My Research
My research addresses the question of who flourishes in life (physically, mentally, and socially), why, and what enhances or hinders healthy life trajectories. My research involves several different research threads:
My research is constantly evolving. For more on my work to date, see my publications and presentation listings. Below are brief summaries of my current lines of research. If you would like more information on these projects or if you are interested in getting involved in my research, please contact me.
- wellbeing measurement;
- big data approaches to psychological study;
- long data approaches for testing sophisticated theories of psychosocial processes underlying health over time; and
- whole school approaches to wellbeing.
My research is constantly evolving. For more on my work to date, see my publications and presentation listings. Below are brief summaries of my current lines of research. If you would like more information on these projects or if you are interested in getting involved in my research, please contact me.
Understanding, Measuring, and Building Health & Well-being
One of my primary areas of research focuses on defining, measuring, and building well-being. Individuals, organizations, and leaders from around the world are turning their attention toward the potential value of assessing subjective well-being. Over the past several years, my colleagues and I developed measures of adolescent and adult flourishing (the EPOCH Measure of Adolescent Well-being; the PERMA-Profiler). I have had a tremendous positive response to these measures, and I am now extending the measures in numerous ways.
The PERMA-Profiler
|
Dr. Martin Seligman, father of the positive psychology field and one of my postdoc advisors at UPenn, has suggested that human flourishing is a multidimensional construct, captured by the acronym PERMA: positive emotion, engagement, relationship, meaning, and accomplishment (Seligman, 2011). Although the domains are debatable, the key argument in this model is that well-being is not a single number indicating how happy a person is, but rather comprises multiple elements, such as finding meaning in life and having good social relationships.
We have developed a brief 23-item questionnaire to directly measure PERMA. The measure has been translated into multiple languages (including Spanish, Chinese, Korean, Dutch, Polish, and Czech). For a talk that I presented on the measure (February, 2014, SAHMRI Research Institute, Adelaide, Australia), see this YouTube video. The measure is now available for use (see my Questionnaires page). |
|
|
The EPOCH Measure of Adolescent Well-beingAdolescence is a very important, though often overlooked, period of life. Building upon the PERMA model, and in collaboration with Dr. Larry Steinberg at Temple University, we developed the EPOCH Measure of Adolescent Well-being. The measure It assesses five positive characteristics that we believe promote health and well-being throughout life (engagement, perseverance, connectedness, optimism, happiness). The measure is now available for use, and is being used in numerous projects in Australia and the US to measure and build well-being (see my Questionnaires page).
|
Workplace Well-beingThe PERMA-Profiler was developed as a general measure of well-being. In collaboration with several organizations, I developed a modified version of the measure, which shifts the context to the work setting. The measure has been translated into French and Spanish, and studies demonstrated the value of the measure in the workplace. Several of my ongoing studies are testing how to effectively use the measure to measure and build employee well-being and company performance. You can take the measure here to find out your profile and ways to improve your wellbeing.
|
|
Big Data Approaches to Psychological Science
Taking measurement to a whole new level, in an exciting collaboration with computational linguists at UPenn, we have created a method that combines big data available through online social media, analytics available from computer science, and theories from psychology to unobtrusively measure and study psychosocial characteristics using online social media such as Facebook and Twitter.
|
As one of the psychologist on the team, I have found that computer science has tools to analyze the growing amount of data available, and social scientists have insights to actually make sense of the information. By combining techniques from computer science with psychological theories, we can investigate the human psyche in an increasingly technologically driven world.
In a first study, I used our open vocabulary approach to identify and visually summarize the dominant naturally occurring words and phrases that most distinguished each Big Five personality trait. The results really blew me away. Distinguishing words and phrases were face valid and provided insight into processes that underlie the Big Five traits. For example, the word clouds to the left show words that distinguished people high on extraversion (top) and high on neuroticism (bottom; this is the PG rated version) |
|
In a second study, I examined age-related differences in language use, and illustrated how the method could be used to test developmental theories (Kern, Eichstaedt, et al., in press, Developmental Psychology). As part of our approach, we can automatically create categories or topics. In the images below, the middle cloud are words distinguishing teenagers (age 13-18; top) and young adults (age 23-29; bottom), and the surrounding clouds are topics distinguishing each group.
We can also trace words across different ages (below). See wwbp.org for more! 
|
|
|
Particularly relevant to social psychology and health, our most recent work examines county level language in relation to life satisfaction, heart disease, stroke, and more. For example, we were able to predict atherosclerotic heart disease mortality better than 10 common predictors, such as gender, socioeconomic status, and health behaviors (Eichstaedt, Schwartz, Kern, et al., 2015, Psychological Science). The image to the right illustrates risky (left) and protective (right) topics of words (again, these are the PG rated versions). Hostile and apathetic words predicted risk, whereas positive experiences related to lower risk.
|
For details on the project, see wwbp.org. And for all sorts of computational social science resources, see lexhub.org.
Long Data Approaches to Understanding Life Trajectories
|
An important question is how physical, mental, and social well-being unfolds over time. That is, what is a healthy life trajectory, and how is that influenced by individual characteristics and social contexts? Where as big data looks at a little information on a lot of people, long data involves extensive information collected on a small group of people over long time periods. My studies have demonstrated that understanding individual trajectories within the context of personality, past experiences, current situations, and future expectations is imperative. To help individuals thrive, we need clearly specified theories of predictors, mechanisms, and outcomes. This approach ensures that resources are spent on the areas that will be most effective in cultivating valued outcomes. Along these lines, my colleague Dr. Howard S. Friedman and I published a foundational article in Annual Review of Psychology, and through various collaborations, my work continues to grow in this domain.
|
|
|
The Terman Life Cycle StudyMy graduate work focused on understanding and predicting healthy aging and mortality risk. Much of my work used a rather unique lifespan dataset: The Terman Life Cycle Study. The study was begun in 1921-22 by Lewis M. Terman at Stanford University as a descriptive study of gifted children. Over 1,500 participants were followed prospectively throughout their lives, completing assessments every five to ten years. We collected death information for most of the participants and developed various psychosocial measures, including child and adult personality, physical activity, healthy aging, and social support. As part of my dissertation, I spent quite a bit of time at the archives collecting data (see pictures from the archives to the left, including ancient file cabinets in an attic that hold materials dating back to the early 1920s (top), and Terman's bust (bottom), which one of the more artistic participants created). My former advisor, Dr. Howard S. Friedman, and colleague Dr. Leslie R. Martin, published a book, The Longevity Project, which describes what we have learned about longevity through the project - I highly recommend it (see Dr. Friedman’s website for more information). Several of the chapters are based on my work. See my publications for articles I have written using this unique dataset.
|
The Harvard Grant Men StudyAnother amazing longitudinal dataset is the Harvard Grant Study, which began in the 1940s. Men in the 1939-1944 classes at Harvard University first completed an assessment, and then were followed throughout their lives, with numerous measures including self-report questionnaires, in-depth medical and psychiatric interviews, and more. Many of the findings in the Terman and Grant studies support each other. George Vaillant, who has worked with the sample over the past 40+ years, recently published an excellent book on the sample. I am fortunate enough to have the opportunity to work with Dr. Vaillant, continuing to study and learn from this amazing group of men. In a project funded by the Templeton Foundation, we examined prospective beliefs in the afterlife.
|
|
Combining the Terman Life Cycle Study with the Hawaii Personality and Health Study. Conscientiousness predicted better health across a 40 year period, partially mediated by education and alcohol use.
Integrating data from 4 studies with over 23,000 participants to examine correlations between adolescent perseverance and adult physical health & behaviors.
|
Integrative Data AnalysisTaking longitudinal data further, my recent work explores innovative methods for combining studies. By taking advantage of existing studies, we can understand how people are traveling in life, and better determine the best points and avenues of intervention. Although it is extremely challenging to work with multiple datasets, by incorporating multiple studies, cohort and study specific effects can be tested, and replicable effects can be distilled. In turn, by approaching existing data with more complex theoretical models, we will better understand factors supporting or changing healthy trajectories, and subsequently improve our measures and intervention strategies.
In a first project, in collaboration with Sarah Hampson, Lew Goldberg, and Grant Edmonds at the Oregon Research Institute, we integrated data from the Terman Life Cycle Study with the Hawaii Personality and Health Study (Kern, Hampson, Goldberg, & Friedman, 2014, Developmental Psychology). Building from this work, we developed TADPOHLS: The Anatomy of Developmental Predictors of Healthy Lives Study, which is a database that categorizes items from longitudinal studies that followed participants from adolescents into adulthood and included measures of psychological well-being, physical health, and psychosocial variables. We developed an extensive classification system, and coded information from 25 different studies. A paper on the database was published in Applied Developmental Science, we are finalizing the database for release to the research community. |
Whole School Approaches to Wellbeing
|
Applying my background in positive psychology, skills in research and methodology, and work with teachers and schools, my research is increasingly focusing on how we can affect positive change across entire schools.
As part of this, we are thinking about how to bring together positive psychology and systems thinking to affect change across educational communities, through what we are calling Positive Systems Science. Check out our blog for updates. |
|
|
Final ThoughtsThroughout my training, I have developed an extensive statistical toolbox, which I creatively apply to address challenging research questions. I take advantage of advanced quantitative techniques, including growth curve analysis, structural equation modeling, survival analysis, meta-analysis, integrative data strategies, and data mining techniques. By understanding what works, for whom, and why, interventions will be better informed and ultimately more effective.
I like to think of my research as a puzzle. I get involved in different projects, often moving in multiple directions at the same time. I apply different methods, address different research questions. Each project is a piece of the puzzle. Over time, the full picture of my research career will come together. It's a journey - stay tuned to see where it leads.
I end with a final quote, by Benjamin Franklin: "Wish not so much to live long as to live well." My hope is that through my research, I can somehow help others to indeed live well, through whatever the journey of life brings. |
Contact me for more information on these projects, or if you are interested in getting involved in my research.
