A Modern Look at Freedman's Box Model

This paper revisits the box model, a metaphor developed by David Freedman to explain sampling distributions and statistical inference to introductory statistics students. The basic idea is to represent all random phenomena in terms of drawing tickets at random from a box. In this way, random samplin...

Full description

Saved in:
Bibliographic Details
Published in:Technology innovations in statistics education 2019, Vol.12 (1)
Main Authors: Sun, Dennis L, Alfredo, Joseph
Format: Article
Language:English
Subjects:
Educational innovations
Probability
Statistics
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper revisits the box model, a metaphor developed by David Freedman to explain sampling distributions and statistical inference to introductory statistics students. The basic idea is to represent all random phenomena in terms of drawing tickets at random from a box. In this way, random sampling from a population can be described in the same way as everyday phenomena, like coin tossing and card dealing. For Freedman, box models were merely a thought experiment; calculations were still done using normal approximations. In this paper, we propose a more modern view that treats the box model as a practical simulation framework for conducting inference. We show how concepts in introductory statistics and probability classes can be motivated by simulating from a box model. To facilitate this simulation-based approach to teaching box models, we developed an online, open-source "box model simulator".
ISSN:1933-4214
1933-4214
DOI:10.5070/T5121044395