Senior Strategy Analyst

## About CPAL

At the Child Poverty Action Lab (CPAL), we believe every child deserves a life filled with opportunity. CPAL operates as an unofficial research and development lab for Dallas — using data to rethink public systems and equipping neighborhood-level partners to succeed, in service of one mission: cutting childhood poverty in Dallas by 50% within a single generation.

We work across five "big bets": Benefits Delivery, Maternal Health, Housing, Criminal Justice, and Public Safety. Each is grounded in evidence connecting childhood experience to adult economic outcomes.

Five design principles guide the work:

- **Start with children and families, and work backward to systems.**
- **A problem well stated is half a solution.**
- **Systems are like a string of Christmas lights.** One broken handoff takes the rest down.
- **Have a bias for action.** Perfect is the enemy of good.
- **Test, learn, and iterate.**

## About the Role

**Department:** Strategic Analytics · **Reports to:** Head of Strategic Analytics

We're looking for a Senior Strategy Analyst who gets excited about hard questions — and has the methodological toolkit to answer them responsibly. Does it sound interesting to combine multiple datasets to figure out where the most people are eligible for SNAP benefits — but aren't accessing them? To determine where a mobile clinic should set up to improve access to contraception? To understand whether students who live further from school are more likely to be chronically absent, and what schools might do about it? To map who gets evicted, how often, and what interventions might change the pattern? If so, this role was built for you.

We're not looking for someone to build machine learning pipelines or architect databases. We're looking for someone who can get to a defensible answer quickly, under real-world time pressure, and who knows when the answer approximates truth well enough to act on. You'll work with independence and judgment, while also collaborating closely with internal teams and community partners.

## What You'll Do

- **Take a fuzzy ask and turn it into an answerable question.** Find the decision behind the request; decide what's worth measuring and what isn't.
- **Work fast with imperfect data.** Interrogate an unreviewed dataset before you build on it — most of the real risk in this job lives in data nobody has checked yet.
- **Conduct applied data analysis using publicly available and administrative datasets** - ACS, CDC WONDER, Department of Education, criminal justice data, TX HHSC, Dallas ISD, Feeding America, internal program data, and others — with a working understanding of their quirks: complex sampling, vintage issues, sample weights, missingness, and outliers.
- **Go beyond the literal ask.** Say what else the requester needs to know, and what they shouldn't say publicly based on what the data can't support.
- **Translate findings into clear, responsible communications for stakeholders who are not statisticians,** helping them understand what the data supports and what it doesn't.
- **Produce clean, well-documented, reproducible workflows** so that your work can be understood, audited, and built on by others.
- **Communicate like a consultant, not an academic.** Turn analysis into a short, decision-ready brief (and a visual, where it helps) that a program lead can use in front of a reporter, a funder, or a city partner — today.
- **Work AI-natively, and verify like it's your name on the answer.** Use AI across research, coding, and QA — but the judgment, the framing, and the final synthesis have to be yours. AI drafts; you decide.
- **Build institutional knowledge.** Leave a trail — assumptions, sources, what you checked — so the next related question is cheaper to answer.

## What Success Looks Like in Your First Year

- Your analysis has changed a resource-allocation decision, a program design choice, or an external message across more than one issue area.
- Program leads come to you with the ambiguous questions, not just the clean ones.
- Your AI-assisted workflows have made you faster without making your answers less trustworthy — you can always say what you checked and why.

## Location

Dallas preferred; flexible for the right candidate. Remote-first candidates are considered, with occasional on-site time in Dallas.

## Hiring Process

This role includes a short practical challenge — roughly 60 minutes total (analysis, a required visual, a written email, and a recorded video walkthrough). It uses a real, unreviewed data file and asks you to answer the kind of question a program leader would actually send you. You must complete the challenge to be considered.

*Open to Sponsorship of Candidates*

## What We're Looking For

More than anything, we're looking for a particular kind of analytical instinct. Can you defend your choice of denominator when there were three reasonable options? Have you had to make a judgment call about how to treat missing data — and can you walk someone through your reasoning? What did you do when you opened a dataset and found that missing, 0, and "no" were all present in the same column, with no codebook in sight? We want someone who has been in those situations, made a call, and can explain it.
- 3–4+ years of applied experience using data to answer real-world questions — not just running models, but knowing which model to run and why.
- Hands-on experience with complex, publicly accessible and administrative datasets and a genuine understanding of what makes each one tricky.
- Strong grounding in statistical methods: regression, sampling and weighting, hypothesis testing, missing data, and the limits of each.
- Fluency in at least one analytical programming language — R, Python, Stata, or similar. Reproducible, well-commented code is a must.
- The ability to work fast under pressure and know when an answer is good enough to act on — without cutting corners that matter.
- A commitment to using data for good — carefully, transparently, and in service of communities.
- Strong written and verbal communication skills, including the ability to explain methodological decisions to non-technical audiences.

Bonus If You Have
- Experience with program evaluation, quasi-experimental design, or causal inference methods.
- Familiarity with geographic or spatial analysis.
- Experience working in a nonprofit, government, or policy-adjacent environment.
- Comfort with AI-assisted research workflows — prompt engineering, LLM-assisted coding or thematic analysis, output validation.

- Competitive salary commensurate with experience
- Health, dental, and vision insurance
- Retirement savings plan with employer match
- Generous paid time off and holidays
- Professional development and learning support
- Work with a small, senior team on problems that change outcomes for children and families in Dallas
- Direct access to program leaders and public-agency partners — your analysis reaches decision-makers, not a dashboard backlog
- AI-native ways of working actively supported and expected

CPAL is an equal opportunity employer. We consider all qualified applicants without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, age, disability, veteran status, genetic information, or any other characteristic protected by applicable law. If you need a reasonable accommodation at any point in the application or interview process, let us know and we will work with you to provide it.