An institutional repository can be used to archive and highlight collections of institutional research products including dissertations, articles, posters, and datasets. The process of ingesting datasets into an institutional repository can be complicated. Because metadata can vary widely depending on how a researcher describes their data, an institutional repository must have mechanisms where data description and metadata can be captured when the data is submitted. Review and appraisal of data submissions will assist in the overall process of ingesting the data into a repository.
Once data is submitted into an institutional repository it has to go through a number of steps in order for it to be preserved and made accessible over time. This process first involves the researcher(s) describing their dataset and signing a submission agreement with specific grant or property rights information that has been agreed upon between the repository and researcher(s). Once this stage is complete the dataset normally must go through a review and approval process where the dataset is analyzed for quality-control in terms of metadata description and file types. The dataset then normally is taken through a transformation process where the metadata descriptors are transferred to XML metadata that is appropriate for that specific repository. Finally, the information package about the dataset (metadata, dataset description, and file type information) is ingested into the repository where it can be archived and made accessible for others to use.