Seeing Diversity with Data
Diversity is often recognized institutionally only when it can be measured. Traditional diversity is therefore based on fairly surface-level attributes, like height, weight, and skin color. But humans differ from each other in infinite ways, and until we can document this diversity, it will be hard to make sure everyone is being recognized, let alone served.
Seeing Diversity with Data is about exploring real-world data, exposing patterns through clever analyses, and using these patterns to construct data-driven stories to bring awareness to the diversity in human experience. Sometimes the focus is on enriching existing stories, sometimes it is about discovering new narratives to expand.
Examples of the first include how diversity increases productivity in science, or how being African American may cause worse pregnancy outcomes, regardless of education or economic status. Examples of the second focus on identifying new variables that might cause disparities in outcome, but that have not yet been recognized (or recognized only by specific communities). My lab is open to both approaches. My own work has focused on the latter, exploring diversity revealed by features in time, like circadian, menstrual, and seasonal rhythms.
An example: by capturing differences in daily rhythms, we illuminate possible biases in education that appear to hurt those with biologically later circadian rhythms.
Chronotype refers to what time of day a person is active - morning larks get up with the sun, while some night owls might just be going to sleep then. Chronotype is largely genetic, arising from normal mutations in "clock genes" that guide an individual's circadian rhythms. It also changes across life in response to normal development. For example, in puberty, most peoples' clocks get a few hours later than the rest of their lives, which is a large reason for the stereotype that teenagers hate to get up early. In reality, we've only just begun to understand chronotype diversity, how it changes across each person's life differently, and what anyone can do to affect their own chronotype.
But there is good reason to pay attention and learn more about chronotype. School times and business hour put an environmental constraint on when people are expected to be up and performing, even though this expectation is closer to some chronotypes than others. As a result, many people are made to perform when our bodies would prefer we slept, and don't get to perform when our minds are sharpest. Being made to get up at the wrong time is one form of "social jet lag." We found that the more social jet lag students experience, the worse they do in classes. Owls have the greatest social jet lag on average, and this correlates to disadvantages in all classes - even later ones.
This Owl Disadvantage" is likely compounded across a life time. Trouble in school means worse grades, which means a harder time getting into college, a harder time finding a job, and a harder time staying healthy. Persistent jet lag, be it physical or social, is a health risk for everything from cancer to infertility, and depression to dementia. As a society, we take no action to mitigate the harm owls incur by living under our more lark-friendly work schedule. We can only talk about owl disadvantage now, because we can finally measure it. If we want to do something about it, we will need to learn more; will remote learning let people individualize their schedules, and will this reduce owl disadvantage? What could brick-and-mortar schools do? What privacy concerns does it raise if schools start monitoring when students sleep? Clearly, much discussion and experimentation is ahead of us, but at least we can now see that this exists to be addressed.
This example shows us how large data over diverse populations reveals unrecognized diversity. That diversity cannot be accounted for and accommodated, let alone celebrated, when it remains invisible. This is the impetus for seeing Diversity with Data.