After years of considering/battling other terms, I decided to put my thoughts on the word "noncognitive" to paper.
When I wrote my last post, I took a risk that I don’t often take: I jumped into the topic of noncognitive skills without actually defining them. It was a moment I’d been anticipating for quite some time. See, when I first started working on noncognitive skills, the use of the term would result in people looking at me as if I had two heads. Today, however, most people either know the term “noncognitive” or feel sufficient social/professional pressure to act as if they do.
That being said, I wanted to devote the second part of this noncognitive series to the issue of terminology. For one, it’s an issue that comes up way too often. As I’ll discuss, people generally don’t like the term and/or don’t find it intuitive. While this isn’t unique, I fear it can dissuade people from exploring the issue and/or drive people to create research and practice from the existing work in this domain.
The second reason is rather personal. Every time someone says, “yeah, but ‘noncognitive’... what does that even mean?” I think to myself, “Do you think we haven’t thought of that??? Do you think that we haven’t tried countless other words in its place?” I’ve seen a host of terms from “psychosocial” to “21st Century Skills” to “soft skills.” Trust me, I would gladly accept any of these monikers if they didn’t have their own problems… which I’ll return to later.
Where does the term come from?
Before defending “noncognitive” as a term, there are a few important points to understand. Generally speaking, the term is used to refer to those characteristics that are unrelated to intelligence, and generally include an array of behavioral, motivational, attitudinal, and dispositional characteristics. You might be asking, “why is something defined by its lack of relationship with something else?”
The real reason is that, despite strong ties to psychology, modern references to “noncognitive” largely emerge from the economics literature. In economics and many other fields, intelligence is referred to as “cognitive ability.” With intelligence - or any other theoretical construct - there are both theoretical and operational ways to define such a vast construct. Theoretically - that is, what truly is “intelligence” - a definition can be difficult to establish. There are many competing theories about its complexity, development, and actualization.
However, when it comes to operationally defining intelligence - that is, how are we going to measure it - there are relatively few ways this is done. Researchers generally rely upon broad-based assessments of things such as critical thinking, problem solving, mathematics, and/or language skills. Thus, measures such as the Armed Services Vocational Aptitude Battery (or “ASVAB”), GED tests, SAT/ACT scores, or other broad-based academic measures commonly represent general cognitive ability in practice.
The term “noncognitive” comes into play when cognitive measures are unable to predict certain outcomes. One of my favorite examples comes from a paper by Nobel Laureate James Heckman and Yona Rubinstein (2001). When comparing occupational outcomes of GED recipients to high school graduates, they found that GED holders had lower earnings and poorer occupational outcomes than high school graduates. Noting that these two groups are similar on cognitive ability, the differences between these two groups must be something else… i.e., noncognitive.
Similarly, other work by Heckman and colleagues (e.g., Heckman, Stixrud, & Urzua, 2006) has shown that the effect of schooling (i.e., the number of years of school attended) on wages and employability is poorly accounted for by cognitive ability alone. Thus, what we gain from school isn’t just cognitive ability, but also these noncognitive characteristics.
Heckman is also generally quick to point out that the precise definition of these noncognitive skills isn’t exactly known. He and Rubinstein conclude their 2001 paper with the following:
“This paper is written in the spirit of “dark matter” research in astrophysics. We have established the quantitative importance of noncognitive skills without identifying any specific noncognitive skill. Research in the field is in its infancy. Too little is understood about the formation of these skills or about the separate effects of all of these diverse traits currently subsumed under the rubric of noncognitive skills. What we currently know, however, suggests that further research on the topic is likely to be very fruitful.” (p. 149)
What are noncognitive skills?
Heckman also notes in several studies that the lack of attention to and consideration of noncognitive factors is likely due to the absence of a unifying theory or measurement. With intelligence, for example, theoretical perspectives may differ, but researchers generally use a select set of measures to operationally define it.
Indeed, varying studies will refer to the importance of motivation, organization, perseverance, or social factors in determining success. This can create confusion about what noncognitive skills are, let alone which ones are of the most importance.
I’ve also thought that Heckman’s comments about the noncognitive space have important theoretical implications. The space of noncognitive skills is broad, diverse, and complex. Thus, the interpretation of many educational psychology efforts (e.g., grit, growth mindset, engagement) oversimplify what noncognitive factors are: a set of behaviors, skills, strategies, mindsets, and other factors that is just as diverse as the cognitive domain. When seeking to understand, measure, or otherwise address noncognitive factors as a whole, it is important to consider an array of variables and not simply one construct or theoretical model.
But how to do that? As you might suspect, there are many ways. As such, I have put together a resources page with references to several frameworks of noncognitive skills. I would highly encourage you to visit that for more information.
The Battle over Terminology
To come back to the issue of the term “noncognitive,” I would lastly like to defend its use. I’ve heard quite a few objections over the years, and cherish the opportunity to use this platform to address them. Personally, I certainly don’t think it’s the best option, but I do think that (a) many of the extant criticisms that have been put forth are not sufficiently problematic to prohibit its use; and (b) I’ve still yet to come across a better alternative.
Perhaps the first complaint is usually, “Well isn’t everything cognitive? How can anything be noncognitive?” This criticism is put to bed rather easily. As mentioned, “noncognitive” doesn’t mean “not dealing with cognition,” it refers to not related to academic ability.
The second approach to terminology is usually an effort to provide an alternative. But in this effort we return to the lack of a unifying theory. Most alternatives are tied to a theoretical perspective, and inevitably some individual or group disagrees with that perspective, thus prohibiting the practical use of the accompanying term.
There are several examples of this. You can continue reading, but let me assure you: we’ve tried them all and none seem to work.
Affective is a term that was introduced as part of Bloom’s Taxonomy, and seems like a reasonable alternative. However, any psychologist usually takes issue with this term because “affect,” in psychology, refers to emotion, and noncognitive skills are far more than emotional responses and strategies. I would argue that socioemotional, a term commonly used in the K-12 space, is similarly limited. (Full disclosure: I did once publish a paper referring to noncognitive skills as “affective.” That paper, and my points here, still stand.)
21st Century Skills is another alternative. One major issue here is that some have used this term to not only describe noncognitive factors, but also to describe a set of complex cognitive skills (e.g., creativity, effective communication, digital literacy) that were largely excluded from explicit attention in the “reading, writing, ‘rithmetic” curricula of the 20th century. For me, part of the allure of “21st Century Skills” dissipated when we actually entered the 21st century. Now that we’re almost a fifth of the way done with this century, the term has lost its aspirational zeal.
Psychosocial is the preferred term of my friend and colleague, Steve Robbins, though one colleague, when searching this term, came across a goth/heavy metal song with the same title… which pretty quickly quelled this term’s viability in our eyes.
Does language really matter that much?
In my opinion? Absolutely. As I mentioned, my biggest opposition to several of these terms is that they’re so easily shot down by critics. While the term “noncognitive” is in some ways diminished by its breadth, it’s also enhanced. So many theoretical models, constructs, and concepts can be folded into this umbrella, it’s almost impossible to say that these factors aren’t important.
Which brings us back to perhaps the most important point: no matter what, the evidence has clearly shown that noncognitive factors matter to academic and professional success, could provide a means of better supporting students and professionals, and are largely ignored in many of our assessment and curricular models.
To me, conveying this message is far more important than figuring out the best term. Let’s just accept it, move on, and hope that a better term comes along the way.
One final personal rant
There’s one last term that is my most loathed and feared option in this space… soft skills. It’s almost inevitable: when explaining my work to a layperson, I’ll wax poetic for several minutes, and at some point a light will go off in their head. “Oh… you mean SOFT skills.”
The use of this term has several important ramifications. First and foremost, the term “soft” implies that these things are vague and unknown. This is hardly the case. Despite the lack of a unifying theory, there are certainly many viable theoretical and operational models that can help a researcher or practitioner make these factors clear and meaningful.
The second implication is that they are “squishy,” not concrete like math, reading, or science, which we (presumably) can train and develop. Again, I will acknowledge that the development of interventions (likely hindered by the aforementioned lack of unifying theory) is perhaps the biggest issue in this field. But there are certainly examples of areas where interventions on psychological constructs have significant impacts on tangible human behavior (e.g., a host of references listed below).
Finally, “soft” does the most damage by diminishing these skills in the eyes of educators, policy-makers, and others. Most of us know and are working to address many of the challenges presented by a system that focuses solely on academic learning. Among those challenges, are exclusion, inequity, and attrition. Ongoing research into noncognitive factors has sought to diminish those negative outcomes. That message is the most important, far beyond whatever word we use to describe it.
References
Heckman, J. J., & Rubinstein, Y. (2001). The importance of noncognitive skills: Lessons from the GED testing program. American Economic Review, 91(2), 145-149.
Heckman, J. J., Stixrud, J., & Urzua, S. (2006). The effects of cognitive and noncognitive abilities on labor market outcomes and social behavior. Journal of Labor economics, 24(3), 411-482.
Mento, A. J., Steel, R. P., & Karren, R. J. (1987). A meta-analytic study of the effects of goal setting on task performance: 1966–1984. Organizational Behavior and Human Decision Processes, 39(1), 52-83.
Kleingeld, A., van Mierlo, H., & Arends, L. (2011). The effect of goal setting on group performance: A meta-analysis. Journal of Applied Psychology, 96(6), 1289.
O'Leary-Kelly, A. M., Martocchio, J. J., & Frink, D. D. (1994). A review of the influence of group goals on group performance. Academy of Management Journal, 37(5), 1285-1301.
Riebl, S. K., Estabrooks, P. A., Dunsmore, J. C., Savla, J., Frisard, M. I., Dietrich, A. M., ... & Davy, B. M. (2015). A systematic literature review and meta-analysis: The Theory of Planned Behavior's application to understand and predict nutrition-related behaviors in youth. Eating Behaviors, 18, 160-178.
Tyson, M., Covey, J., & Rosenthal, H. E. (2014). Theory of planned behavior interventions for reducing heterosexual risk behaviors: A meta-analysis. Health Psychology, 33(12), 1454.
Yeager, D. S., & Walton, G. M. (2011). Social-psychological interventions in education: They’re not magic. Review of educational Research, 81(2), 267-301.