---

Education

---

Publications and Conference Proceedings

1. Lang, D. N., Chen, Y., Paepcke, A., & Stevens, M. (2020). Course reviews reveal gender differences and other scientific insight about the students who submit them. EdArxiv.
2. Lang, D., Chen, G., Mirzaei, K., & Paepcke, A. (2020, March). Is faster better? a study of video playback speed. In Proceedings of the Tenth International Conference on Learning Analytics & Knowledge (pp. 260-269).

3. Lang, D. Stenhaug, B. Kizilcec, R. 2019. Keystrokes, Edit Distance, and Grading Rules: Psychometric Properties of Short Answer Items. AERA 2019.

4. Mongkhonvanit, K., Kanopka, K., & Lang, D. (2019, March). Deep Knowledge Tracing and Engagement with MOOCs. In Proceedings of the 9th International Conference on Learning Analytics & Knowledge (pp. 340-342). ACM.

5. Benotti, L., Bhaskaran, J., Kjartansson, S., & Lang, D. (2018). Modeling Student Response Times: Towards Efficient One-on-one Tutoring Dialogues. In Proceedings of the 2018 EMNLP Workshop W-NUT: The 4th Workshop on Noisy User-generated Text (pp. 121-131).

6. Domingue, B. W., Lang, D., Cuevas, M., Castellanos, M., Lopera, C., Mariño, J. P., ... & Shavelson, R. J. (2017). Measuring student learning in technical programs: A case study from Colombia. AERA Open, 3(1), 2332858417692997.

7. Lang, D., Kindel, A., Domingue, B., & Paepcke, A. (2017). Making the Grade: How Learner Engagement Changes after Passing a Course. Proceedings of Educational Data Mining 2017.

8. Furlong, F., Lang, D., & Takhtamanova, Y. (2014). Drivers of mortgage choices by risky borrowers. FRBSF Economic Letter, 01.

9. Furlong, F. T., Takhtamanova, Y., & Lang, D. (2014). Mortgage Choice in the Housing Boom: Impacts of House Price Appreciation and Borrower Type. Federal Reserve Bank of San Francisco.

10. Hobijn, B., Krainer, J., & Lang, D. (2011). Cap rates and commercial property prices. FRBSF Economic Letter, 2011, 29.

11. Lang, D., & Lansing, K. J. (2010). Forecasting growth over the next year with a business cycle index. FRBSF Economic Letter, 2010, 29.

---

Unpublished Working Papers and Posters

Unpublished Working Papers

• Lang, D. (2017) Digital Canvases and Remote Tutoring.

• Anderson, R. Lang, D. Lee, H. (2017) Modeling Innovation Diffusion in an Online Tutoring Network.

Posters

• Kanopka, K. Lang, D.(2019) Adversarial Examples for Neural Automatic Essay Scoring Systems.

• Kanopka, K. Lang, D.(2018) Deep Knowledge Tracing and Engagement in MOOCs.

• Lang, D.(2018) Air Inequality: A Study of California Schools

• Bennoti, L. Bhaskaran, J. Kjartansson, S. Lang, D.(2018) Modeling Student Response Times: Towards Efficient One-on-one Tutoring Dialogues.

---

Presentations

Recorded Presentations

• UW Data Science for Social Good 2019: Peer Support https://youtu.be/Sp61cGvxn00?t=5348

• Bay Area Learning and Analytics Conference (2019) https://youtu.be/ThRQPWcmfYE
  

Other Presentations

• Bay Area Learning and Analytics Conference (2019). "Clickstreams, Video Engagement, and Course Performance."

• Bay Area Learning and Analytics Conference (2018). "Teacher Professional Development: Social Networks and Influence Maximization."

• Bay Area Learning and Analytics Conference (2018). "Heterogeneous Effects of Cellphone Credit Incentives on Mobile Learning in Africa."

• Center for Education Policy Analysis Alumni Research Conference, Stanford University (2018). "Air Quality Inequality: A Study of San Francisco Schools."

---

Research Partnerships

Yup- Yup is a text-messaged based tutoring company

• Completed Projects: Modelling student response time to tutor questions

• Completed Projects: A randomized controlled trial evaluation of a digital whiteboard in a remote tutoring environment

• Ongoing Projects: Diffusion of teacher practice using social network analysis

• Ongoing Projects: Measuring the impact of badges and gamification in a tutoring platform

Reup- Reup is a college-coaching company that encourages students to re-enroll in college after dropping out

• Ongoing Projects: Predicting college re-enrollment using coaching transcripts and natural language processing

Eneza- Eneza is a Kenyan learning platform that provides primary and secondary education via SMS and feature phones

• Ongoing Projects: Understanding the effect of microcredit incentives in educational platforms

• Completed Projects: Predicting psychometric features of items using natural language processing

---

Awards

• Best Paper Award. (CS 230: Deep Learning) “Deep Knowledge and Engagement Tracing” Kanopka, K. Lang, D. Mongkhonvanit, K. (2018)

• Travel Scholarship. Berkeley Initiative for Transparency in the Social Sciences Research Transparency and Reproducibility Training (2018)

• Travel Scholarship. 21st Summer Institute in Statistical Genetics (SISG) (2015)

---

Projects and Code

• Data Science for Social Good https://uwescience.github.io/DSSG-Peer-Support

• Deep Knowledge Tracing github.com/klintkanopka/dkt2

• Tutor-Response-Time Models github.com/davidnathanlang/cs224u-project

---

Skills and Programming Languages

• Expert: Stata, R

• Proficient: Matlab, Python, SQL

• Novice: TensorFlow, PyTorch

Technical Skills: Social Network Analysis, Natural Language Processing, Computer Vision, Machine Learning, Causal Inference, Regression Analysis, Econometrics, and Psychometrics

---

Employment

• (Summer 2019–Present) Data Science Fellow, University of Washington

  • Identified counseling and peer support strategies using natural language processing tools

  • conducted social network analysis on network platform with more than 500,000 thousand users

  • Met with platform stakeholders and recommended a set of interventions to improve users mental health

• (2018–Present) Researcher, Stanford Graduate School of Business

  • Conducted interviews regarding the XPrize competition, Adult Literacy, incentive prizes, and AI in educational settings

  • Assisted in the organization of a conference on Market Shaping (attendees include Hal Varian, Al Roth, Susan Athey, and other experts in the field of market design and machine learning)

  • Wrote a summative white paper of findings and conference proceedings

• (2015–Present) Ph.D. Candidate, Graduate School of Education, Stanford University

  • Conducted natural language processing on text-message-based tutoring services to improve lesson quality and student engagement

  • Recruited and ran randomized controlled trials on a mobile tutoring network

  • Analyzed Stanford online course clickstream data and recommended policy and feature changes to improve student completion rates

  • Worked with San Francisco Unified School District to develop models to identify dyslexic students and predict student absences

• (2015–2018) Partner/Consultant, Data Dream Team Consulting LLC,

  • Worked with airlines to predict customer frequent-flier redemption

  • Cleaned and massaged a terabyte plus transactional database for preprocessing

  • Discussed and defended modelling choices with clients and financial auditors

  • Recommended changes to frequent-flier programs based on findings

• (2014–2015) Workforce Analyst, University of California Office of the President

  • Evaluated systemwide policies on faculty hiring committee composition and its effects on minority hiring outcomes

  • Reviewed and modeled faculty salary equity studies for all ten campuses

  • Conducted survival analysis based on UC faculty data and other data sources to predict faculty retirement

  • Conducted longitudinal analysis of faculty advancement/separation using a competing risk hazard model

• (2008–2014) Research Analyst/Associate, Federal Reserve Bank of San Francisco

  • Spoke to a wide range of academic, public, and professional audiences about the economic outlook and recent monetary policy decisions

  • Fact-checked President Janet Yellen’s speeches

  • Used data-mining techniques to improve collections on the Treasury’s billion-dollar debt portfolio on a joint project with McKinsey and the Treasury

  • Created and cleaned a variety of datasets which utilized the Consumer Expenditure Survey, McDash Mortgage Data, and other financial datasets

  • Created a data-retrieval add-in that improved productivity and saved our department several thousand dollars relative to proprietary alternatives