Learn quantitative methods
Each year, ICPSR provides a comprehensive, integrated program of basic and advanced training in research design, statistics, data analysis, and social science methodology.
Transfer data between programs
Understanding research data files
Use data responsibly
Most social science data sets are generated as a result of confidential research on individuals or organizations. Therefore, when conducting research using data sets, it is important to abide by the following principles:
- Use data for statistical analysis and reporting of aggregated information only, and not for investigation of specific individuals or organizations.
- Make no use of the identity of any person or establishment discovered inadvertently.
- Produce no links among datasets that could identify individuals or organizations.
- Do not redistribute the data to anyone other than current students, faculty,or staff of MIT.
- Only use the data for non-profit, scholarly research and instructional purposes.
- Cite the data appropriately in your publications.
In addition, if you make use of any confidential data in your research, you should contact COUHES (Committee on the Use of Humans as Experimental Subjects), the MIT office that reviews and approves research involving human subjects. See also our page on the use of restricted data.
Search in Barton for resources on statistical analysis methods using the following subject headings:
- Mathematical statistics
- Social sciences -- Statistics
- Social sciences -- Statistical methods
- Social sciences -- research
- Social sciences -- methodology
- or name the specific method about which you want information (e.g. Multivariate analysis)
Consult the following reference books:
- Dictionary of statistics / Graham Upton and Ian Cook
- Dictionary of statistics & methodology : a nontechnical guide for the social sciences / W. Paul Vogt. (in print)
- Encyclopedia of statistical sciences / founder and editor-in-chief, Samuel Kotz
- Sage dictionary of statistics : a practical resource for students in the social sciences / Duncan Cramer and Dennis Howitt. (in print)
- Sage encyclopedia of social science research methods / Michael S. Lewis-Beck, Alan Bryman, Tim Futing Liao, editors. (in print)
Consult the Harvard-MIT Data Center Research Technology Consulting Service for:
- Data analysis support and programming advice
- Statistical methodology questions
Match geographic areas
When doing data analysis, many researchers need to know the relationship between different geographic areas (e.g. what counties lie within a particular metropolitan area). Following are some resources of use in matching geographic areas.
Note: when matching geographic areas, it is important to understand:
- If your data will span any length of time, boundaries can change over time. Therefore, many of the resources listed will specify that you're matching boundaries based on the boundary definitions of a particular year.
- Some users need to match Zip Codes to other geographic areas. However, difficulties in defining the land area covered by zip codes led the Census Bureau to designate ZCTAs (ZIP Code Tabulation Areas), generalized area representations of Zip Codes that enable more precise geographic calculation.
For help in matching geographic areas, see MIT GIS Services.