skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Explore Research Products in the PAR
It may take a few hours for recently added research products to appear in PAR search results.


  • NSF Public Access
  • Search Results
  • Math Content Readability, Student Reading Ability, and Behavior Associated with Gaming the System in Adaptive Learning Software

This content will become publicly available on July 14, 2026

Title: Math Content Readability, Student Reading Ability, and Behavior Associated with Gaming the System in Adaptive Learning Software
Recent research on more comprehensive models of student learning in adaptive math learning software used an indicator of student reading ability to predict students' tendencies to engage in behaviors associated with so-called "gaming the system." Using data from Carnegie Learning's MATHia adaptive learning software, we replicate the finding that students likely to experience reading difficulties are more likely to engage in behaviors associated with gaming the system. Using both observational and experimental data, we consider relationships between student reading ability, readability of specific math lessons, and behavior associated with gaming. We identify several readability characteristics of specific content that predict detected gaming behavior, as well as evidence that a prior experiment that targeted enhanced content readability decreased behavior associated with gaming, but only for students that are predicted to be less likely to experience reading difficulties. We suggest avenues for future research to better understand and model behavior of math learners, especially those who may be experiencing reading difficulties while they learn math.  more » « less
Award ID(s):
2000638
PAR ID:
10658123
Author(s) / Creator(s):
; ; ; ;
Editor(s):
Mills, Caitlin; Alexandron, Giora; Taibi, Davide; Lo_Bosco, Giosuè; Paquette, Luc
Publisher / Repository:
International Educational Data Mining Society
Date Published:
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; ; ; ; ; ; (, International Educational Data Mining Society)
    Mills, Caitlin; Alexandron, Giora; Taibi, Davide; Lo_Bosco, Giosuè; Paquette, Luc (Ed.)
    Students' reading ability affects their outcomes in learning software even outside of reading education, such as in math education, which can result in unexpected and inequitable outcomes. We analyze an adaptive learning software using Bayesian Knowledge Tracing (BKT) to understand how the fairness of the software is impacted when reading ability is not modeled. We tested BKT model fairness by comparing two years of data from 8,549 students who were classified as either "emerging" or "non-emerging" readers (i.e., a measure of reading ability). We found that while BKT was unbiased on average in terms of equal predictive accuracy across groups, specific skills within the adaptive learning software exhibited bias related to reading level. Additionally, there were differences between the first-answer mastery rates of the emerging and non-emerging readers (M=.687 and M=.776, difference CI=[0.075, 0.095]), indicating that emerging reader status is predictive of mastery. Our findings demonstrate significant group differences in BKT models regarding reading ability, exhibiting that it is important to consider—and perhaps even model—reading as a separate skill that differentially influences students' outcomes."]} 
    more » « less
  2. ; ; ; ; ; ; (, IEEE)
    This Work-in-Progress research paper focuses on the game-based learning environment of GeoExplorer, a digital learning tool that simulates Cone Penetration Testing (CPT) – a field test that civil engineers conduct to understand soil properties. This paper aims to investigate the gendered patterns of student interaction with GeoExplorer, if any, and the associated gendered patterns of content comprehension and self-efficacy related to one's ability to pursue CPT in the field, if any. As such, we ask: (1) How, if at all, does gender affect a student's belief in their ability to apply skills gained through engagement with GeoExplorer in real world environments? and (2) How, if at all, do prior gaming experience and gender impact the GeoExplorer play experience? Four open-ended semi-structured interviews regarding students' experience with gameplay, gender identity, and learning were analyzed. Preliminary analyses used narrative and grounded theory approaches. Our early findings indicate a lack of student insight on the role of their gender identity on their gameplay experience and the presence of a gendered impact of gameplay on student learning and self-efficacy in carrying out CPT in the field. 
    more » « less
  3. ; ; ; ; ; ; ; ; ; ; et al (, Frontiers in Psychology)
    Knowledge of the relations among learners' socio-emotional characteristics and competencies as they engage in mathematics and reading is limited, especially for children with academic difficulties. This study examined the relations between anxiety, motivation, and competence in mathematics and reading, within and across domains, in an academically-diverse set of 8–13-year-old learners ( n = 146). To measure anxiety and motivation across domains, we paired existing measures of math anxiety and reading motivation with researcher-developed analogs for reading anxiety and math motivation. Participants completed standardized assessments of mathematics and reading, anxiety and motivation surveys for math and reading, and a measure of nonverbal cognitive ability. Results showed high internal consistency for all anxiety and motivation scales (Cronbach's alpha = 0.76–0.91). Pearson correlations showed that within and across domains, participants with higher competence had lower anxiety and higher motivation. Higher anxiety was also associated with lower motivation. Regression analyses showed that for both math and reading, within-domain motivation was a stronger predictor of competence than anxiety. There was a unidirectional across-domain relation: socio-emotional characteristics for reading predicted math competence, after accounting for nonverbal cognitive ability, age, gender, and within-domain anxiety and motivation. Results contribute to knowledge of the socio-emotional characteristics of children with and without learning difficulties in association with reading and math activities. Implications of a unidirectional socio-emotional link between the two domains can advance research and theory of the relations among socio-emotional characteristics and competence for academically-diverse learners. 
    more » « less
  4. (, Frontiers in psychology)
    null (Ed.)
    The goal was to identify the domain-general cognitive abilities and academic attitudes that are common and unique to reading and mathematics learning difficulties that in turn will have implications for intervention development. Across seventh and eighth grade, 315 (155 boys) adolescents (M age = 12.75 years) were administered intelligence, verbal short-term and working memory, and visuospatial memory, attention, and ability measures, along with measures of English and mathematics attitudes and mathematics anxiety. Teachers reported on students’ in-class attentive behavior. A combination of Bayesian and multi-level models revealed that intelligence and in-class attentive behavior were common predictors of reading accuracy, reading fluency, and mathematics achievement. Verbal short-term memory was more critical for reading accuracy and fluency, whereas spatial ability and mathematics self-efficacy were more critical for mathematics achievement. The combination of intelligence and in-class attentive behavior discriminated typically-achieving students from students with comorbid (D = 2.44) or mathematics (D = 1.59) learning difficulties, whereas intelligence, visuospatial attention, and verbal short-term memory discriminated typically-achieving students from students with reading disability (D = 1.08). The combination of in-class attentive behavior, verbal short-term memory, and mathematics self-efficacy discriminated students with mathematics difficulties from their peers with reading difficulties (D = 1.16). Given the consistent importance of in-class attentive behavior, we conducted post hoc follow-up analyses. The results suggested that students with poor in-class attentive behavior were disengaging from academic learning which in turn contributed to their risk of learning difficulties. 
    more » « less
  5. ; ; ; ; (, Proceedings of the 15th International Conference on Educational Data Mining)
    Research into "gaming the system" behavior in intelligent tutoring systems (ITS) has been around for almost two decades, and detection has been developed for many ITSs. Machine learning models can detect this behavior in both real-time and in historical data. However, intelligent tutoring system designs often change over time, in terms of the design of the student interface, assessment models, and data collection log schemas. Can gaming detectors still be trusted, a decade or more after they are developed? In this research, we evaluate the robustness/degradation of gaming detectors when trained on older data logs and evaluated on current data logs. We demonstrate that some machine learning models developed using past data are still able to predict gaming behavior from student data collected 16 years later, but that there is considerable variance in how well different algorithms perform over time. We demonstrate that a classic decision tree algorithm maintained its performance while more contemporary algorithms struggled to transfer to new data, even though they exhibited better performance on unseen students in both New and Old data sets by themselves. Examining the feature importance values provides some explanation for the differences in performance between models, and offers some insight into how we might safeguard against detector rot over time. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email