An interview
We recently had a chat with Kai Horstmann, about his review article titled, “Assessing Personality States: What to Consider when Constructing State Measures”, which is scheduled to appear in an upcoming issue of EJP. Kai is a junior professor at Humboldt-Universität zu Berlin, Germany.
Read on to learn more about his work on personality state measures below!
Q: Hi Kai! Can you tell us a bit about yourself and how you became involved in personality psychology?
I am 30 years old, I live in Berlin and I currently work as a Junior Professor (this is similar to being an assistant professor outside of Germany) of the Psychological Assessment of Person-Situation-Dynamics at Humboldt-Universität zu Berlin. My research primarily focusses on the assessment and prediction of personality states. My other research areas include the description of situations and the development of measures for situational characteristics.
I got into personality psychology likely because one day, my former PhD advisor, Matthias Ziegler, invited me to work with him as a PhD student. Psychological assessment has always interested me very much, and I like the fact that psychological assessment plays an important role in so many – if not all – fields of psychological research. Now, my research focus is on the intersection of personality and assessment. This focus is also a result of my work with John Rauthmann who has also influenced me quite a bit (in a positive way of course, thanks, John!), but probably also due to the fact that measurement is valued so highly in personality psychology.
Q: What is your study about?
Psychological research has seen a strong increase in the use of experience sampling studies. In those studies, participants are repeatedly asked about their current emotions, feelings, thoughts, behavior, their surroundings, and so on. Although this research design is not new at all, it has become quite popular, possibly due to the fact that nearly everyone has a smartphone nowadays. In personality psychology, most experience sampling studies focus on the assessment of so-called personality states (that is, momentary manifestations of a trait, for example dancing as an expression of extraversion or reading a book as an expression of openness). And although we know a lot about how to design and evaluate measures for personality traits, we do not know that much about the development and evaluation of measures for states.
In our review, we take a look at how personality state measures have been developed and how estimates for their reliability and validity were obtained. Based on our review, we then highlight possible ways on how to estimate the reliability and validity of experience sampling measures in general. Take a look at validity, for example. When validating a personality state measure, most articles that we reviewed took the average of all states and correlated this with the corresponding trait. If the correlation is high, this is interpreted as validity evidence for the state score. But it is validity evidence for the average state score, and this is not the same. So, if one is interested in seeing if a state measure really captures the expression of a trait, then the individual state must also be validated. And this should not be very complicated: If one were to validate a state measure for neuroticism, it would be possible to examine students before and after the exam. The state neuroticism score of the students should be higher before the exam than after the exam. Of course, this requires the assumption that facing an exam leads to higher neuroticism, in the short term. This is also why we argue that validation requires a strong theory and hypotheses. We hope that, at least in the long term, some best practices will emerge.
We also show how personality theory can inform the development of an initial item pool for state measures. If a theory suggests a link between a trait and an outcome, then the state must necessarily be the link between the trait and the outcome. Conscientiousness is for example related to better physical health. Making a workout plan and buying groceries may be the state that links these two. We exemplify this in more detail by generating a set of items that could be used to assess conscientiousness states. We hope that this article will inspire further research on the development and validation of state measures, so that the field can develop some best practices for the assessment of (personality) states.
Q: Where do you see yourself in the (near) future?
Research-wise, I will continue to examine personality states and how personality states can inform personality theory. For example, I think it is a very interesting question to examine which of our daily thoughts, feelings, or behaviors form the picture that we have about ourselves, or in other words, which personality states will be integrated into the (self-reported) personality trait. Investigating this question will require more refined state measures as well as information about a person’s current situation, for example by obtaining informant reports about a person during their everyday life.
Concerning other areas of my life, I hope I can make it to the final four in German lacrosse once more, before this career comes to an end. But given the current situation with COVID-19, this will unfortunately not happen this year.
Q: Do you have any tips or advice for young researchers?
Having a great supervisor is probably most important, and I got really lucky with Matthias. At the same time, having a great mentor, (that is, someone who does not grade your work in the end) such as John was also very helpful. I think there are so many different approaches to writing articles, teaching, and so on, and realizing this has helped me a great deal. So, the first advice would be to find a mentor who can give you all of their honest advice and unfiltered opinions. This may not always be a pleasure, but it is very helpful.
I think it is similarly important, though, to make very good friends sooner or later within your group, people that are the same age or career stage as you are. Having friends in academia you can turn to and ask almost anything without feeling ashamed a tiny bit is the best thing that can happen to you.
Q: Who is your favorite scholar?
Well, this is a very difficult question, of course. There are so many excellent young scholars. However, because you probably want to have an answer to this question, I think it is Clemens Stachl, for me. Clemens works on extracting information on personality traits from mobile sensing data, using machine learning. His research is truly interesting and of extremely high quality – as far as I can tell. Moreover, he works with an interdisciplinary group of people, which I like. Also, he is genuinely a nice, humble, and friendly person I always enjoy talking to, and he also gives great workshops (trust me…).