An example of a data management plan.
Data Management Plan Example
Paper by Hart, 2020, available at https://figshare.com/articles/preprint/Example_of_a_Data_Management_Plan/13218743 under a CC By 4.0 license.
[This data management plan is an example of one that I have submitted planning to have fully open data after the first publication. You will want to change many details throughout for your own needs.]
Roles and responsibilities.
Data collection and analysis will be handled by the PI, assisted by the project coordinator, graduate students and other staff and Investigators. Metadata training and generation will be completed in consultation with FSU University Libraries Data Services Team.
Over the course of this project, the PI will be responsible for the anonymization of the data, the security of the information about human subjects, and management of all the data related to the project. Additionally, the project coordinator will schedule quarterly meetings with University Libraries Data Services Team to review and augment the implementation of this data management plan.
As dictated by Florida State University’s Data Management Policy, the intellectual property to the data will remain with Florida State University, stewarded by the PI. In the case of the PI moving on from the University, stewardship of the data will become the responsibility of University Libraries, with the understanding that the PI is always identified as the creator and author of the data.
- All survey and behavioral testing data (Preschool year, 400 families, kindergarten year, 280 families, first grade year, 224 families), saved in .csv format.
- Questionnaire form given to parents, saved in .txt format.
- Blank copy of the child testing protocol, saved in .txt format.
- Blank copy of the math language child testing protocol, saved in .txt format.
- SAS code used to create .csv datasets.
- Copy of any pamphlets and presentations created, saved in .pdf format, posted on www.figshare.com.
- Research articles and presentations originating from the proposed project will posted in Florida State University’s open access research repository. The publicly accessible documents will also be linked to from the PI’s website (www.XXXX.com) and figshare.com.
Period of data retention.
The project team will retain the data for their own use up until the first publication (anticipated to be in the second year of the project). Following that publication, all project data will be deposited and openly shared through LDbase.org. LDbase is a domain specific repository, housing data related to learning differences. This will allow for ease of access and discoverability for education and developmental sciences researchers. LDbase was built in partnership with NIH, Florida State University (FSU) and FSU Libraries, meaning permanent storage is guaranteed, and the data will be reviewed and re-curated for utility. As such, LDbase aligns as closely as possible to FAIR data principles (Findable, Accessible, Interoperable, and Reusable), and is fully compliant with IRB, HIPPA, and GDPR. De-identified data will be stored using an ODC-by license and will be available to the public indefinitely. Access to all identifiable information about the subjects of the study will be restricted to the immediate research team throughout the project to comply with human subject policies. Summary data will also be accessible in the publications that are generated from this study.
Data format and dissemination.
All collected data will be saved in non-proprietary formats. Full datasets will be made available in .csv format, accompanied by a .txt file describing in generalities the demographic data collected and the methods of the research study.
Metadata will be comprised of two formats - contextual information about the data in a text based document (.txt) and a standardized .xml file. These two formats for metadata were chosen to provide the full explanation of the data (text format) and to ensure compatibility with international standards (xml). The text file will be the human readable version of the xml file. Florida State University Libraries Research Data Services team will provide consulting on descriptive metadata throughout the project to ensure accessibility, discoverability and reusability.
Data will be identified with code numbers, not names. Files will be named descriptively to indicate brief details about the content and version of the data.
Data will be disseminated through LDbase.org, which will ensure that data is publicly available after project completion. The data will be deposited into LDbase.org upon publication of the first study associated with this project, likely in the second year of the project. All contextual details will be provided in the codebook, which will also be accessible in LDbase.org. All ethical, privacy, and sensitive information will be anonymized in the released data sets, in compliance with IRB protocols.
Data storage and preservation of access.
During the course of the study, all data and documentation will be backed up monthly onto two external hard drives, one remaining onsite at Florida State University, and the other at the home of the PI. All data, including files, code and codebooks will be made open access to support replication and reusability of this project. Upon deposit, LDbase.org will become the primary agent of storage, preservation and access to this data.
Policies and provisions for re-use, re-distribution and production of derivatives.
The data as published in the research articles will be reusable under standard publication terms, requiring citation to the article and following policies and guidelines of academic writing. The full data sets as posted in LDbase.org will be reusable under an ODC-by license. All publicly accessible data will be reusable as it applies to continued scholarly purposes. Commercial reuse will be prohibited, following data ownership policies at Florida State University. These qualifications will be made explicitly clear in the metadata, README files, and data sets themselves.
Ethics and Privacy.
During all parts of the project, the data will be accessible only by certified members of the project team. All members of the team will complete the CITI subjects protection training program. Stored data will be de-identified to the extent possible. Names of participants (including informed consent forms) will be kept in a separate locked cabinet, away from the project data. Contact data for the families will be maintained electronically on the single project computer, with only a single linking key used to associate the families with their data. Participants will be made aware of the open data policy during informed consent procedures.
Dissemination of Research
Our plans for dissemination of research, beyond making our data open access as described above, are described in more detail in the Broader Impacts section. These include traditional publications (Open Access being the preferred, but at the very least all papers stored on FSU’s DigiNole) and conference presentations, blog posts and infographics, social media use, reporting individualized results to families, and broader summative findings to the general public via newsletters and brochures, as well as incorporating our findings into preservice teacher training, preschool consortium boards, and teacher PD.