,Loren Collingwood began his political consulting work in 2003, working as a field associate and analyst for the Washington, DC, based polling firm, Greenberg Quinlan Rosner. There, he learned about survey sampling, questionnaire design, survey fielding, and qualitative focus group and in-depth interview methods. He received his Ph.D. in political science from the University of Washington in 2012, with a particular focus on quantitative methods. In particular, he helped direct the university's Washington Poll, managing the day to day operations and call center. Throughout his time in graduate school he continued to provide advanced statistical analysis to political consulting firms and candidates. More recently, Collingwood has moved into redistricting and racially polarized voting consulting where he helps assess whether voting systems are racially representative. Today, Collingwood is an associate professor of political science at the University of New Mexico. In addition to his academic publishing, Collingwood has handled more than 100 consulting jobs ranging from complex sample design, survey weighting, regression analysis, factor analysis, cluster analysis, maximum difference experiments, field experiments, and ecological inference. Consultants and political organizations often turn to him to handle the most difficult part of a data collection or analytical effort. He is used to working with consultants in high pressure fast-paced campaign environments.

Some of the services Collingwood offers:
1. Survey sampling and sample design. In 2017 and 2018 I developed a complex cluster design to more appropriately generate margin of error estimates for the National Democratic Institute's surveys in Iraq.
2. Questionnaire design. I have worked in polling for 15 plus years and can help develop and test survey scales for latent dimensions, and survey experiments to test messaging, etc.
3. Survey weighting. I can weight survey or other data to known population benchmarks and produce frequency questionnaires and crosstabs as needed.
4. Propensity Score Sample Matching. Using modern causal inference techniques, I weight online panel data to known sample benchmarks to a priori generate more representative data from non-probability samples.
4. Regression Analysis. I regularly conduct regression analysis for clients based on observational survey data. I have conducted regressions for candidates and parties across the United States, South America, Asia, and Europe.
5. Maximum Difference (Max Diff). Max Diff is a choice modeling technique used to hone message testing. I have developed tailored software to handle max diff data and report results in understandable ways that works with clients' existing data-processing systems.
6. Factor and Cluster Analysis. Many clients seek a more nuanced understanding of sensitive issues. Often we turn to scaling to arrive at statistically appropriate measures.
7. Ecological Inference (EI). Jurisdictions interested in assessing voter turnout by race often turn to aggregate data to infer individual-level voting behavior. I am the lead author on the R statistical package, eiCompare. This program provides a one-stop-shop for conducting EI -- from geocoding, to gathering Census data, to producing and checking results.