In this module, you will explore the following aspects of responsible data management:
The module consists of:
Research data are the recorded factual material commonly accepted in the scientific community as necessary to validate research findings. Research data are not limited to raw experimental results and instrument outputs; they encompass associated protocols, numbers, graphs, tables, and charts used to collect and reconstruct the data. They include numbers, field notes or observations, procedures for data analysis and/or reduction, data obtained from interviews, or surveys; computer files and databases, research notebooks or laboratory journals, slides, audio/videorecordings, or photographs.
Research data may be in hard-copy form (including research notes, laboratory notebooks, or photographs) or in electronic form, such as computer software, computer storage/backup, or digital images.
Research data do not generally include: unreported preliminary analyses of data, drafts of scientific papers, future research plans, peer reviews, or communications with colleagues; trade secrets, commercial information, or materials necessary to be held confidential until they are published, or similar information protected by law; personnel, medical, or similar information, the disclosure of which would constitute a clearly unwarranted invasion of personal privacy.
Research materials are tangible physical objects from which data are obtained such as environmental samples, biological specimens, cell lines, derived reagents, drilling core samples, or genetically-altered microorganisms. While these are not considered to be research data, they should be retained consistent with disciplinary standards.
Two characteristics of data are authenticity and integrity (Macrina, 2000).
Authentic data are those that honestly and accurately represent the work conducted. Data collection methods should be designed to minimize mistakes/errors.
Integrity of data speaks to the trustworthiness of data throughout the research data lifecycle. It involves the appropriateness and proper execution of the collection method. Researchers should ensure that the study design and methods are appropriate, and that technology involved is reliable. They should ensure that their design and methods are rigorous (i.e., avoid questionable research practices) and free from bias (e.g., selecting a method in order to obtain results that favor a specific conclusion). Further, they should ensure that they have the requisite authority/permission to collect the needed data (e.g., human subjects, vertebrate animals, biological material). Finally, researchers should take measures to protect the integrity of data after data are collected, during analysis, and when they are stored for long-term retention or preservation.
"Research data management concerns the organisation of data, from its entry to the research cycle through to the dissemination and archiving of valuable results. It aims to ensure reliable verification of results, and permits new and innovative research built on existing information" (Whyte, 2011).
Data management is an integral part of good research practice and comprises activities pertinent to all stages of the research data lifecycle. Data management involves establishing strategies for the creation, use, secure storage, and ongoing accessibility of research data. Data management activities contribute to organizing, documenting, finding, archiving, and re-using your data.
Responsible data management:
Collection and generation of data are integral aspects of research and scholarly activity. In such activities, data:
The integrity of research and scholarly activity depend on accurate, detailed, organized, complete, and accessible data.
UNH is responsible for stewardship of data generated or collected as part of funded research and scholarly activity that is conducted:
This responsibility derives from the Office of Management and Budget (OMB) Circular A-110, Subpart C.53. Further an increasing number of sponsors (e.g., National Science Foundation [NSF], National Institutes of Health [NIH]) have requirements for sharing research data and disseminating research results.
UNH’s stewardship responsibilities include, but are not limited to:
In conducting research as part of the UNH community, investigators are obligated to assist UNH in fulfilling its responsibilities, including the management and sharing of research data.
To fulfill its stewardship responsibilities, UNH adopted a Policy on Ownership, Management, and Sharing of Research Data. The UNH policy delineates rights to and responsibilities for research data, as well as required data management practices that ensure:
The UNH policy applies to all members of the UNH community, including, but not limited to, faculty, staff, and graduate and undergraduate students. It applies to research supported by federal and non-federal funds as well as to unfunded research activities undertaken at UNH.
The federal government normally assigns ownership of data generated in activities that it funds to a grantee institution (i.e., UNH), not to an individual (i.e., principal investigator).
In most situations, UNH owns research data generated by its faculty, staff, and students (see Section 5 of the policy for exceptions). UNH assigns custody of the research data that it owns to the faculty or staff principal investigator, or sponsoring principal investigator in the case of students, who is expected to discharge his/her custodial responsibilitiesFor purposes of the UNH policy, custodianship is the physical possession of and direct responsibility for protecting research data, including accurate recording and proper retention, maintenance, access, sharing, and disposition of the data.
in accordance with responsible data management practices.UNH faculty and staff investigators may appeal a determination of UNH ownership of research data to the Senior Vice Provost for Research, Economic Engagement and Outreach (see Section 12 of the UNH policy on Ownership, Management and Sharing of Research Data for details [the Senior Vice Provost for Research position has been renamed]).
UNH graduate student investigators may appeal determinations of UNH ownership of research data to the Dean of the Graduate School. UNH undergraduate student investigators should direct an appeal to the undergraduate college/school Dean (see the UNH policy on Ownership, Management and Sharing of Research Data for details).
Researchers and scholars are responsible for the recording, management, security, and retention of research data in activities that they conduct or oversee. UNH has developed the following resources to assist researchers with their responsibilities:
In exercising their responsibilities, researchers and scholars should adopt an orderly system of data organization for each project or activity, and should communicate the chosen system to all individuals who participate in the project/activity. This system of data organization should include, for example, a file naming convention and folder structure, version control, plan for file sharing, and procedures for recording and documenting data.
A research data lifecycle depicts the different activities involving data in a research project. They include:
The Data Documentation Initiative (DDI) was one of the first organizations to conceptualize a data lifecycle (Corti et al., 2014). Since then, others have adapted it, including the U.S. Geological Survey.
Research data management planning is essential to ensure efficiency in research projects, and to ensure long-term management of data. A data management plan helps to organize all data-related activities, and to ensure that all individuals involved in a project are aware of their responsibilities. Federal agencies are increasingly requiring data management plans as part of funding applications.
A data management checklist is a useful tool to employ at the start of a research project to ensure that all pertinent issues have been identified and addressed, and that an individual has been assigned to each.
To ensure success of a research project, Corti et al. (2014) state "It is crucial that roles and responsibilities are assigned and not just presumed" (p. 29). This is especially important in collaborative research projects.
Most federal agencies also require researchers to develop data management plans that are submitted as part of proposals for funding. These plans describe the data that will be created in the research, how those data will be managed, and how those data will be preserved, archived, and made accessible. Such plans generally should describe the following components:
All data should be recorded truthfully, accurately, and contemporaneously with their production or observation.
Researchers, scholars, and students should ensure that their data records:
Investigators should record research data consistent with the standard practices of their discipline. In the absence of such standards, UNH’s minimum standard is that research records are written/recorded, dated, and identified by the project title and name(s) of the individual(s) conducting the activity, experiment(s), or other investigation(s).
To ensure the authenticity and integrity of data when recording, researchers, scholars, and students should strive to ensure to:
To ensure the authenticity and integrity of data when recording, researchers, scholars, and students should strive to ensure to (cont.):
For more record keeping best practices, see pages 8-12 of Clinical Tools, Inc., Guidelines for Responsible Data Management in Scientific Research.
Originals of other related data products and descriptions (including calibration data for the instrumentation, information on the operational conditions, and any transformations applied to the data to make them useful) must be kept with the corresponding data.
A detailed description must contain exact information on the original data formats and what has been done to the data.
The source code of any software that was used to prepare the data must also be documented and archived.
Whatever the organizational system adopted, researchers and scholars should ensure that all personnel involved with the project/activity, including key administrative personnel, understand and adhere to the system.
Electronic Lab Notebooks (ELNs) or Electronic Research Notebooks (ERNs) are software tools that can be used to replace traditional paper-based lab or research notebooks. While ELNs can make research documentation easier, it is important for researchers to maintain good record-keeping and data management practices while incorporating ELNs into their workflow.
Overall, ELNs can promote efficiency, organization, collaboration, data integrity, and accessibility in research. They provide:
ELNs come in various options: open-source or commercial, free or fee-based, and customizable systems. There is not a one-size-fits-all ELN solution for every lab or research group. Each ELN has different features that might work better for different research needs, disciplinary practices, and researcher preferences. When considering switching to an ELN, researchers should consider:
When considering switching to an ELN, researchers should consider (cont.):
UNH does not have a separate policy for ELNs. Use of ELNs at UNH should comply with current policies for research data and expectations of funding agencies.
Data analysis involves the systematic examination and interpretation of collected data to derive meaningful insights and draw conclusions. The specific approach to data analysis depends on various factors, such as the research questions or hypothesis and the type of data being analyzed. Ensuring accurate and appropriate data analysis is essential for maintaining research integrity.
A data analysis plan provides a structured approach to guide researchers through the analysis process to ensure that they systematically and transparently analyze their data. Simpson (2015) emphasizes the importance of having a clear analysis plan even before data collection begins. This plan guides researchers from the initial stages of summarizing and describing the data to answering research questions or testing hypotheses and supports the reproducibility and rigor in the analysis.
Data analysis plans generally include (Cressman & Sharp, 2022):
Researchers need to address the following when planning for data analysis:
When conducting data analyses, researchers should follow established accepted practices for data analysis in their field or justify unconventional approaches. Researchers should be aware of several important issues when conducting data analysis, including the following :
There are various tools and techniques available for data analysis, each serving specific analytical needs. Researchers should choose those that best suit their needs for effective data analysis. Options for tools include:
Options for tools include (cont.):
Researchers should explore other available options and assess their suitability based on their requirements. For a list of software tools that are available at UNH, see USNH Academic Software Applications.
Researchers should present data analyses objectively and transparently without introducing bias in how they present methods, visuals, or sample descriptions. Researchers should use version control systems for code and data, and clearly document all data analysis procedures, workflows, and decisions to improve reproducibility and enable anyone to rerun analysis.
Proper documentation of digital image processing steps is essential to ensure transparency and traceability in research. While processing techniques can be valuable for achieving research goals, it is important to present images in a format that accurately represents the original data and to avoid inappropriate manipulations. Further, researchers should provide a comprehensive description of image processing approaches to enable others to reproduce the workflow (Aaron & Chew, 2021).
Questionable practices that researchers adopt that lead to mistakes and misrepresentation of images include:
Researchers should establish an image processing workflow, document that workflow, and document all changes made to an image. The following are guidelines for best practices in image processing:
The following are guidelines for best practices in image processing (cont.):
All relevant observed data should be included in an analysis unless there are valid reasons for excluding or modifying data. Reasons for excluding or modifying data, especially when such exclusions or modifications could alter the findings or interpretation of results, should be recorded and disclosed (The National Academy of Sciences, 1992).
Acceptable reasons for excluding data include instrument malfunction, subject attrition, specimen instability, accidental disruption of procedures, or irrelevance to the hypothesis.
Data imputation is a procedure used to estimate data for missing values. All imputed data should be identified, accompanied by an explanation of the imputation method.
Data manipulation refers to changing, excluding, or creating data, usually without disclosure. The following methods of data manipulation are considered unethical as they distort or misrepresent the true results of the research or scholarly activity.
At UNH, fabrication and falsification of research data are regarded as acts of scholarly misconduct.
Researchers and scholars should provide reasonable access to data by any individual who has participated in the project/activity in which the data were collected or generated.
Prior to starting an activity, researchers and scholars should delineate in writing for each member of the group (or for each student or other personnel involved in the activity) which parts of the data may be copied and taken by the individual if he/she leaves the group (or ceases the activity).
The federal government has the authority to access any data collected or generated in activities that it has funded.
UNH may need to access data and in certain situations take custody of data, such as in patent disputes, allegations of data misuse, subpoena, or Freedom of Information Act/Right to Know Law requests. This responsibility lies with the UNH Senior Vice Provost for Research, Economic Engagement and Outreach (SVPREEO).
If a copyright or patent application might emerge from a study, researchers and scholars should promptly contact the office of UNHInnovation for information and assistance.
Researchers, scholars, and students must adhere to any limitations on access to data stipulated by a relevant agreement, policy, or regulation. For instance, provisions to protect the privacy of human subjects and the confidentiality of data collected from them may prohibit making raw data accessible except to authorized individuals.
Researchers, scholars, and students are expected to share data as well as unique research materials essential to the replication or extension of reported findings consistent with the standards of their discipline where such sharing is not limited by a relevant agreement, policy, or regulation (The National Academy of Sciences, 1992).
Benefits of sharing data include:
In 2013, the federal Office for Science and Technology Policy (OSTP) directed each federal agency with over $100 million in annual research and development expenditures to develop a plan to increase public access to the results ~ digital scientific data and scientific publications ~ of federally funded research.
As a result, most federal agencies require that data and unique materials, such as cell lines and DNA sequences, generated or collected in activities supported by their funds be shared with others in a timely manner after the associated research results have been published or provided to the sponsoring agency.
For example, NASA requires that articles are made publicly available within 12 months of publication, and that data are made publicly available at the time of publication, or within a reasonable time period after publication, which is stipulated in the data access plan.
Some journals also require sharing of materials and/or data described in articles they publish.
Data is part of the open science movement. Open data are "online, free of cost, accessible data that can be used, reused, and distributed provided that the data source is attributed and shared alike" (FOSTER).
Tangible research materials should be shared with individuals who are not affiliated with UNH only by specific written agreement, such as a Material Transfer Agreement.
UNH retains a non-exclusive, irrevocable, royalty-free license to use all research data for purposes of internal research, education, and/or protection of intellectual property when the data are generated at or under the auspices of UNH.
The required retention period is the time period for which researchers and students should retain data securely. It is generally at least three years from the date of data collection, termination of a sponsored agreement under which the data were collected, or publication based on the data, whichever is longer (controlling period).
, and return the data to UNH upon request. UNH-owned original research data must remain at UNH when data are:Students and other study personnel may normally take a copy of UNH-owned research data related to their research when they leave UNH.
Ways to support FAIR data include depositing data in an established data repository (with appropriate access), assigning a persistent identifier (such as a DOI), and providing sufficient documentation with data.
Principal investigators/custodians are responsible for the physical storage and security of research data during collection and retention periods, consistent with the standard practices of their discipline and/or the terms of a sponsored agreement.
Of particular importance are issues involving confidentiality and general management of data obtained from human subjects, security of research data against theft or loss, and maintenance of backup or archival copies of research data that may be needed in the event of a disaster, as well as any software. Storage methods should be secure yet allow access by authorized individuals.
Research data and associated materials/correspondence must be retained in sufficient detail and duration to allow appropriate response to questions about research accuracy, authenticity, primacy, and compliance with laws and regulations governing the conduct of research.
The recordkeeping systems/practices used by researchers should allow unmediated access by UNH over their entire retention period. Of particular importance are instances in which a principal investigator leaves UNH.
The UNH policy requires that research data be retained securely for at least three years from the date of data collection, termination of a sponsored agreement by which the data were collected, or publication based on the data, whichever is longer (controlling period).
If custodians permanently leave UNH and take research data with them, they should notify their Dean/Director of the location of such data. They are obligated to hold the data in trust for UNH and return the data if requested to do so. The data must not be disposed of within the required retention periodThe required retention period is the time period for which researchers and students should retain data securely. It is generally at least three years from the date of data collection, termination of a sponsored agreement under which the data were collected, or publication based on the data, whichever is longer (controlling period).
without written permission of UNH’s Senior Vice Provost for Research, Economic Engagement and Outreach (SVPREEO).Examples of situations in which data may be required to be retained beyond the controlling period include, but are not limited to:
"Data confidentiality refers to protection of data from unauthorized access and disclosure, including means for protecting personal privacy and proprietary information" (NIST, National Cybersecurity Center of Excellence).
Researchers should:
By prioritizing data confidentiality, researchers uphold the core values of science, comply with laws and regulations, and build trust with research participants and partners.
Securing certain types of data is required for regulatory compliance purposes or due to the sensitive nature of the information involved. These include:
Securing certain types of data is required for regulatory compliance purposes or due to the sensitive nature of the information involved (cont.):
For more information visit https://www.usnh.edu/data-services/regulated-data.
Sharing research data related to a published article or otherwise making data publicly available may raise questions about intellectual property rights. These include:
Sharing research data related to a published article or otherwise making data publicly available may raise questions about intellectual property rights (cont.):
The following are some best practices for protecting research data.
The following are some best practices for protecting research data (cont.).
To protect research data during storage and transfer, researchers should consider the following:
To avoid unauthorized use, disclosure, or publication of data, researchers must obtain permission before using or sharing data. This includes obtaining permission from owners of proprietary data and adhering to specified terms and conditions and obtaining informed consent from human subjects and explaining study purpose, data collection methods, use of data, sharing, and risks. Researchers must honor agreements or restrictions on data use and sharing and respect confidentiality of data.
Sometimes it is necessary to use or adhere to written agreements for protecting sensitive information, confidentiality, and intellectual property when using or sharing. Common agreements include:
1. The student may take all original data, including lab notebooks and digital files, as long as she leaves a copy with her faculty advisor (Sponsoring PI).
2. The student may take a copy of all data but the original data, including lab notebooks and digital files, must remain with her faculty advisor (Sponsoring PI).
3. If the student was not paid in the form of salary or wages by UNH, then she owns the data she generated.
4. The student’s faculty advisor (Sponsoring PI) must retain only the raw data; recordings of experimental procedures and digital analyses are not research data.
5. A student may appeal ownership of her data by writing to the UNH Graduate School Dean (if a graduate student) or the undergraduate college/school dean if an undergraduate.
1. The graduate student owns the research data so he should have access to the data.
2. The student has no right to access the data as he has left the research group.
3. The faculty advisor should have entered into a written agreement with each member of the research group at the start of the project delineating which data each individual could access and/or copy and take if s/he left UNH or the group.
4. As the student participated in the project where the data were collected, he should have reasonable access to the data, even though he is no longer a member of the research group.
5. As soon as the student graduated from UNH, according to UNH policy, he should have been eligible to receive a copy of the data from the project.
Click on the image below to read the case study that was presented at the beginning of this module.
Once you have finished all of the review questions click ’Certify Completion’.
