Data and Analytics

DIME Analytics’ primary portfolio includes:  Development Research in Practice: the DIME Analytics Data Handbook, which covers the data workflow at each stage of an empirical research project; the DIME Wiki, a one-stop shop for practical guidance and implementation resources for impact evaluation; ietoolkit and iefieldkit, Stata code packages featuring commands to routinize common impact evaluation tasks; data visualization libraries in R and Stata; and Manage Successful Impact Evaluation Surveys, an open-access virtual training covering best practices at all stages of the primary data collection workflow.


DIME Analytics creates tools that improve the quality of impact evaluation research for all. The team supports high quality research across the DIME portfolio, offers public trainings, and develops tools that are freely available to the global community of development researchers. We take advantage of the concentration and scale of research at DIME to develop and test solutions to ensure data work quality across our portfolio, and to make public training and tools available to the larger community of development researchers who might not have the same capabilities.

Public Resources and Training

  • Manage Successful Impact Evaluation Surveys
    A fully virtual course, in which participants learn the workflow for primary data collection, with a focus on remote adaptations during a pandemic. The course covers best practices at all stages of the survey workflow, from planning to piloting instruments and monitoring data quality once fieldwork begins. There is a strong focus throughout on research ethics and reproducible workflows. The course uses a combination of virtual lectures, readings, and hands-on exercises.
    2021 Course Materials | 2020 Course Materials
  • Research Assistant Onboarding Course
    This course is designed to familiarize Research Assistants and Research Analysts with DIME's standards for data work. By the end of the course's six sessions, participants will have the tools and knowledge to implement best practices for transparent and reproducible research. The course will focus on how to set up a collaborative workflow for code, datasets, and research outputs. Most content is platform-independent and software-agnostic, but participants are expected to be familiar with statistical software.
    2021 Course Materials
  • DIME Wiki
    One-stop shop for impact evaluation research solutions. The DIME Wiki is a resource focused on practical implementation guidelines rather than theory. It is open to the public, easily searchable, and suitable for users of varying levels of expertise.
  • Development Research in Practice: the DIME Analytics Data Handbook
    The Handbook leads the reader through a complete empirical research project, providing links to continuously updated resources on the DIME Wiki as well as illustrative examples from a real-life DIME project in Rio de Janeiro. The handbook is intended to train users of development data how to handle data effectively, efficiently, and ethically. The handbook is accompanied by a free virtual course, fully open to the public.
  • DIME Continuing Education Series
    DIME Analytics offers regular trainings on technical topics for DIME staff and affiliates. Recent topics have included: Data Processing in Python Using Pandas, Introduction to APIs, Writing reusable code in Stata using programs and adofiles, Spatial Data in Stata, Optimizing Survey Length, GitHub Pull Requests, and Introduction to Python for Stata users. All Continuing Education materials are publicly available.
  • R for Advanced Stata Users
    This open-access virtual course is designed to familiarize participants with the R programming language and software environment, focusing on common tasks and analysis in development research. The course will build upon comparisons to Stata syntax, and assumes familiarity with the use of do-files, including macros and loops. Modules include: descriptive analysis, data processing, data visualization, and geospatial data.

Reproducible Research

  • Computational Reproducibility Assessment  (code review)
    DIME Analytics conducts computational reproducibility check prior to publication for all DIME researchers and for any interested researchers outside of DIME, for any working paper or publication. The checks verify that a third-party, using the same materials and procedures used by the research team, can exactly reproduce the tables and figures in the research paper. DIME Analytics provides detailed feedback on code organization and assistance with the creation of replication packages. Analytics also offers structured peer code review sessions for research assistants, pre-publication code review, and technical support with data and code publication. Contact dimeanalytics@worldbank.org for details.
  • Reproducible Research Agenda
    Regular events and trainings to promote reproducible research, in collaboration with the Berkeley Institute for Transparency in Social Science (BITSS) and other partners. Events in 2019 included Making Analytics Reusable (eventblog) and Transparency, Reproducibility and Credibility: a Research Symposium (eventpresentationsblog).
  • Quality Assurance for Data Acquisition
    DIME impact evaluations rely on original data, either acquired from partners or generated directly through surveys. DIME Analytics supports all DIME projects to ensure that all acquired data is of the highest quality.
  • Survey Review
    Open source resources for data collection (e.g. checklists), and technical support to ensure high-quality data at each stage of the process: procurement of data collection firms, survey instrument design and programming, enumerator training, ensuring ethical research, and data quality assurance.
  • Measurement Conference
    Annual conference on data and measurement innovations in partnership with the Center for Effective Global Action (CEGA). 
    2021: Emerging Data and Methods in Global Health Research (blog)
    2020: Data Integration and Data Fusion
    2019: Crisis Preparedness and Response
    2018: Artificial Intelligence and Economic Development

Open Source Research Tools

We take advantage of the scope and scale of DIME research to develop and test econometric and technical solutions and develop public tools. 

Experts

Maria Ruth Jones

Survey Specialist
Photo of Roshni-Khincha, DIME Team

Roshni Khincha

Data Coordinator
Photo of Luiza-Cardoso-De-Andrade, DIME Team

Luiza Cardoso De Andrade

Junior Data Scientist
Photo of Luis-Eduardo, DIME Team

Luis Eduardo San Martin

Data Coordinator
Welcome