In our stats classes, we need MANY examples to convey both theories behind and the computation of statistics. These examples should be memorable. Sometimes, they can make our students laugh, and sometimes they can be couched in research. They should always make our students think.
In this spirit, I've collected three small examples from the Pew Research Center's Race and Ethnicity archive (I hope to update with more examples as time permits). I don't know if any data collection firm is above reproach, but Pew Research is pretty close. They are non-partisan, they share their research methodology, and they ask hard questions about ethnicity and race. If you use these examples in class, I think that it is crucial to present them within context: They illustrate statistical concepts, and they also demonstrate outcomes of racism.
Lessons: Racism, ANOVA theory: between-group differences, post-hoc tests
While this is descriptive data, I think that it illustrates several ANOVA concepts. Participants reported whether or not they had experienced several different situations. It demonstrates how life experiences in America are influenced by race. It can also be used to get your students thinking about ANOVA.
You can think about ethnicity as the factor with four levels (Asian, Black, Hispanic, White).
If you are teaching ANOVA, this data visualization illustrates between-group differences, both in terms of statistics and life in America. You can ask your students to speculate on which item has the most between-group variance and the least between-group variance. You could also think about post-hoc results for these findings. For the question "Been unfairly stopped by the police", Asian and Hispanic respondents may not differ significantly from one another. Still, Black and White respondents vary considerably from each other and from Asians and Hispanics.
Lessons: Racism, categorical data, nominal data, free-response options, research methods, psychometrics
This article describes the options provided by the US Census Bureau for reporting race (their term), and how far those options have evolved since the first US census (1790). It also illustrates how racist attitudes can be seen in the way the government elected to describe the race of American citizens. It gives a little hope that things are changing (at least at the Census Bureau).
In 1790, the only options for race were: a) Free white males and female, 2) All other free persons, and 3) slaves.
Until 1960, the person recording your census data would pick your race.
Americans couldn't describe their backgrounds by selecting multiple races until 2000
Regardless, the Census Bureau is trying harder in 2020, and the new way of recording race provides examples of categorical variables, qualitative data, free response.
| https://www.pewresearch.org/fact-tank/2020/02/25/the-changing-categories-the-u-s-has-used-to-measure-race/ |
Furthermore, if you want to discuss politicized data collection, you could also dig into Hansi Lo Wang's reporting for NPR on this topic. It is excellent.
Lessons: Racism, Chi-Square theory: observed data, expected data
Pew Research created a graph to illustrate one facet of systematic racism in health care: Black people are dying of COVID-19 in numbers that are disproportionate and alarming. (Aside: I also teach my students about this, framing it in data-driven ML, using this article from Nature)
I think this graph does an excellent job of conveying a concept at the core of chi-square: The distance between expected data (The black portion of the population) versus the observed data (Black share of COVID-19 deaths) is illustrated by state.
 |
| https://www.pewresearch.org/fact-tank/2020/06/04/black-americans-face-higher-covid-19-risks-are-more-hesitant-to-trust-medical-scientists-get-vaccinated/ft_20-06-01_covidracehealth_3/ |
Comments
Post a Comment