AISP Toolkit Feb25 2025 - Flipbook - Page 51
Racial Equity in Data Analysis
Data analysis is the stage at which available data are explored in order to develop 昀椀ndings,
interpretations, and conclusions. Data analysis can be as simple as calculating descriptive statistics,
such as counts of program participants or the percentage of participants who achieved a certain
outcome. Analysis can also include measuring longitudinal trends, identifying causal relationships
between interventions and outcomes, or creating complex models that predict participant behavior.
Data analysis involves a complex series of decisions about the questions being asked, the data
and methods used to answer them, and how results will be interpreted to inform conclusions
and recommendations. Decisions about data disaggregation, in particular, require careful
consideration. On the one hand, disaggregating data can shed light on the unique experiences
of small populations and those glossed over in other analyses. However, creating a subgroup
has implications, and may shift the focus of analysis to a speci昀椀c population that is already
over-surveilled. Another key area of decision-making in analysis is whether to use quantitative
data, qualitative data, or a mixed methods approach. We 昀椀nd that too often, the opportunity
to strengthen quantitative analysis by weaving in qualitative data and other forms of contextual
data, such as the social and political history of race in the local area (see Hacking Into History and
Delaware PDG), is overlooked. Finally, decisions about how you frame and tell the story of the data
matter (see CalEnviroScreen 4.0 and Library of Missing Datasets). Though we will cover these
decisions more in the section on Racial Equity in Reporting & Dissemination, it is worth noting that
framing begins in the analysis stage, and solely relying on statistical outputs will not necessarily lead
to insights or empower people to take action. Across the board, engaging with individuals who have
lived experience and with trusted community advocates can strengthen these decisions, leading to
more meaningful and robust analyses.
CENTERING RACIAL EQUITY THROUGHOUT THE DATA LIFE CYCLE
Incorporating a racial equity lens during data analysis starts with having the right mix of people
to develop and execute a strong analytic plan. This includes subject matter experts with deep
understanding of the existing evidence and most relevant questions to ask (see Indiana MPH in the
Work in Action); the “data people” who know how to clean and wrangle the available data, assess
data quality, and apply proper statistical methods; and, when appropriate (see Foundations for
Community Involvement), community members with lived experience of the issue being studied who
can ensure that analytic approaches are aligned to community need and support the interpretation
of results (see Northside Achievement Zone and Wilder Research).
47
