An interview
EJP's special issue "Personality and Social Structure" is out now, and we talked to Rodica Damian about her paper, "Whose Job Will Be Taken Over by a Computer? The Role of Personality in Predicting Job Computerizability over the Lifespan". Rodica is Assistant Professor of Psychology at the University of Houston.
Q: Hi Rodica! Can you tell us a little bit about your study?
According to economists, 19% to 49% of jobs are at risk of being computerized over the next 10 to 20 years. We had this large dataset that included 346,000 individuals from a representative US sample. They were first assessed when they were 16 years-old, their job titles were recorded 11 and 50 years later. We used something called "computerizability scores" -- these had been previously assigned to American job titles by a team of economists working together with scientists in robotics and machine learning and indicate each job’s probability of being taken over by computers in the next two decades. We found that people who were more interested in science and arts, who were more extraverted and conscientious, more intelligent, and of higher socioeconomic status tended to select into jobs that had a lower probability of being automated in the future. Education accounted for about 60% of each of the effects, but direct effects of personality traits and vocational interests remained.
Q: What do you think are the implications of your findings?
These findings suggest that training vocational interests and personality traits could potentially help people to be more prepared for the future labor market. Specifically, education could potentially be improved to better prepare people for the labor market by training more STEM interests, artistic interests, and promoting personality traits like conscientiousness. Future studies would have to investigate what interventions might be best suited for achieving these goals. Throughout history, people have always been able to create new jobs in the face of automatization. People haven’t faced complete unemployment, because they have managed to create new jobs through increasing levels of education. But, if the current education doesn’t cut it anymore, we will have to change education so that people have the skills to create new jobs.
For people to remain employed, I think it is important to already start preparing for the future generations so that they have the skills that make them adaptable. And of course, you don’t know for sure that even if you prepare people, that is a guarantee that the labor market will be able to absorb all these newly trained persons. For example, maybe we don’t need a billion scientists! But even if that is the case, I think that more targeted education will allow for more people to have at least a fighting chance.
Q: What made you interested in the computerizability of jobs?
The idea came to me a couple of years ago when I was reading The Economist one evening. The cover depicted a tornado sweeping away office clerks. The accompanying article featured a discussion in which economists were really worried. What I noticed was that there was no research available on what you could do to better prepare people, and what skills you might need to train to make them more adaptable to the changing job market. I am generally interested in what makes people thrive, including in the workplace, and in general well-being. So, this subject really fits with my interests. Additionally, because I have friends who work in different fields, I am often encouraged to tackle problems that are not only interesting to psychologists, but that are also interesting to the broader public.
Q: What are your future plans for your research?
I have some ideas for this particular line of work. I believe creativity might also be important in predicting selection into less computerizable jobs – although it was not measured in the research we are talking about. I would also be interested in replicating the study’s findings in samples from other countries, and in longitudinal studies with more than two time points so I can investigate how changes in skills relate to changes in job selection. I also think that it would be interesting to look at the reverse effects. For example, it might be that some jobs help you acquire skills that are related to selecting into less computerizable jobs in the future. Going forward, another area I’m interested in is diversifying experiences which affect your life trajectory. Diversifying experiences are life experiences or turning points in people’s lives, ranging from adversity, like trauma or abuse, to eye opening experiences like fending for yourself living in a foreign culture. These events seem to often bend the arc of people’s lives, sending some people into downward spirals, while propelling others to achieve great feats. I want to understand what it is about people that marks them for these different paths when they go through the same diversifying experience. Once we understand what can help or hurt people’s success and well-being in response to the life experiences thrown at them, we can develop useful interventions.
Q: Is job computerizability a relevant topic in your own professional life?
I think we are pretty safe from computerizability as researchers in psychology – barring some unpredicted scientific revolution – because our jobs involve creativity and a lot of non-routine tasks. In fact, economists found that the job “Research Psychologist” had a probability of computerization of only 0,004 and was ranked 17th in the list of all 925 jobs coded for computerizability, where rank #1 was the least computerizable job. So, that’s nice!
On the one hand, I do want to say that computerizability can play a huge role in helping us perform the more routine parts of our jobs in faster, and more efficient ways. The most obvious places where computerizability has had an impact is in data collection, data analysis and open science. With data collection we have gone from paper-and-pencil to the administration of online questionnaires, and even looking at people’s social media traces. And then, in terms of data analysis, because we have better statistical packages we are now able to fit models that were not even imaginable 10 years ago. That’s amazing! However, on the other hand, I also think that computerizability can sometimes be dangerous because it lowers the entry bar for doing psychology, as people can just use stats packages in combination with rules of thumb. I think this has contributed to the replicability crisis. To use a really graphic metaphor: it’s like using a chainsaw versus an axe. You can saw quicker but you can also injure yourself if you don’t use it properly. But, in terms of open science, I think computerizability has given us the tools to really practice open science as we can easily store large datasets online. We now have less excuses not to share files, because it takes only a second to upload them.
Q: You've been an assistant professor at the University of Houston for a couple of years, now! Do you have any advice for young researchers?
Haha, I’m not sure if I’m experienced enough to impart advice to “young researchers,” am I that old? Also, there is no advice that works for everyone, but I will try.
Enjoy grad school and especially your post-doc (I 100% recommend postdocs if you can find a good mentor), I think we don’t realize as students and postdocs how much time we have for research, so try to make the most of that. This may sound trivial to some, but I don’t think it’s obvious to many students: The more you advance in your career, the more responsibilities fall on you, so even though grad school and your postdoc might seem like a lot of work, it actually can be a relaxing time to dedicate to research, when you are protected in your advisor’s bubble, assuming you are lucky and have a good advisor. Learn a lot of statistics and keep learning more statistics all the time. Find research topics that you love, do your work thoroughly, work as hard as you can to be productive, try to have a fun life and plenty of social support, and be persistent, don’t get discouraged, and apply to many jobs, and then apply again, until you find your place.