Alexander Baxter

About Me

My name is Alexander Baxter, and I am a PhD-trained Data Analyst working at the US Census Bureau (since Aug 2023). I have over 7 years of experience working as an applied researcher (in both academic and government settings). I have a strong track record of turning messy datasets into actionable insights for stakeholders.

Please see below for my sample work portfolio.

Contact

Email: alexander.baxter94 [at] gmail [dot] com

Other Sites: LinkedIn · Google Scholar Page . Full List of Publications & Awards

Portfolio

Data Cleaning: Exploring Data Entry Errors (DOJ)

Data Dashboard: Interactive Graphs in R Shiny App

Data Visualization: Predictors of Educational Achievement

Meta-Analysis: Initial Impressions Predict Later Romantic Pursuit

Data Mining: Genome-Wide Analysis of Personality and Health

Machine Learning, Data Reduction, & Model Fitting

Power Analysis via Simulation and Bootstrapping

Item Analysis and Survey Development

Oral & Poster Presentations: Compatibility and Relationships

Data Cleaning: Exploring Data Entry Errors (DOJ)

In this technical report, I explored potential data entry errors as part of my work at the California DOJ Research Center. I worked with a team of four others to analyze a large dataset (3 million+ rows) for an annual report on racial profiling by police officers. Stakeholders wanted to analyze whether there were racial disparities in how long police stops lasted. However, an initial analysis showed that a small subset of stops had an unusually long stop duration, which suggested potential errors in data entry. My assignment was to investigate these cases. I identified four types of errors that could potentially explain these patterns in the data and proposed several solutions to reduce these errors in future waves of data collection.

PDF

Data Dashboard: Interactive Graphs in R Shiny App

This is an interactive data dashboard that I built (in R Shiny) for one of my dissertation research projects. The study looked for associations between 11 different social factors and vasopressin binding (a hormone associated with various social behaviors) across 14 different brain regions. The primary report focused on broad trends across the measures we collected. I created this app for readers who wanted additional information about the results of the study so that they could explore the data themselves and visualize specific analyses of interest. The app also allows users to visualize the data's variability across subjects and brain regions.

Web App

Data Visualization: Predictors of Educational Achievement

This is a report that I wrote to demonstrate my ability to perform descriptive analyses and create summary data visualizations using R Markdown. The report explores factors associated with educational achievement in a public data set from the National Center for Education Statistics.

PDF · R Code (R Markdown File)

Meta-Analysis: Initial Impressions Predict Later Romantic Pursuit

In this first-authored study, I used three data sets (550+ participants) to assess whether first impressions during speed-dating predicted later romantic pursuit. I meta-analyzed the results across studies, and found that impressions based on compatibility and popularity were the strongest predictors of later dating. This project demonstrates my ability to communicate results to both technical audiences (in a journal article) and lay audiences (in press interviews).

PDF · PNAS Article · Data, R Code, & Preregistration . Newsweek Coverage · Altmetric: Top 5% Attention Score

Data Mining: Genome-Wide Analysis of Personality and Health

In this first-authored study, I explored whether infant personality measures were associated with biological markers of health and immunity (methylation in immune cells) across over 1 million sites in the genome. This project demonstrates my ability to identify key trends from large quantities of data and my ability to work with large data sets in R.

PDF . Journal Article

Machine Learning, Data Reduction, & Model Fitting

This is a statistics lab that I wrote to teach students how to perform machine learning analyses via regularization (ridge and lasso regression). The lab uses a data set from scientific study on the social lives of people that identify as asexual, and demonstrates how machine learning can be used to identify the best predictors of an outcome (sexual identity) across many psychological and social predictors (10+). The lab also teaches students about the danger of overfitting a regression model with too many predictors, and demonstrates that regression analyses can be evaluated by testing the model in a hold out data set that was not used for training the original model.

PDF · R Code (Markdown File) · Associated Data

Power Analysis via Simulation and Bootstrapping

This is a statistics lab that I wrote to teach students how to use bootstrapping and data simulation to plan studies (power analyses) and estimate population parameters (prevalance rates), even when limited information is available and must be compiled from different sources. The lab also teaches students how to apply these principles to estimate confidence intervals for data analysis.

PDF · R Code (Markdown File)

Item Analysis and Survey Development

This is a group project that I did as part of a research methods and measurements course. The assignment was to create a 10-question survey to measure a psychological construct. We then had to administer the survey and analyze the resulting data to assess whether our measures had psychometric reliability and subscale reliability (determined via correlational analysis and factor analysis).

PDF

Oral & Poster Presentations: Compatibility and Relationships

In this 7 minute oral presentation, I present an overview and preliminary results on a psychology research project. The study investigated how initial compatibility between potential partners contributes to relationship formation, and found converging lines of evidence in both humans and monkeys.

This presentation was given at The Love Consortium 2021 Virtual Research Conference.

Youtube Video

This is a poster that I designed and presented at an academic research conference (on the same project described above). This poster demonstrates my ability to create aesthetic and engaging presentations that effectively communicate research results.

This poster was presented at the Society for Personality and Social Psychology 2022 Research Confence.

PDF of Poster

Page updated

Google Sites

Report abuse