How to access MPPS Datasets
The MPPS datasets are being deposited into two different collections at ICPSR:
-
Public/Student-use datasets contain only portions of each survey, broken into themed collections of data.
For example, the Fall 2013 MPPS survey included batteries of questions on fracking, wind power, the Great Lakes, and Detroit's bankruptcy. Each of those topics has been split out into its own sub-dataset. These small datasets are ideal for use in the classroom. They are also useful for academic researchers who might want to explore the MPPS data before submitting an application to ICPSR for access to the restricted-use datasets. These Public/Student-use datasets are available for free and immediate downloading via ICPSR.
Restricted use datasets require ICPSR approval before they can be accessed. They represent full MPPS surveys; they have not been broken into sub-datasets as is the case for the Public/Student-use datasets described above.
For example, the Fall 2013 MPPS restricted-use dataset includes all of the survey questions from that wave, including those on fracking, wind power, the Great Lakes, and Detroit's bankruptcy. The MPPS staff do still modify these restricted-use datasets to remove jurisdiction and respondent identifiers and to recode other variables in order to protect respondent and jurisdiction confidentiality. However, it is theoretically possible that a researcher might be able, in some rare cases, to use enough variables from a full dataset to identify a unique jurisdiction, so access to these datasets is restricted and approved on a case-by-case basis. Researchers must agree to protect respondent and jurisdiction confidentiality in order to access these restricted-use datasets, and ICPSR provides further protection, as described below.
Researchers interested in using these restricted-use datasets should first review the codebooks on our website (linked below) to see the full list of variables including those not found in the Public-use datasets described above. Those interested in applying for access to these restricted-use datasets should email ICPSR for more information. Once an application is approved, data remain on ICPSR file servers and are accessed and analyzed by the researcher virtually. Analyses may be reviewed for disclosure risk prior to transferring results to the researcher. The fee to access restricted-use data in the openICPSR VDE covers one user for a period of one year. More information is available through ICPSR.
See the restricted-use datasets.
- Merging outside data to MPPS data: Researchers may be able to merge their own data to MPPS data at the jurisdiction level. Researchers should contact CLOSUP staff by email to [email protected] or phone, 734-647-4091 to discuss this possibility.
Data-finding aids
The MPPS surveys have covered a very wide range of topics. To find data of interest, use one or more of the following finding aids:
- The MPPS question database
- Individual survey pages
- MPPS questionnaires
Dataset citation information
Any use of the MPPS datasets should be cited as follows
- Public use datasets: Center for Local, State, and Urban Policy. Michigan Public Policy Survey Public Use Datasets. (2017). [Data set]. Inter-university Consortium for Political and Social Research [distributor]. http://doi.org/10.3886/E100132V28
- Restricted use datasets: Center for Local, State, and Urban Policy. Michigan Public Policy Survey Restricted Use Datasets. (2017). [Data set]. Inter-university Consortium for Political and Social Research [distributor]. http://doi.org/10.3886/E100148V11
Respondent and jurisdiction confidentiality
The MPPS assures respondents that their identity, and the identity of the jurisdictions they represent, will remain confidential at all times. Therefore, CLOSUP modifies both public-use and restricted-use datasets in order to protect that confidentiality. Jurisdiction names and identifiers are stripped from the data, as are respondent identifiers. Other variables that might be used to identify individual jurisdictions are recoded into broader categories, or are also stripped from the datasets.