Key Attributes of a Modern Statistical Computing Tool

Abstract

In the 1990s, statisticians began thinking in a principled way about how computation could better support the learning and doing of statistics. Since then, the pace of software development has accelerated, advancements in computing and data science have moved the goalposts, and it is time to reassess. Software continues to be developed to help do and learn statistics, but there is little critical evaluation of the resulting tools, and no accepted framework with which to critique them. This paper presents a set of attributes necessary for a modern statistical computing tool. The framework was designed to be broadly applicable to both novice and expert users, with a particular focus on making more supportive statistical computing environments.A modern statistical computing tool should be accessible, provide easy entry, privilege data as a first-order object, support exploratory and confirmatory analysis, allow for flexible plot creation, support randomization, be interactive, include inherent documentation, support narrative, publishing, and reproducibility, and be flexible to extensions. Ideally, all these attributes could be incorporated into one tool, supporting users at all levels, but a more reasonable goal is for tools designed for novices and professionals to 'reach across the gap,' taking inspiration from each others' strengths.

Making Data Science Count In and For Education

“…This delineation contributes to an issue: Learners eventually require functionality that the tool they have used does not provide, whereas the tools used by professionals remain challenging to begin to use. McNamara (2019) recommends that developers of statistical tools recognize that individuals analyzing data are likely to use different tools over time, and so it is necessary to "build (either technically or pedagogically) an onramp toward the next tool" (p. 382). In this way, those designing (and studying the impacts of) tools for learners can be informed by the high-performing software used by professional statisticians and data scientists.…”

“…This will involve coordinating research at the K-12 and undergraduate levels and between mathematics education, statistics education, science education, and computer science education. This will also involve creating (and researching) opportunities for data science learners to use statistical and data science-related tools that are designed not only for learning, but also for professional data science practice (McNamara, 2019;Rosenberg et al, 2020), even if, at first, learners must use tools designed for professionals in a more constrained way. Learners must also be socialized into the conventions of data scientists that go beyond tools: conventions such as dispositions toward open science and privacy and ethical issues.…”

Making Data Science Count In and For Education

“…Learners eventually require functionality that the tool they have used does not provide, whereas the tools used by professionals remain challenging to begin to use. McNamara (2019) recommends that developers of statistical tools recognize that individuals analyzing data are likely to use different tools over time. So it is necessary to "build (either technically or pedagogically) an onramp toward the next tool" (p. 382).…”

“…This will involve coordinating research at the K-12 and undergraduate levels and between mathematics education, statistics education, science education, and computer science education. This will also include creating (and researching) opportunities for data science learners to use statistical and data science-related tools that are designed not only for learning but also for professional data science practice (McNamara, 2019;Rosenberg et al, 2020), even if, at first, learners must use tools designed for professionals in a more constrained way. Learners must also be socialized into the conventions of data scientists that go beyond tools: conventions such as dispositions toward open science and privacy and ethical issues.…”

More confidently uncertain? Teaching learners to apply Bayesian methods to make sense of scientific phenomena

“…A key expected contribution has to do with Bayesian methods being more accessible to learners. Another contribution concerns how existing tools may need to be changed to support a Bayesian approach to modeling and inference may be designed in line with what is known about the development of statistical software that is useable by learners (McNamara, 2019). As a technique that depends on computation, this work has some implications for research on computational thinking, especially in science education (Sengupta et al, 2013).…”