arXiv subjects

Alex Reinhart

Publications and source records attributed to Alex Reinhart.

11 recordsLinked to original sources

arXiv2026

Analyzing Students' Statistics Writing Before and After the Emergence of Large Language Models

The ability to communicate statistical results to domain experts and stakeholders is an important goal of the undergraduate statistics and data science curriculum. However, as large language models (LLMs) have become more accessible, a major concern is that students are offloading important cognitive tasks to generative AI. Using a corpus of over 1,600 undergraduate students' data analysis reports from 2021 to 2025, we show how students' writing style and verb usage have become more similar to that of LLMs. This shift is most pronounced in the first and fifth quintiles of students' reports, which roughly map onto the introduction and conclusion sections, respectively. At the same time, we demonstrate that students' writing style has become more similar to that of statistics experts with the addition of LLMs. We end by discussing the implications of our findings for statistics and data science educators. In particular, we propose alternative modes of assessment that still emphasize statistical thinking, such as targeted writing assignments for structuring a report introduction.

stat.AP↗

arXiv2024

Do LLMs write like humans? Variation in grammatical and rhetorical styles

Large language models (LLMs) are capable of writing grammatical text that follows instructions, answers questions, and solves problems. As they have advanced, it has become difficult to distinguish their output from human-written text. While past research has found some differences in surface features such as word choice and punctuation, and developed classifiers to detect LLM output, none has studied the rhetorical styles of LLMs. Using several variants of Llama 3 and GPT-4o, we construct two parallel corpora of human- and LLM-written texts from common prompts. Using Douglas Biber's set of lexical, grammatical, and rhetorical features, we identify systematic differences between LLMs and humans and between different LLMs. These differences persist when moving from smaller models to larger ones, and are larger for instruction-tuned models than base models. This observation of differences demonstrates that despite their advanced abilities, LLMs struggle to match human stylistic variation. Attention to more advanced linguistic features can hence detect patterns in their behavior not previously recognized.

cs.CL↗

arXiv2022

Think-aloud interviews: A tool for exploring student statistical reasoning

Think-aloud interviews have been a valuable but underused tool in statistics education research. Think-alouds, in which students narrate their reasoning in real time while solving problems, differ in important ways from other types of cognitive interviews and related education research methods. Beyond the uses already found in the statistics literature -- mostly validating the wording of statistical concept inventory questions and studying student misconceptions -- we suggest other possible use cases for think-alouds and summarize best-practice guidelines for designing think-aloud interview studies. Using examples from our own experiences studying the local student body for our introductory statistics courses, we illustrate how research goals should inform study-design decisions and what kinds of insights think-alouds can provide. We hope that our overview of think-alouds encourages more statistics educators and researchers to begin using this method.

stat.OT↗

arXiv2021

Big data, big problems: Responding to "Are we there yet?"

Bradley et al. (arXiv:2106.05818v2), as part of an analysis of the performance of large-but-biased surveys during the COVID-19 pandemic, argue that the data defect correlation provides a useful tool to quantify the effects of sampling bias on survey results. We examine their analyses of results from the COVID-19 Trends and Impact Survey (CTIS) and show that, despite their claims, CTIS in fact performs well for its intended goals. Our examination reveals several limitations in the data defect correlation framework, including that it is only applicable for a single goal (population point estimation) and that it does not admit the possibility of measurement error. Through examples, we show that these limitations seriously affect the applicability of the framework for analyzing CTIS results. Through our own alternative analyses, we arrive at different conclusions, and we argue for a more expansive view of survey quality that accounts for the intended uses of the data and all sources of error, in line with the Total Survey Error framework that have been widely studied and implemented by survey methodologists.

stat.AP↗

arXiv2020

Expanding the scope of statistical computing: Training statisticians to be software engineers

Traditionally, statistical computing courses have taught the syntax of a particular programming language or specific statistical computation methods. Since the publication of Nolan and Temple Lang (2010), we have seen a greater emphasis on data wrangling, reproducible research, and visualization. This shift better prepares students for careers working with complex datasets and producing analyses for multiple audiences. But, we argue, statisticians are now often called upon to develop statistical software, not just analyses, such as R packages implementing new analysis methods or machine learning systems integrated into commercial products. This demands different skills. We describe a graduate course that we developed to meet this need by focusing on four themes: programming practices; software design; important algorithms and data structures; and essential tools and methods. Through code review and revision, and a semester-long software project, students practice all the skills of software engineering. The course allows students to expand their understanding of computing as applied to statistical problems while building expertise in the kind of software development that is increasingly the province of the working statistician. We see this as a model for the future evolution of the computing curriculum in statistics and data science.

stat.CO↗

arXiv2019

Self-exciting point processes with spatial covariates: modeling the dynamics of crime

Crime has both varying patterns in space, related to features of the environment, economy, and policing, and patterns in time arising from criminal behavior, such as retaliation. Serious crimes may also be presaged by minor crimes of disorder. We demonstrate that these spatial and temporal patterns are generally confounded, requiring analyses to take both into account, and propose a spatio-temporal self-exciting point process model which incorporates spatial features, near-repeat and retaliation effects, and triggering. We develop inference methods and diagnostic tools, such as residual maps, for this model, and through extensive simulation and crime data obtained from Pittsburgh, Pennsylvania, demonstrate its properties and usefulness.

stat.AP↗

arXiv2018

A Review of Self-Exciting Spatio-Temporal Point Processes and Their Applications

Self-exciting spatio-temporal point process models predict the rate of events as a function of space, time, and the previous history of events. These models naturally capture triggering and clustering behavior, and have been widely used in fields where spatio-temporal clustering of events is observed, such as earthquake modeling, infectious disease, and crime. In the past several decades, advances have been made in estimation, inference, simulation, and diagnostic tools for self-exciting point process models. In this review, I describe the basic theory, survey related estimation and inference techniques from each field, highlight several key applications, and suggest directions for future research.

stat.ME↗

arXiv2016

Sequential nonparametric tests for a change in distribution: an application to detecting radiological anomalies

We propose a sequential nonparametric test for detecting a change in distribution, based on windowed Kolmogorov--Smirnov statistics. The approach is simple, robust, highly computationally efficient, easy to calibrate, and requires no parametric assumptions about the underlying null and alternative distributions. We show that both the false-alarm rate and the power of our procedure are amenable to rigorous analysis, and that the method outperforms existing sequential testing procedures in practice. We then apply the method to the problem of detecting radiological anomalies, using data collected from measurements of the background gamma-radiation spectrum on a large university campus. In this context, the proposed method leads to substantial improvements in time-to-detection for the kind of radiological anomalies of interest in law-enforcement and border-security applications.

stat.ME↗

arXiv2016

Multiscale spatial density smoothing: an application to large-scale radiological survey and anomaly detection

We consider the problem of estimating a spatially varying density function, motivated by problems that arise in large-scale radiological survey and anomaly detection. In this context, the density functions to be estimated are the background gamma-ray energy spectra at sites spread across a large geographical area, such as nuclear production and waste-storage sites, military bases, medical facilities, university campuses, or the downtown of a city. Several challenges combine to make this a difficult problem. First, the spectral density at any given spatial location may have both smooth and non-smooth features. Second, the spatial correlation in these density functions is neither stationary nor locally isotropic. Finally, at some spatial locations, there is very little data. We present a method called multiscale spatial density smoothing that successfully addresses these challenges. The method is based on recursive dyadic partition of the sample space, and therefore shares much in common with other multiscale methods, such as wavelets and Pólya-tree priors. We describe an efficient algorithm for finding a maximum a posteriori (MAP) estimate that leverages recent advances in convex optimization for non-smooth functions. We apply multiscale spatial density smoothing to real data collected on the background gamma-ray spectra at locations across a large university campus. The method exhibits state-of-the-art performance for spatial smoothing in density estimation, and it leads to substantial improvements in power when used in conjunction with existing methods for detecting the kinds of radiological anomalies that may have important consequences for public health and safety.

stat.ME↗

arXiv2015

Detecting changes in maps of gamma spectra with Kolmogorov-Smirnov tests

Various security, regulatory, and consequence management agencies are interested in continuously monitoring wide areas for unexpected changes in radioactivity. Existing detection systems are designed to search for radioactive sources but are not suited to repeat mapping and change detection. Using a set of daily spectral observations collected at the Pickle Research Campus, we improved on the prior Spectral Comparison Ratio Anomaly Mapping (SCRAM) algorithm and developed a new method based on two-sample Kolmogorov-Smirnov tests to detect sudden spectral changes. We also designed simulations and visualizations of statistical power to compare methods and guide deployment scenarios.

physics.data-an↗

arXiv2014

Spatially-Aware Temporal Anomaly Mapping of Gamma Spectra

For security, environmental, and regulatory purposes it is useful to continuously monitor wide areas for unexpected changes in radioactivity. We report on a temporal anomaly detection algorithm which uses mobile detectors to build a spatial map of background spectra, allowing sensitive detection of any anomalies through many days or months of monitoring. We adapt previously-developed anomaly detection methods, which compare spectral shape rather than count rate, to function with limited background data, allowing sensitive detection of small changes in spectral shape from day to day. To demonstrate this technique we collected daily observations over the period of six weeks on a 0.33 square mile research campus and performed source injection simulations.

physics.data-an↗