Let’s give credit where credit is due. Dallas County and City of Dallas public officials have not won any accolades recently for decisions regarding programs that impact the quality of life for its multicultural community. The recent fiasco surrounding the distribution of the COVID-19 vaccine is the most recent example, but two other recent programs also come to mind – the Census 2020 Campaign and the lingering presence of food deserts in South Dallas. As I will argue, these programs have one important thing in common that has posed a barrier to their success: the absence of quality research to guide decisions regarding public programs that are targeted to Black and Latino residents.
In 2020, I published a book entitled “The Culture of
Research” that discusses the importance of conducting sound research in
culturally and linguistically diverse communities and the consequences to
decision making when such studies are missing or poorly conducted. Following
are some insights derived from this book that should help the reader understand
how sound research with multicultural communities could have produced improved
outcomes in the management of the local COVID-19 vaccination program, the
Census 2020 Campaign, and solving the mystery surrounding the persistence of
food deserts in South Dallas.
COVID-19 Vaccine Awareness, Registration and Distribution
The local confusion and mismanagement associated with the
COVID-19 vaccine distribution can be traced originally to the absence of
guidance and transparency at the federal level, and the related decision by the
previous administration to allow states to define their own independent
strategies with a minimal financial support from the federal government.
Nonetheless, the COVID-19 vaccine distribution dilemma in Dallas County, Texas presents a good case study on decisions that public officials in urban communities should not make. Indeed, the series of inconsistent and questionable decisions resulted in considerable public confusion and frustration with vaccine registrations, availability and distribution. Some of these missteps included the following:
- Contradictory messages from county and city public officials;
- Over-reliance on an Internet strategy to inform and register residents, many who lacked online access or own a computer, or were not comfortable with technology;
- Inconsistent support in Spanish and other languages;
- Placement of testing and vaccination sites in higher income while providing limited access in the more vulnerable areas; and
- Transportation barriers that prevented some residents to travel to testing or vaccination sites.
Figure 1:
The problem associated with vaccine distribution in
communities of color is deeply concerning because they are the most likely to
experience the more serious medical consequences from the coronavirus. Importantly, these missteps in decision
making could have been avoided with communications that were better coordinated
by public officials and engagement of experts with significant experience
engaging multicultural persons.
The decision to use an Internet vehicle for the vaccination
campaign is very likely the reason that white, higher income residents continue
to be more successful in getting vaccinated. Although some public officials
supported the idea of targeting zip codes in South Dallas that included some of
the most vulnerable Black and Latino residents, the Texas State Health
Department[2]
immediately issued a threat to withhold vaccine doses allocated to Dallas
County if the targeting was implemented.
This threat was a direct contradiction to recommendations by the National
Academy of Sciences that support vaccine community intervention programs that
are targeted to the most vulnerable communities. [3] Thus, Dallas County and City of Dallas public
officials learned the hard way that launching a public vaccination program in
linguistically and culturally diverse communities require less reliance on
technology and more reliance on outreach efforts that take the vaccines to the
residents. Ironically, while the state
threatened to withhold vaccine doses if local officials employed a zip-code
targeting approach in South Dallas, the use of an Internet strategy as the
primary form of communication accomplished the same outcome by vaccinating
higher numbers of white, higher-income residents who resided in the northern
parts of Dallas County.
South Dallas Food Deserts
Why have mainstream supermarkets avoided South Dallas food deserts that are populated by lower-income Blacks and Latinos? [4] This question inspired me to conduct a geospatial analysis using crime, demographic and supermarket expenditure data to examine the common reasons cited by supermarket executives to explain the avoidance of communities like South Dallas – such as high crime, low population density, lower household median income and insufficient food expenditures. The study revealed that crime patterns were often inflated by previous investigators and news stories, and that the annual food-at-home expenditures in several food deserts in South Dallas were adequate to sustain the annual sales of a mainstream supermarket. Flawed crime analyses, stereotypes of urban retail, and an apparent disdain for Black and Latino customers appeared to drive site selection decisions in South Dallas.
Over the past two decades, the
City has floundered millions of taxpayer dollars on ill- conceived investments that
failed to produce positive changes in the supermarket options for this
community. Worst yet, a market demand study of community residents – a
traditional practice to measure supermarket opportunities -- has never been
conducted in South Dallas. Such a study would have provided supermarket and
site selection executives the statistical evidence needed for an investment
decision. In the meantime, South Dallas residents will be forced to continue
shopping outside of their community for healthy, affordable food or visit the
less desirable dollar stores. Once City
public officials decide that the South Dallas community is deserving of a high-quality
supermarket experience, a professional, high quality market demand study is the
best approach for making this a reality.
If supermarket redlining practices continue in South Dallas despite
solid evidence of its retail potential, it might be a good idea to recruit a
supermarket chain from outside of Dallas County or Texas that reveals a greater
interest in serving Black and Latino consumers in urban communities.
The Census 2020 Campaign
In January 2020, Dallas County and the City of Dallas funded
a $1.9 million Census
2020 Campaign to provide a comprehensive strategy to boost response rates in
hard-to-count communities that were populated by lower-income Blacks and
Latinos. The team selected to conduct the campaign submitted a report
summarizing the multitude of campaign activities that they conducted from
February to August of 2020 to target these HTC communities. To monitor progress on this campaign, I
produced maps on a monthly basis that illustrated the cumulative self-response
rates by census tracts that were provided by the Census Bureau. Figure 2 below shows that the final self-response
rates reported by the Census Bureau were decidedly lower in the southern and
northwest parts of the city where HTC Blacks and Latinos resided. In fact, the table of Overall Self-Response
Rates indicates that Dallas County ended with one of the lowest self-response
rates (63.9%) compared to other large Texas counties. Consequently, the Census Bureau was required
to deploy many more field interviewers in order to minimize the potential population
under-count, an especially difficult task during the pandemic. Despite its best intentions, the Census 2020
Campaign funded by the County and City appeared to fall short of its intended
goal in hard-to-count communities and will likely lead to the loss of millions
of federal dollars for local programs. Although the pandemic posed a barrier to
response rates during this period, the burden on Dallas County was likely
similar for all other counties considered here.
Figure 2: Dallas County 2020 Self-Response Rates by Census Tract and City Service Area
Part of the challenge in completing the Census 2020 questionnaires can be traced to the reliance that the Census Bureau placed on using an online survey as their major data collection strategy. In past censuses, the Census Bureau relied primarily on a mail questionnaire, while data collection for the annual American Community Survey has utilized a mixed mode strategy that included mail questionnaires, telephone interviews, personal interviews and online surveys. Not surprisingly, Black and Latino respondents to the American Community Survey have opted for telephone and personal interviews more often than whites or Asians, while online surveys were the least chosen option.
Figure 3 below presents the percentage of Dallas County households
that completed the Census 2020 using an online survey. The map presents the cumulative Internet
self-response rates for the 2020 Census as of October 28, 2020. Of the seven City Service Areas (CSAs), the
Central, Southeast, South Central, Southwest and Northwest CSAs are populated
primarily by lower income Blacks and Latinos. It is clear that these CSAs
included census tracts (highlighted in red) with the lowest online response
rates, while the numerous other census tracts (highlighted in yellow) showed modest
online response rates. The highest
online return rates were realized for census tracts in the northeast and north
central CSAs that were populated primarily by white, higher-income residents.
Throughout 2020, public officials in Dallas County and City
of Dallas were aware of the poor performance of the Internet to encourage poor
Blacks and Latinos to complete the 2020 Census.
Why then was the Internet the main vehicle used for communications related
to COVID-19 vaccine awareness, registration and distribution? Good research and multicultural expertise would
have been beneficial to decision makers during this period.
Based on my past 45 years of experience in conducting
surveys of multicultural populations, it is my opinion that the Census 2020
Campaign sponsored by Dallas County and City of Dallas was not guided by the best
expertise regarding the strategies for successfully engaging multicultural population
segments in surveys and the biennial census. If it had been, Dallas County might have
experienced a higher ranking in Census self-response rates in comparison to the
many Texas counties that did not allocate any funding for a Census 2020
campaign.
The challenges facing public officials to ensure a
satisfactory quality of life for all community residents have become more
complex and will require careful planning using the best expertise in
understanding and engaging culturally and linguistically diverse communities. Public officials must resist the temptation
to take the path of least resistance by overlooking or dismissing the need for
solid research to guide decisions that impact the quality of life of multicultural
residents. Dallas County and City of Dallas public officials learned the hard
way that engaging culturally and linguistically diverse residents is a complex
task that requires multicultural expertise and support from community organizations.
As the population of urban areas like Dallas County continues to grow and
evolve demographically, the challenges to respond more effectively to important
community needs and events will become more challenging. Let’s hope that public officials will be
better prepared to respond.
Reference Notes
[1] Garcia,
N. and Jimenez, J. (2021, Jan. 28). White
Dallas residents outpace Blacks, Hispanics in registering for COVID vaccine. Dallas Morning News, Accessed at: https://www.dallasnews.com/news/public-health/2021/01/29/white-dallas-residents-outpace-blacks-hispanics-in-registering-for-covid-vaccine/
[2] Choi,
J. (2021, Jan. 21). Texas threatened
to reduce vaccine supply to Dallas County over plan to focus on 'vulnerable'
ZIP codes. The Hill. Accessed at: https://thehill.com/homenews/state-watch/535294-texas-threatened-to-reduce-vaccine-supply-to-dallas-county-over-plan-to
[3]
National Academy of Sciences, Engineering and Medicine (2020). Framework for
equitable allocation of COVID-19 vaccine. Washington, D.C.: National
Academy Press.
[4] Rincón,
E.T. and Tiwari, C. (2020, March 23). Demand metric for supermarket site selection:
A case study. Papers in Applied Geography, Accessed at: https://doi.org/10.1080/23754931.2020.1712555