580 California St., Suite 400
San Francisco, CA, 94104
Key research themes
1. How can induced student errors enhance mathematical understanding and conceptual thinking?
This research area investigates error-eliciting problems designed to intentionally provoke specific student errors in mathematics, particularly misconceptions and procedural misapplications. It focuses on how planned and expected errors serve as educational tools to foster deeper reasoning, justification, and conceptual understanding rather than to be avoided. Understanding these errors enables educators to harness them as springboards for inquiry and critical thinking in mathematics learning.
Key finding: Introduces and categorizes error-eliciting problems that deliberately induce common mathematical misconceptions such as 'multiplication always makes bigger' and 'division always makes smaller.' Demonstrates through empirical... Read more
2. What cognitive-pragmatic mechanisms underlie error production in second and third language learners?
This theme explores multilingual language errors from a cognitive-pragmatic perspective, integrating theories such as Relevance Theory, Mental Models Theory, and the Graded Salience Hypothesis. It examines how learners’ interlanguage errors are influenced by their motivation to produce relevant utterances, the salience of meanings, and their mental representations when processing L2 and L3. The focus is on understanding errors beyond traditional transfer and interference, emphasizing learners’ pragmatic strategies and context-dependent processing in producing and comprehending foreign languages.
Key finding: Demonstrates that many L2 and L3 errors unexplainable by transfer or overgeneralization can be accounted for by learners’ attempts to maximize relevance under limited or distorted contextual understanding, as predicted by... Read more
Key finding: Analyzes negative lexical transfer and interference arising from cognates and false friends, illustrating how lexical similarities across L1, L2, and L3 cause systematic errors at multiple linguistic levels including... Read more
Key finding: Reviews diverse sources and types of learner errors in EFL, including those due to mother tongue interference and overgeneralization, emphasizing the significance of context and learner cognitive processes in understanding... Read more
3. How can statistical and computational methods quantify and evaluate errors in environmental data and computational applications?
This theme addresses the quantitative measurement, correction, and propagation of errors in environmental data analysis and computational processes. It includes the development and application of automated tools for raster data error assessment, novel algorithms for evaluating floating-point computational accuracy, and statistical methods for uncertainty quantification in diagnostics. The focus is on improving precision, robustness, and interpretability of error metrics to support reliable environmental modeling and numerical simulations.
Key finding: Presents the design and implementation of an automated toolbox computing key error metrics (RMSE, NRMSE, MBE, NMBE, MAE, NMAE) for raster datasets within GIS platforms, facilitating cell-by-cell error quantification against... Read more
Key finding: Introduces encapsulated error arithmetic encapsulating both computed results and approximated numerical errors within a dedicated data type, enabling direct measurement of significant digits and numerical accuracy with lower... Read more
Key finding: Develops generalized probabilistic formulas quantifying diagnostic trustworthiness and error probabilities considering multiple states of diagnostic objects, tools, and human operators. This framework accounts for diagnostic... Read more
4. What are common linguistic error patterns in L2 writing, and how do they relate to learners’ native language and proficiency?
This theme investigates learners’ written language errors, particularly in spelling, syntax, and morphosyntax, analyzing the nature, frequency, and sources of these errors across diverse learner populations. It also evaluates automated writing evaluation tools’ accuracy and the influence of L1 interference and overgeneralization on error patterns, aiming to inform language teaching and error correction practices.
Key finding: Identifies and quantifies dominant spelling errors in written English by third-semester university students, revealing prevalent patterns that suggest specific orthographic challenges faced by learners, providing empirical... Read more
Key finding: Finds that mother tongue interference is the principal cause of syntactic errors among Sudanese secondary students writing in English, with learners exhibiting inadequate semantic and syntactic understanding, highlighting the... Read more
Key finding: Analyzes 120 written compositions of Algerian EFL learners, categorizing morphosyntactic errors including word order, subject-verb agreement, verb forms, and morpheme addition/omission, attributing these errors to... Read more
Key finding: Compares grammatical errors between young learners and adults at elementary proficiency, showing adults’ errors predominantly stem from mother tongue interference while children’s errors mostly result from overgeneralization,... Read more
Key finding: Evaluates the accuracy of Grammarly’s automated error detection on L2 English learners’ texts, finding an overall accuracy of approximately 79% with higher reliability in form-related errors (about 92%). However,... Read more
5. What statistical and algorithmic methods can improve error detection, correction, and interpretation in data and software systems?
This area focuses on methodologies for detecting, projecting, correcting, and interpreting errors in software architectures, program analysis, and self-reported data. It includes formal approaches such as error projections in program models, correction of heaping errors in variables via validation data, and metrics for error propagation probabilities in software components. The goal is enhancing error management through analytical, algorithmic, and statistical techniques applicable to large-scale software and data systems.
Key finding: Defines error projections as sets of program nodes associated with reachable error paths in abstract program models and introduces annotated error projections that include counterexamples and abstract stores, enabling precise... Read more
Key finding: Proposes a novel statistical correction method for heaping errors in self-reported data using external validation datasets, effectively reducing bias in the estimation of means, variances, and regression parameters. The... Read more
Key finding: Develops analytical and empirical models to estimate error propagation probabilities between components and connectors within software architecture, employing stochastic information flow and entropy functions. The work... Read more