With 99 Measures of Well-Being, How Do You Choose?
Introducing a hierarchical framework of well-being.
Posted September 7, 2020 | Reviewed by Kaja Perina
Please note this blog post is co-authored by Dr. Fallon Goodman - who can be found at http://psychology.usf.edu/faculty/fgoodman and @FallonRGoodman
Understanding human well-being is one of the most important goals of psychological science. For good reason: most humans want to feel good, avoid feeling bad, and be free of maladies. Researchers seek to understand the psychological, biological, and environmental factors that influence whether a person experiences high well-being.
Well-being is central to human behavior. Our daily decision making is influenced by perceptions of what our well-being (and the well-being of others) will be in the future. Whether to hit the gym or binge watch Cobra Kai. Who to vote for in a presidential election. Which apartment or house to purchase. Well-being is increasingly the target of interventions. Do individuals who receive a training or therapeutic program end up, on average, better off than those who did not? What works best for whom, under what circumstances, and how? Until we unpack the “black box” of what is well-being, our ability to enhance well-being will be hampered.
One barrier to understanding well-being is the rich marketplace of ideas. Clinical psychological science has predominantly focused on maladies, disorders, symptoms, and dysfunction. A person without a mental disorder is assumed to be high in well-being, and a person with one or more disorders is assumed to be low in well-being. Well-being and distress have been pitted as opposite ends of the same continuum. Recognizing the need to empirically distinguish well-being from distress, Dr. Ed Diener outlined a model of subjective well-being (SWB), demonstrating that positive and negative affect operate relatively independently and offer unique contributions to the concept of happiness. SWB included positive and negative affect, but also cognitive evaluations of the quality of life, including specific domains (e.g., satisfaction with work, family, friends, romantic life, leisure) and temporal frames (i.e., past, present, future). Together, these three components – positive affect, negative affect, and life satisfaction – form SWB.
With a tractable model and corresponding self-report measures, Diener laid the groundwork for the empirical study of well-being. What is measured matters, and the science of well-being—often referred to as positive psychology—took off. SWB, however, drew criticism by philosophers and psychological scientists claiming important components of well-being were left out. Surely well-being could not be reduced to emotional experiences and a vague subjective judgment of how one’s life was going. Researchers introduced a whirlwind of new models and constructs. The result: as of 2016 (and more since then), researchers, therapists, coaches, and policy makers must wade through a staggering 99 published self-report measures of well-being with 196 different components. With over 99 measures of well-being to choose from, studying well-being is an intimidating task. Which are the best measures? Which model best captures reality? Which model should you choose for your specific research questions?
To define and model what well-being is, it is essential to separate out what it is not. Unfortunately, across subfields of psychology, the same measures are often used to both measure and predict well-being. The result has been a conflicting body of work on the components, causes, correlates, and consequences of well-being. To help address this problem, our team proposes a hierarchical framework of well-being that organizes existing models in a parsimonious manner. We illustrate how well-being can have a single, overarching well-being construct at the top of the hierarchy as well as distinctions between lower-level components, such as the Aristotelian distinction between ‘hedonic’ (e.g., positive affect) and ‘eudaimonic’ components (e.g., meaning in life). With a hierarchical framework of well-being, researchers and practitioners do not have to choose between higher and lower-order well-being. They can address (and measure) both general well-being and specific lower-order components that are relevant to a given situation, population, and/or research question.
A hierarchical framework of well-being
Our model proposes that well-being is hierarchical, with general well-being as a single factor at the top that subsumes lower levels of increasing specificity (Disabato, Goodman, & Kashdan, under review; preprint). General well-being is defined as perceived enjoyment and fulfillment with one’s life as a whole. The structure of this definition implies that well-being is 1) subjective, 2) about oneself, 3) about a person’s life as a narrative or story, and 4) includes affective (enjoyment) and nonaffective (fulfillment) components. Of less importance than the structure are the particular words. It is unlikely that researchers will agree on a precise definition of well-being, but our hope is that there is broader agreement on its features. Underneath the general well-being factor are four levels: lenses (different perspectives from which well-being is conceptualized), contents (homogeneous topic areas within the lenses), characteristics (clearly defined components of well-being that offer practical value in dissecting human experiences), and contexts (characteristics that arise in particular situations or contexts and/or a narrow aspect of a particular characteristic). For example, psychological well-being might be a lens of well-being; meaning-making as a content area within it; purpose in life, significance, and coherence as characteristics of meaning-making; and work-related purpose, work-related significance, and work-related coherence as contexts in which work related characteristics unfold. This hierarchical structure is akin to common models of intelligence (e.g., g-factor) and personality (e.g., Big Five), where different constructs arise at various levels of the hierarchy.
A question that naturally emerges from a hierarchical framework is what components fall under the well-being umbrella and what do not. We offer a pragmatic solution: define and measure well-being with constructs that are free from specific contexts. There is an important quality in our hierarchical model that is neglected in discussions about well-being—our model is agnostic about what leads to or causes well-being. We refer to this as a content-free approach. Well-being is personal and, by definition, subjective. People differ in what they value, strive for, and draw meaning from. For a religious person, organized religion may increase their well-being; for an atheist, organized religion may decrease their well-being; for an agnostic person, organized religion may be unrelated to their well-being. Each of us holds a complex set of beliefs about well-being that are rooted in our personal experiences, values, and cultural context. A content-free approach to well-being does not wash away these individual differences – content-laden activities and values such as religion remain as person-specific predictors of one’s overall, subjective assessment of how their life is going. Our approach ensures that well-being is a similar construct across people and does not differ based on individual preferences.
This approach sidesteps the unachievable task of reaching consensus among scientists and practitioners, a task embroiled with personal biases. These biases become clear after dissecting a small batch of the 196 components used to operationalize well-being in existing self-report surveys (Linton et al., 2016): achievement, eco-awareness, family, faith/religion, financial wellbeing, friendships, parenting, peace of mind, purpose in life, motivation, pleasure, sex life, trust, vacations, and work. Scientific progress is contingent on candid dialogue and careful scrutiny of research ideas. It becomes increasingly difficult to be an objective critic when the object of critique is intertwined with deeply held beliefs and personal investments. If we were to intensively study a group of people and test which components predicted their well-being, we can anticipate considerable variability. A team of researchers who decide a priori which experiences lead to well-being for human beings across the globe is almost assuredly biased.
Rather than pick and choose the ingredients of well-being, a less arbitrary approach that prevents tautologies is to measure well-being with measures that are independent of any (presumed) causes. In a model of well-being with content-free subjective measures, individual differences trump personal preferences – people derive well-being however they want. With our framework, we offer an opportunity for a less biased, more transparent, and clearer approach to answer questions about “what works,” “for whom,” “under what circumstances,” and “how.”