Greetings, my fellow statisticians. This summer has proven to be quite remarkable. The spring and early summer brought hopes of a new normalcy in a quasi-post-COVID world. But in the midst of our record heat wave and wildfires galore, a resurgence appears to be upon us thanks to the delta variant.
We will need to tap our reservoirs of resilience yet again to weather the months ahead. But we have weathered this storm before, so although the path is likely treacherous, we know what we need to do to navigate safely. Please continue to wear your masks per CDC guidance, and if you have not yet been vaccinated, please consider doing so (unless your doctor has advised otherwise). Your family needs you, your friends need you, and we—your ASA colleagues—need you.
For all our statistician heroes at the FDA or CDC or in pharma, public health, and the medical sciences and anyone else helping to end this pandemic, we thank you for helping to save lives and stem the suffering. Hang in there.
I would also like to thank all who have supported me as I navigate the senate confirmation process for the position of US Census Bureau director. In last month’s Presidents Corner, I wrote about the value of vulnerability. I can proudly say I embraced that value heading into my confirmation hearing. Regardless of what the future holds, I will always cherish the expressions of support I received from the statistical community, especially fellow ASA members. I only hope I have the opportunity to pay it forward. But for now, let me tell you what I am really thinking about: the value of empowering people in our statistical work.
We have all heard the saying, “statistics tell a story,” and I genuinely believe it. But I know the storyteller is a key player in any statistical inference venture. Context can be crucial. We statisticians tout the objective power of mathematical calculations, statistical theory, and inferential statistics. And indeed, we present seemingly objective empirical findings to researchers, who then interpret and synthesize with aplomb and add to their knowledge base.
Not surprisingly, different storytellers can develop different stories from the same set of analytic results. What appears to be a tangled mess to some may be an intricately woven tapestry to others. Indeed, different perspectives can lead to vastly different interpretations when viewing exactly the same statistics. All interpretations can be informative and contribute to a dialog that leads to insight. And that is why I value and seek alternative perspectives, be they in statistical studies, project management, leadership decision-making, or even career development.
Here is a story of how community engagement—specifically, perspectives offered by community members—added great value to a policy research project in which I was involved tangentially. The setting was Austin, Texas, and the topic was police profiling. Community concerns over police profiling of Black and Latino youth in east Austin had come to a head and tensions were rising alarmingly. Colleagues of mine at the Urban Institute secured funding to conduct a research project undergirded by a community-based participatory research design. Under this model, the police department leadership and some officers would work closely with community leaders from the affected neighborhood (and Urban Institute researchers) to define the problem to be addressed and help guide the research to better understand community concerns and police perspective, which in turn would hopefully lead to policy solutions.
Naturally, a big concern was that reliance on solely anecdotal information might not reflect the broader community sentiment and unnecessarily influence the direction and tone of the conversations between the community and police representatives. So, the research team decided to conduct a sample survey of households in the affected neighborhood and use the results to generate insights and policy recommendations for building a better, more productive policy-community partnership.
A questionnaire was developed and tested, a sample was drawn, and interviewers were trained and deployed over a long, hot summer. The completed interviews were gathered, processed, and tabulated.
Because this was a community-engaged study, rather than having the Urban Institute researchers synthesize the results, develop findings, and make policy recommendations, a different approach was taken. The Urban Institute researchers crafted sets of data visualizations from the survey data composed of easy-to-read charts and graphs. Then, a “data walk” was conducted at a local community center in the target neighborhood.
At the data walk, both community members and police staff—from the police chief to the officers who patrolled the neighborhood—participated in an evening event focusing on the survey results and their interpretation. Everyone gathered, reviewed the empirical data, and provided their own interpretations, with a researcher guiding a discussion at stations featuring a specific set of results on a single topic (e.g., satisfaction with some aspect of policing or perceptions of safety in the neighborhood).
Motivated by the survey findings, attendees also offered policy recommendations to ameliorate the tensions between police and community. The data walk was a little tense at first because of the unspoken power dynamic between the police and community members, although there was a base of trust that had been nurtured over the course of the project that helped ease tensions as the evening wore on.
During the data walk, both the researchers and police noted a seemingly contradictory pair of results that had the potential to call into question the integrity of the survey. The findings showed the community overwhelmingly believed the police were profiling the neighborhood youth and were highly dissatisfied. But other findings showed a high level of support for the police doing their job. Both researchers and police asked how that could be and concluded something was awry with either the survey questions or the implementation of the survey design.
However, the community participants knew better. Their unique perspective facilitated an understanding of results that led to knowledge gain. They pointed out that the results were not contradictory; both were valid sentiments of the community. They explained it this way: If they are home at night and someone is trying to break into their house, they call the police who then rush to them and avert the crime and often detain a suspect. They very much appreciate and value that response. But they do not appreciate police pulling over their neighbors, friends, and family members who are returning home from school or work after a hard day. They highly value the role of policing in public safety but are adamantly against profiling of community members because of their skin color.
This explanation made a lot of sense. And without a community engagement model, researchers might still be scratching their heads over what they thought represented an illogical data pattern.
Once this was articulated by community members, the data walk continued with renewed enthusiasm and innovative policy recommendations were jointly developed by the community members and police in attendance. Everyone learned from this experience, especially me.
Although I was not formally involved in this project, participating in the data walk opened my eyes to the value of alternative perspectives. You do not have to be a PhD-level substantive researcher or statistician to be a critical thinker. Your life experiences matter and can contribute to insights. Community members who have a stake in their communities can offer cogent insights if we only give them a chance to participate, to have a voice.
I have now been a big believer in the value of community engagement for years. I believe in nurturing environments in which different perspectives can be offered. Without that, a researcher can run the risk of missing a key finding and culturally relevant policy solutions.
Please remember that we—as statisticians—are also critical thinkers and can offer an alternative perspective that can benefit a research project. We can also encourage researchers to consider community engagement models to help enrich the understanding of results and the potency and relevance of their recommendations. When it comes to research, giving “power to the people” can prove invaluable in the interpretation of results.
Leave a Reply