Identifying flexible predictors of academic success is important for helping students with learning differences as well as for creating targeted interventions. However, real-world limitations often make it necessary to cut down the number of factors researchers consider, while still keeping track of important differences.
In this study, researchers at University of California, Berkley combined machine learning (artificial intelligence that can learn and generalize from data) with a person-centered technique to identify key cognitive and motivational factors that set learners apart and predict early academic growth. Using a large and diverse dataset, machine learning identified three strong predictors: prior reading skills, motivation, and working memory. The researchers highlighted that if traditional variable-centered approaches — which treat factors as independent —were used, then this combination of factors would not have been found.
Kindergarten Learner Profiles
What they discovered were five distinct learner profiles that predicted growth in academic outcomes. To help interpret the profiles the researchers created descriptive labels, rather than diagnostic categories, based on the students’ standardized scores in prior reading achievement, working memory and motivation: 1. High Overall Functioning; 2. Moderate Strengths; 3. Low Motivation and Reading with Average Working Memory; 4. Improving but Initially Low; 5. Consistently Low Functioning.
These profiles showed significant stability across kindergarten and first grade regardless of socio-economic status and diagnostic categories. It is important to note, however, that this stability does not mean that the traits are fixed, but rather that the academic environment is stable and lacks targeted interventions. For example, students in Profile 3 (Low Motivation and Reading with Average Working Memory) could represent a hidden group of disengaged learners who have the cognitive abilities yet lack the motivation or foundational literacy necessary to fully engage with academic tasks. This could then result in them being overlooked in traditional assessments that prioritize skill deficits alone.
The researchers conclude that early learning is shaped by students' strengths and challenges in areas such as motivation, working memory and prior knowledge. They suggest that interventions need to begin by the start of kindergarten and should be based on how children learn, not just what they know.
StepUp Note
This research identified 3 key factors which help children become independent learners: reading skills, working memory, and motivation. StepUp programs reinforce each of these areas of learning. Group movement exercises encourage each child to remember as much as they can, without fear of failure. Fear of failure can limit a child’s ability to do their best work. Phonics and vocabulary exercises build the decoding skills which contribute to fluent decoding, and the vocabulary skills which enable them to match spoken language to written words. Finally, daily practice helps children gain both skills and fluency. Fluency turns skills into tools for new learning!
Note by Nancy W Rowe, MS, CCC/A