Noncognitive Skills and Their Relationship With NAEP Performance by NAEP Achievement Levels

• In Event: Noncognitive Constructs and Academic Outcomes: Using NAEP Data to Understand Differences in Achievement

Abstract

Noncognitive constructs, often referred to as noncognitive or soft skills, are traits and patterns of thoughts and behaviors (Borghans et al., 2008) such as perseverance, and self-discipline. Recently, there has been a growing emphasis on development beyond cognitive skills, often referred to as academic skills, such as reading and mathematics. A growing body of research suggests an importance that noncognitive constructs play in students’ success in schools and for many aspects of students’ lives (Frank, 2020; Lee & Shute, 2009).

Although there has been quite a bit of research suggesting that noncognitive constructs are positively related to students’ academic achievement (Bohrnstedt et al., 2020; Zhang et al., 2021), many of them focus on the overall relationship between noncognitive constructs and achievement, often assuming that these relationships are the same for all levels of academic performance, which might not be true. This study examines whether these relationships vary depending on students’ academic performance.

Data for this study are from the NAEP 2019 grades 4 and 8 reading assessment including students’ responses on noncognitive construct items and their background characteristics gathered using NAEP student questionnaires. Five noncognitive constructs—persistence, academic self-discipline, enjoyment of complex problems, confidence in reading, and interest in reading—will be analyzed.

First, this study uses descriptive analysis to examine the distribution of noncognitive constructs by NAEP achievement levels (i.e., below the NAEP Basic, NAEP Basic, and NAEP Proficient and above). Then, the relationship between these constructs and NAEP performance is examined by NAEP achievement level (AL) to compare how the magnitude of these relationships differs by students’ achievement level using multiple group regression analysis. Additionally, any variations that may exist among students who perform below the NAEP Basic level are examined by equally dividing this group into three based on the NAEP reading score distribution.

Preliminary results suggest that the distribution of noncognitive constructs varies depending on students’ reading performance. For example, the percentage of fourth graders performing below the NAEP Basic level who reported low persistence is higher than the percentage of fourth graders performing at or above NAEP Basic with the same level of persistence, as expected. The relationship between persistence and reading achievement also varies by AL. For example, the relationship was stronger for students performing below the NAEP Basic compared to students performing at NAEP Basic or above. Grade 8 results show similar patterns. The entire analysis results for all five noncognitive constructs at both grades 4 and 8 will be presented at the conference presentation if accepted.

Findings from this study are expected to have two significances. First, the results will provide empirical evidence whether the same distribution of noncognitive constructs and the relationships between these skills and academic achievement can be assumed across students’ academic performance levels. Second, if distribution of these skills and the relationships vary across academic performance levels, the results can shed light on more effective educational policy and practice based on more accurate understanding of the relationship, e.g., more targeted and differentiated support depending on performance level.
 

Authors

• B. Jasmine Park, American Institutes for Research
  Presenting Author

• Alison Filbey, American Institutes for Research
  Non-Presenting Author

• Grace Ji Cheng, American Institutes for Research
  Non-Presenting Author