sample sizes, it limits how you can analyse the data, making it harder to produce persuasive models of pathways between mechanisms.
As a concrete example of how the behavioural instinct has been blunted, Doliński points to a piece of research on physical intimacy (from the journal issues he reviewed) that did make the effort to go into the field. Experimenters visited a clinic providing flu shots, but did they observe how waiting patients interacted, or how closely they approached the receptionist or stood in line? No: they simply asked people more questions.
This matters because declarations of behavioural intentions do not always translate to behaviours. For instance, lottery winners rarely donate their cash to charity, even though more than a quarter of survey respondents claim they would give substantially. Also, while the very robust “bystander effect” shows that people are less likely to act to counter a transgressive act when others are present, people asked to describe their hypothetical actions believe they would be as likely to act whether several other people are also present or not. Similarly, recent work canvassing opinions on the Milgram experiment showed that most people believe they would not have carried out the experimenters’ orders, even though most people do.
Doliński worries that social psychologists doing the important work of field studies, and those studying behaviour more generally, are being shut out of the best journals in their field. Social psychology is already going through a process of reform as it responds to concerns about research quality. Perhaps it should take this opportunity to also consider adjusting its incentive systems to make more of research that does the best it can to see how psychological manipulations impact the world.