Replication Datasets

Replication datasets are datasets and other materials deposited by an investigator that allow another researcher to replicate the analysis in a published article, book, or dissertation. Often these datasets are subsets of the original datasets, containing only those variables necessary for replication of published findings. Many journals, publishers, funding agencies, and graduate departments require investigators to deposit replication datasets in a public archive. Depositing data with the ICPSR fulfills these requirements.

Note: This is a partial list of replication studies at ICPSR and may not list all replication datasets currently archived. Newer replication datasets are added to openICPSR, our self-deposit system.

What is Included in Replication Datasets

Investigators are encouraged to deposit everything that is needed to replicate the results of the study. This includes all data, computer programs, sets of computer program recodes, and extracts of existing data files used in the analysis along with the article itself. Investigators must include a description of the data and methodology that is detailed enough to permit another researcher to replicate the findings.

ICPSR's Role

ICPSR does not approve or alter datasets in any way. They are distributed in exactly the same condition and format sent by the data contributor.

Once an author has submitted a replication dataset, ICPSR will make it available at no cost to the user.

Confidentiality and Copyright

Investigators submitting replication datasets to ICPSR are entirely responsible for ensuring the confidentiality of research subjects. ICPSR does not review these datasets for disclosure risk.

Investigators are also required to affirm that they have the right to publish this material.

Depositing Replication Datasets

Authors may deposit replication datasets using openICPSR, our self-deposit system. Data files are uploaded securely through a Web browser.