Data Handling

Introduction

Data is the foundation of research. Accordingly, data integrity is paramount. The first step in
good data management is designing an appropriate research plan that ensures collection of
meaningful and unbiased data. By doing this, it will not waste resources including manpower and
funding and will lead to protect human and animal subjects.

Proper data collection, storage and sharing to the research plan are also vital. “Data sharing”
means to make data available to others for validation. If data is not recorded to allow others to
validate findings, their integrity could be called into question. In some circumstances, data must
be withheld from others rather than shared, for example, when securing the intellectual property
inherent in new inventions, protecting the confidentiality of human research subjects and
addressing national security considerations. Who owns data collected in an academic
environment is an often misunderstood area.

Learning Objectives

The Data

“Data” is a broad term used both in everyday conversation and within the academic realm. Data
can be numerical values, visual images or observational records and exist in any media or format.

Different disciplines have different notions of what constitutes data. Data can range from material
created in a laboratory such as an electrophoresis gel or a DNA sequence, to information obtained
in social science research such as a filled-out questionnaire, video and audio recordings or
photographs. Data can be astronomical measurements, microscope slides, climate patterns, cell
lines, field notes, soil samples or results of statistical analyses.

Proper data management is required throughout the research lifecycle from protocol development
to the archiving or disposal of research materials at the end of a project. Researchers must be
prepared to address questions that may arise at each stage.

Legal-Regulatory Concepts

Researchers are required to comply with all applicable laws, regulations, rules and guidelines. It
is also important that they keep in regular and close communication with those who are around
them in order to develop the knowledge to identify the right appropriate data management
techniques and what new measures to take depending on the situation. A researcher who infringes
on the applicable laws, regulations, rules or guidelines may be subject to punishment.
 When one must “comply with” or “conform to” ..., it requires to meet the following:
 National government statutes and regulations, local government regulations and
administrative guidelines of the national or local government level
 Certification standards of private bodies (specialized agencies)

 Policies and rules set by individual’s institution

 Professional ethical standards
 Social norms

Compliance requirements constitute the minimum standards for responsible conduct which every
scholar must not fall below. Among the listed requirements above, these requirements include
complying with statutes and regulations, certification standards and institutional rules.

Integrity, however, is more than just compliance with laws and organizational rules. Integrity is
based on ethical commitments, sources of which include professional disciplines and social
norms.

Privacy and Confidentiality

Data protection requires rules about who may access information and under what conditions
because it sometimes involves privacy issues and statutes for protecting personal information.

A specific statute that applies to data handling is different depending on the nature of the data and
various other factors.

Personal identity data, such as names and addresses, are protected by law as personal information
which includes the information relating to individuals’ health and financial conditions. In the U.
S., “health” data are protected by the Health Insurance Portability and Accountability Act
(HIPAA) and all personnel working in the field of medical services and clinical studies are
required to be trained without exception. Also “educational background” data are protected by the
Family Educational Rights and Privacy Act (FERPA) and “financial” data are by the Financial
Services Modernization Act (FSMA).

In Japan, these data are subject to protection under the laws for the protection of personal
information[Note] if they identify a specific individual. Even if the data per se cannot identify a
specific individual, if the data “is easily verified with other information and a specific individual
will be identified”, it is also subject to protection by law. For example, regarding a set of linkable
anonymous information, if the relevant institution has a table to link the information with a code
or similar to assigned to the specific individual, the information is deemed to be personal
information.

[Note]: Personal information held by the national government, local governments or independent
administrative cooperations (and national university corporations) is protected by a special act
applying to each of these categories. The “Act on the Protection of Personal Information” applies
to other organizations such as private bodies.

Rules that apply to data protection are different in each county and it appears that the European
Union (EU) imposes the strictest controls. If researchers conduct their work outside of Japan,
they must be aware of the relevant regulations and standards applicable to the particular country.
There are guidelines that apply to the medical research area in Japan as follows:
* Ethical Guidelines for Human Genome/Gene Analysis Research (Partially amended as of
November 25, 2014)
* Ethical Guidelines for Medical and Health Research Involving Human Subjects
(December 22, 2014 Public Notice by MEXT)
* Guidelines on Gene Therapy Clinical Studies (Partially amended as of November 25,
2014)
* Act on the Safety of Regenerative Medicine (November 25, 2014 by MHLW)
Please see the website of Ministry of Health, Labour and Welfare for updated information on
relevant regulations and guidelines.

Permission to Collect and Use Data

Many laws, regulations and guidelines require researchers to obtain permission to collect and use
data prior to a project. Factors that must be considered include 1) the kinds of data, 2) the origin
of the data, 3) the purposes for collection. Researchers often lack of knowledge to deal with these
issues. Then, researchers are strongly urged to seek consultation with expertise in their own
organization.

Some examples of these rules as follows:

 In order to collect data relating to human subjects, it is generally required to submit a
written research plan and approved by the institutional review board. In the case of
clinical studies, although this procedure may vary depending on the facility, usually an
institution has its ethical review board, ethical review committee for human genome/gene
analysis and/or a committee to review the studies (clinical trials) conducted in compliance
with the GCP (these are commonly named “IRBs”). Research plans involving animal
experiments may also be subject to review from an ethical perspective (i.e. causing
minimal pain, using alternatives means as much as possible, minimizing the use of
animals). On June 1st 2006, MEXT issued “Basic Guidelines for Conducting Animal
Experiments and Related Activities in Research Institutions”.
 Also, permission (a written approval or consent) might be required from the individual
human subjects who are the source of the data.
 There may be a requirement to itemize of when and how data are going to be used.
 There may be a requirement to describe how data will be protected during a period of
storage and how they will be disposed of at the end of a study.

 There may be a requirement to submit plans for data sharing with research subjects and/or
the general public.
Data Acquisition
 Data Selection

“Data selection” is the process of determining appropriate data types and sources as well as
suitable instruments for collection of data. The data selection process generally is unique to each
discipline and driven primarily by the nature of the project, existing literature and accessibility of
desired data sources.

No research project has unlimited funds this means, selection of appropriate data is usually
constrained by issues of cost and convenience. Also collection of data is not selected solely from
the viewpoint of obtaining the right conclusion and additionally, researchers must recognize the
possibility that cost and convenience could compromise the quality of the research outcome.

Establishing clear data selection standards prior to a project can help to prevent improper data
collection practices such as exclusion of data that would not match the hypothesis.

How Much Data Should be Collected?

Different disciplines have different standards for how data are collected and what constitutes
acceptable rigor for the reliability and validity of results. This is one of the reason why a review
of the existing literature on a topic carefully in advance is imperative when designing a research
plan. For example, if a project includes a quantitative element, a key component of any protocol
design will be the sample size (number of cases or “n”) and the decision hinges on how large an
error is acceptable for an expected effect when estimating an appropriate sample size. If research
designs with an “n” and it’s too small, it will not contribute to the knowledge base. This is
especially the case where animal or human subjects are involved in the project. If the sample size
is too small, the data might be unusable and the project might have to be repeated for a rigorous
test of hypothesis. If the project has an “n” and it is too large, this can also be ethical issue. Due
to the subjects which may needlessly be suffering from unpleasant experiences and pains or even
lose their lives. To avoid these errors and ensure the development of an ethical study design on
the bases of reliable statistical procedures such as a power analysis, it should be used to determine
an appropriate “n”.

Researchers should consider to seek specialized methodological and statistical expertise at any
time when they are uncertain of their abilities with respect to data selection or other
methodological issues (specifically, methods such as design of experiments). In general, it is
unethical to commence a research project if the researcher does not have sufficient
methodological skills to complete the relevant work.

Data Acquisition Methods

Data collection means the process of gathering and measuring variables of interest in an accepted and
systematic fashion. Rigorous collection methodologies may enable researchers to answer the research
questions, test hypotheses being set in advance and evaluate the outcomes.

Although data acquisition methods vary by discipline and data types, the emphasis on ensuring accurate
collection remains the same. It is important to understand both the selection of appropriate data
collection instruments and clearly delineated instructions to prevent any possible error. Researchers are
also required to keep records of the process and methods they use to collect data appropriately.
Personnel with the relevant expertise may also be appointed to take charge of accurate data collection,
quality control and quality assurance if necessary.

Consequences that can arise from improper data collection include:

 The inability to answer to the questions accurately for the objective of the research
 The inability to repeat and validate the study

 Leading to inaccurate findings

 Wasted resources

 Misleading other researchers to pursue fruitless studies

 Misleading decision-making in public policies or private companies

 Causing futile harm to human and animal subjects

While the degree of impact from faulty data collection varies by discipline and the contents of the
project, poor collection practices may cause serious harm as example of when flawed results are
used to support public policy recommendations. Effective methods make the detection of errors
easier whether they are made intentionally or unintentionally. It is critical that researchers have
sufficient methods and skills to collect data properly and it is the same as the case of data
selection. If they are not confident in themselves, outside experts should be sought to review
procedures prior, during and after data collection.

Developing a rigorous and detailed recruitment and training plan for all members of the research
team is crucial. This can help to prevent the problem of staff members unintentionally deviating
from the research plan. In addition, the structure of communication must be delineated clearly for
everyone to be informed about any change in procedure.

For some projects, it may be appropriate to visit the sites where data is collected to observe the
process directly. Whether the data is quantitative or qualitative, a regular review and audit of
records should be performed to verify that collection is proceeding according to the procedures
described in manuals.

Data Management

Storage and Protection

Research data must be stored securely and protected from theft at all times both during or after a
research project. Reliable security policies and procedures should be established for data stored
electronically and in the physical form such as paper files, journals and notebooks respectively.

Regarding to information protection, three core elements must be remembered: confidentiality,

availability and integrity.

Confidentiality

Confidentiality refers to limiting information access, disclosure to authorized users and

preventing disclosures or access by unauthorized ones. In any research field involving human
subjects, respecting the rights, dignity and privacy of the subjects is paramount.
This is important as a part of ethical and professional responsibilities. In the U. S., confidentiality
is a legal requirement for the researchers as well.

In Japan, secure management obligations are imposed on researchers’ institutions by statutes and
ethical guidelines to prevent personal information and data from being disclosed. In addition,
several ethical guidelines impose confidentiality obligation on Principal Investigators and
researchers. If a researcher is a physician, dentist or pharmacist, noncompliance of the
confidentiality obligation is subject to a criminal offence.

In the U. S., federal regulations require attention to the privacy of research subjects including
their personal information.[1][2] HIPAA adds appropriate safeguards requirements for most
research data derived from health care records. Upholding confidentiality requires specification of
data handling responsibilities and privileges.[3][4] That is, making such rules as who may handle
which portion of the data, at what point during the project, for what purpose etc. in advance as
legal requirement.

If research data are maintained on a personal computer, it is essential to follow basic security
steps such as keeping the computer in a physically lockable place, updating software regularly as
a defense against computer viruses and spyware, using individual passwords rather than shared
ones as a protection from illegitimate access and following other secure computing practices.
Many organizations provide technical resources as mentioned above therefore, an individual
researcher is not required to find an expert in this area.

These steps are particularly critical for systems that are maintained outside of a normal office
environment such as at a researcher’s home where there is generally much less of a physical
security system and also an absence of the technical protections typically provided for devices
which is managed on an organization’s network.

Data Access

While protection from illegitimate access must always be important, the greater risk is that the
data might become unavailable to anyone. Therefore, measures must be considered protection
from risks such as fire, water or other environmental damage, or common technical failures such
as hard disk malfunctions.

It is important to make frequent, periodic backup copies of data sets if possible and store these
copies with a password in a secure off-site location that is protected both from intruders and
environmental threats. Primary resources, collections and archives must be preserved in a secure
and safe manner appropriate to the material. Adequate records of what data is stored, and where
will make easier to find when needed and prevent future loss.

Most universities and research institutions have a common data center which provides a safer
storage space than a personal computer. In many cases, a system itself creates periodical backup
copies. Researchers should check if the system is available for their use and if it is, they should
use it.

Integrity

Integrity has the third element of information protection which refers to the trustworthiness of
information as described below:
 It includes the concept that data have not been altered inappropriately either by accident or
deliberate misconduct after they have been recorded.
 It concerns that the data is from the person or entity that the user believes it is theirs and is
not from a different person by error.
 Integrity may even include the integrity of data recording which is the notion that the
person or entity in question entered the correct information reflected the actual
circumstances and under the same circumstances would generate identical data.

A key method to ensure data integrity, whether storing or archiving data and manually or
electronically, is configuration management. This involves keeping track of data on all the media
and formats used during different stages of the project. For example, the raw data could consist of
a digital recording in an oral history project which is transcribed to an electronic written
document, and then excerpted for the use in a print publication. “Configuration management” is
the method that is keeping track of all data and upgrading it to newer media or formats as
necessary during the life of a particular project.

Data collection and storage increasingly involve electronic systems which provides convenience
and cost savings. Yet, they also raise the issue of complexity of data protection. A single system
failure could compromise an entire data set. Such failures may occur without warning even if the
researchers have taken reasonable precautions. Some important records still remain on paper and
that case likely won’t change. Regardless of the medium, appropriate precautions are necessary to
maintain the integrity and accessibility of data.

Retention and Disposal

Data handling procedures should describe when, how and who may handle data for storage,
retrieval, disclosure, archive and disposal purposes.

How long should the data be kept after a project is over? The answer usually depends on:
 The nature of the project including potential ongoing interest in or need for the data.
  Costs of maintaining the data over the long run.
 Contracts or guidelines set by the research sponsor.

For example, under U.S. Department of Health and Human Services (DHHS) requirements,
research records must be maintained for at least three years after the last financial statement.
Other federal regulations or institutional guidelines may require that data be retained for a
different period of time.

In Japan, documents relating to research are generally required to be maintained for five years
after the completion of the project.

Although medical records are required to be maintained for five years, there is no clear
requirement of the period for research records to be maintained. In many scientific journals
require to maintain submitted manuscripts for five years after publication and there may be
institutional rules for the period of these records to be maintained.

In the case of clinical trials, commonly named “GCP Ministerial Ordinance” (Ministerial
Ordinance on the Standards for the Implementation of Clinical Trials on Pharmaceutical
Products) requires institutional review boards and medical institutions implementing clinical
trials to maintain records by whichever is later date on that manufacture and sales approval is
granted or three years have passed from the completion of clinical trials.[5]

Retaining data on paper files or electronic media for a long time after the end of a project may
increase the chances of unauthorized access. Risks may increase when researchers leaving the
project, transferring to another institution, retiring from the institution or die without establishing
proper data management procedures including securing disposal or archival storage.

Disposal of Data

Disposal of confidential data requires care and technical expertise to ensure that the data cannot
be retrieved or reconstructed. Multiple-pass erasures are required when disposing of magnetically
recorded data stored on recording discs such as computer hard drives. Optical media may need to
be over-written, if possible, or shredded.

If researchers lack the required expertise and tools, appropriate technical resources from their
organization should be sought or a third-party disposal contractor to be hired. Disposal of data
stored in laboratory notebooks or paper files may also be asked to the disposal contractor which is
same as the case of the electronic media.

Falsification of Data
Improper selection of data
Possible improper behavior when researchers conduct an experiment to prove their own
hypothesis is to delete or otherwise edit data that appears to fall outside of the range of expected
results. Taking part in this behavior, either intentionally or unintentionally, to misrepresent results
constitutes research misconduct as distorting experiment results and misleading conclusion.

Alteration of Images
Today, most images on research article are digitized. With the advent of digital imaging
techniques coupled with inexpensive and easy-to-use software, the doctoring of images from
scientific research has become much easier. Examples have come to light recent images
submitted for publication that have had parts removed or parts of other images pasted into them.
It is obviously unethical practice that is altering an image to change of its meaning or eliminating
aspects which is unmatched with the conclusion a researcher is trying to prove. However, other
changes that may be acceptable is to improve the contrast to make easier to see important features
as long as the alteration is minor and does not change the overall meaning of the image. Many
journals do not have guidelines on acceptable practices for alternation of images. It is essential for
each researcher to use their own judgment to ensure that Images are altered to the extent which is
not changing the meaning of the image.

Data Analysis
Statistical Examination

Data analysis refers to the process of systemically applying statistical or logical techniques to
describe, illustrate, condense and evaluate data. The method of analysis is determined by the
approach taken in the data collection and the type of actual data. Regardless of the method of
analysis, the aim of statistical examination is to produce an accurate research conclusion on the
bases of a solid data foundation.

Researchers must attend to a number of questions regarding data analysis, the most important of
which is the need for the relevant skills. Researchers are sometimes overconfident and assume
they have received sufficient training. This can easily lead to unintentional errors. If researchers
have any doubts about their level of expertise, review of proposed protocols is critical particularly
proposed analytic methods.

Analysis Methods Should be Identified Before Data Acquisition

As well as data selection criteria, the selection of statistical analysis methods always should
precede data collection. If it is delayed in the research process, this may increase the risk that
analytic choices will be made based on which method produces the most favorable results. In
other words, it will increase the possibility of bias. Each discipline has developed accepted
practices for data analysis. If an unconventional approach is used, it is crucial to state clearly that
it is the method, explains why shows how this new and possible untested method of analysis is
being used as well as how it differs to the traditional methods.

Whether statistical or non-statistical methods are used, researchers should clarify the limitations
and possible biases of their methods to both themselves and their readers or reviewers. Regardless
of whether one studies quantitative or qualitative phenomena, researchers use a variety of tools to
analyze data in order to test hypotheses, discern patterns of behavior and ultimately answer
research questions. Failure to understand or acknowledge data analysis issues can compromise
the integrity of the research.

Publication and Reporting

Publication to scientific journals or reporting in academic conferences is the process of preparing

research findings and disseminating them to the scholarly community for making the research
results open to criticism. Disciplines advance through reporting, observing and listening to
research findings at professional meetings and in journals. There must be a robust trust
relationship between authors and readers/audiences regarding the accuracy and truthfulness of
data.

In addition to the stage of data acquisition and analysis as discussed above, the practice of
ensuring research integrity extends to the stage of documenting and preparing results for
publication. Researchers may experience many ethical challenges in this final stage due to:
 Publication pressures stemming from professional competition and job insecurity
 The lack of proper guidance or bad examples from advisors regarding publication

 The lack of required methodological skills for publication

 Unclear guidelines for publication, lack of penalties or a perception that there is little
chance of getting caught
 Ego (desire for achieving academic degree or promotion) or vanity
 Tempting opportunities for personal financial gain
 Language barriers for presentation
 Pain in writing/reporting

Researchers who demonstrate a lack of integrity while engaged in data reporting and publishing
could profoundly harm the direction of future research efforts and cause loss of trust in the
relevant area. Ultimately, they could also risk the well-being and safety of the general public and
research subjects.

In Japan, in relation to publication of research findings, it has been pointed out that there is a
tendency of disseminating exaggerated significance of research findings to the general public.
Although some researchers consider that it is due to problem in media reports because mass
media generally takes the role of disseminating information to the general public, the
exaggeration may be partly caused by researchers themselves. Researchers have a desire to attract
public or private fund for supporting their research and are also under pressure to attract such
fund which could lead them to report research findings with exaggeration as related to the first
example in the list above. Since a huge sum is allocated from the national budget to support a
major research project, researchers should be more self-disciplined as a result, such an important
project will not be jeopardized due to loss of public trust in research which may be triggered by
the practice of exaggerated reporting.
 Sources of guidance for promoting good data, reporting practices and publishing are obtained
from respected faculty advisors and department chairpersons, university policies, codes of
professional ethics and government rules and regulations. Deficiencies in training or a lack of
awareness of existing policies, codes, or rules may increase the possibility of a deviation from the
acceptable ethical standards of practice.

Problems relating to the integrity of data reporting and publication include:

 Misrepresentation of data quality or of the data itself
 Fabrication or falsification of data

 Arbitrary interpretation of data in an effort to produce significant results

 Unfair evaluation of prior research in order to exaggerate the significance of individual’s
own research
 Misleading discussion of observations
 Reporting conclusions which are not supported by data
 Failure to disclose conflicts of interest
 Plagiarism
 Misattribution of authorship (e.g., ghost authorship: failure to attribute authorship to
someone who has made a significant contribution to the research; gift authorship:
attributing authorship to someone who has not made a significant contribution to the
research)

All researchers should be aware of these problematic practices that could compromise the
integrity prior to reporting and publishing data.

Data Sharing and Ownership

[Note] Under the Japanese laws, “ownership” is established only on tangible objects and is not on
information. However, some of the information including “intellectual properties” such as patent
and “know-how (or trade secret)” may be owned exclusively based on special provisions of
statutes. In addition, rules of a research institution may stipulate that information as research
results belongs to the institution. As following, the word “ownership” or “property right” is used
in the context that information is exclusively owned by or belongs to a particular person or entity.

In order to prevent problems, it is important to make decisions prior to a study begins regarding
how to handle the offspring of tangible biological materials such as cell lines, mice and vectors
used in research as well as any modifications or alterations produced through genetic engineering
involving these tangible biological materials. When tangible biological materials are being
exchanged between multiple institutions, it has become common to enter into a Material Transfer
Agreement (MTA).

Issues of Ownership
Data “ownership” generally means that a researcher possesses and has responsibility for the
information. It covers the range of rights and obligations with respect to data collection and
sharing. Information control is subject to restrictions by technical capabilities to access, create,
modify or package it as well as by legal-regulatory constraints.

As discussed previously, rights and obligations related to ownership of data depend on the data
type (for example, health, education and financial) as well as in the facility where the data are
collected. Rights and obligations also may be governed by a “data-use agreement” for a particular
study, which sets limits on how the data are handled including whether it can be transferred to
third parties or disposed of at the end of a project.

All members of a research team should review their institution’s policies and rules with respect to
data ownership and make sure that they have fully understood the contents.

In Japan, research institutions (such as universities, research centers and hospitals) usually have
institutional working regulations or other rules concerning intellectual properties in which
ownership of information is stipulated. These rules often include provisions regarding
“information relating to research results” or “information obtained in the performance of one’s
duty” in which matters relating to data ownership are stipulated. In addition, some institutions
may require written consent such as by “oath” at the time of employment or participation in a
project in order to ensure that data ownership belongs to the institutions.

For example, graduate students and post-doctoral fellows participating in a research project may
sometimes falsely assume that the data, which they have collected, belongs to them. However,
post-doctoral fellows are in fact employees of the university and any tasks achieved during their
fellowship are considered part of their professional duties. In such cases, the university is
normally the owner of the data. On the other hand, because employment relationship does not
usually exist between a university and its graduate students in Japan, the circumstances are a
slightly different and search is not a professional duty (however, graduate students who are
employed by the university as teaching or research assistants are subject to the same treatment as
post-doctoral fellows). Graduate students are granted admission into a university on the
understanding that they will abide by the university rules. University regulations often state that
even if they are not employees of the university, data obtained by graduate students basically
belongs to the university since it was obtained by using the university’s facilities and funds
therefore, the university is the owner of the data. Of course it is not always the case and the head
of the laboratory or the principal investigator should discuss with graduate students prior to
starting a research project for the students to understand and respect the university rules on data
ownership.

Also rules of universities often stipulate when a faculty member conducts a research project under
a contract with the university i.e. as assignment from the university being involved in a research
project using the university’s facility, staff or funding (this case is also deemed as “duties of
employees”), the university owns the rights to the data. As for patent rights, an individual
employee who is the inventor initially owns the rights and those rights are transferred to the
university in accordance with the university’s rules for employee inventions.
In addition, if there is a third party sponsor under a collaborative research agreement or a
sponsored (contract) research agreement, stipulations regarding intellectual property rights may
be included in an agreement with the sponsor or the research funding agency.

In the case of government funded research, although the university usually owns the data, often
delegates the responsibility for data management to the principal investigator (PI). Then the PI
takes responsibility for data collection, recording, storage, retention and disposal. When data are
published, the copyright is retained by the PI who typically transfers it to a journal.

Recently, many universities and public research institutions have established regulations
requiring that their researchers refrain from bringing research data and materials that are not the
property of the institution from outside, and conversely, refrain from taking out research data that
is the property of the institution. There has been increasing awareness that research data and
materials are the property of the institution, and it is a major violation to take them out without
permission and use them to publish papers or apply for patents.

In addition, researchers must be cautious when taking research data and related materials in or out
of the country since their actions may violate the country’s laws related to security export control
(in Japan, the Foreign Exchange and Foreign Trade Act [“Foreign Exchange Act” for short]) and
carry the risk of criminal prosecution*. To take research data and related materials out of the
country, researchers need to apply to the relevant METI agency and get authorization from the
Minister of Economy, Trade and Industry. Designated cargo and delivery destinations are listed
in Appendix 1 of the Export Trade Control Order, and designated technologies and delivery
destinations are listed in the Appendix to the Foreign Exchange Order. Researchers need to be
aware of that the regulations apply not only to cargo but also to technologies.

* It may be a good idea to access METI’s webpage on security export control to check whether the
goods you plan to bring in or out of the country are subject to regulations, or to make an inquiry to the
contact number listed on that page.

Laboratory Notebooks

Ownership of data and data books developed by undergraduates, graduates, and postdoctoral
fellows on a research project generally belongs to either the grantee institution or the PI.
Therefore, trainees should not assume that they will be allowed to take “their data” when they
leave. Trainees should discuss with their advisors or institutions as appropriate arrangements need
to be made in advance to minimize misunderstandings.

The Bayh-Dole Act of 1980

The Bayh-Dole Act of 1980 in the United States allowed universities to control the intellectual
property such as patents generated from federally funded research that they conduct. In Japan,
too, similar provisions have been stipulated in 1999 (commonly named “Japanese Bayh-Dole
provisions”).[6] Universities could exclusively license its intellectual property to gain profits.
Many universities have benefited from the licensing revenue that has flowed from this type of
arrangement however, these actions have been criticized as “sell-out” of universities to business.

When researchers leave institutions where they conducted research, they usually must negotiate
agreements to retain their grants and data. In the case of an industry or privately funded research
project under a collaborative research agreement or a sponsored (contract) research agreement,
data are usually co-owned by the sponsor and the university as they are deemed to be research
results or information relating thereto. In some cases, there is such a stipulation that the sponsor is
the sole owner of the data. However, in this case also, the right to publish data may remain in the
researcher. On the other hand, even in the case of co-ownership, there may be such a stipulation
that the data obtained by the research are subject to confidentiality obligations or requirement for
the sponsor’s prior consent in order for a researcher to publish the data.

The emphasis is that researchers should review the policies of their institutions as well as the
terms of any contract signed with a funding source. It is equally important in collaborations, all
parties should have a clear understanding of who will determine how the data will be distributed
and shared prior to starting data collection between faculty peers, students, or institution staff.

Rights and Obligations of Sharing Data

Professional norms and a long scientific tradition promote the principle of “openness”. In other
words, it is assumed that researchers will publish significant research results and engage in the
free exchange of information. Such sharing of information benefits the research enterprise: other
researchers learn new techniques, gain new perspective and avoid wasting social resources by
repeating unfruitful experiments.

The costs and benefits of data sharing should be considered in their ethical, institutional, legal and
professional dimensions. In the U.S., many federal agencies encourage data sharing. For example,
the National Institutes of Health (NIH) and National Science Foundation (NSF) has a Data
Sharing Policy.

Sharing has benefits for individual researchers since it enables to lead to collaborations. Yet,
some information should not be released because of privacy and human subject protection
concerns. Also there are times when releasing data may potentially cause harm to the public, a
typical example of which is illustrated by the controversy over publish papers on the bird flu
virus[7].

Further, releasing unpublished data at an early stage of research can jeopardize a researcher’s
ability to develop the research based thereon and his advantage to be the first finding. Data that
could lead to a patent also should not be shared before completing certain procedures. However,
it can be said that researchers are generally expected to share raw data after publication as long as
doing so does not create an undue financial burden.

Researchers may face a dilemma of balancing between willingness to share data in order to
facilitate scientific progress and the obligation to sponsors, collaborators and subjects to preserve
and protect data from disclosure, although, they must learn how to handle such dilemma. Signed
agreements of nondisclosure such as non-disclosure agreement or research agreement involving
non-disclosure liability between researchers and their corporate sponsors may stop data from
being published or shared with colleagues. The appropriateness of these agreements largely
depends on the specific circumstances. For example, from the view point of complying with a
confidentiality agreement, it may appear ethical not to publicly disclose research data that
suggests a possible negative impact on social welfare, what about from the humane perspective?
Even in such circumstances, recklessly disclosing data will not necessarily be considered as the
right course of action.

In any case, the key point is to understand the issues affecting ownership and sharing of their
research data in order to make decisions that balance the promotion of scholarly inquiry against
protection of the interests of the parties involved.

Clinical Trial Data, Conflicts of Interest and Research Misconduct

In a recent incident that caused controversy in Japan, it was reported that it was an employee of
the company selling the drug who performed the statistical analysis, not a researcher of the
university that had conducted the study in an investigator-initiated clinical study of a particular
drug. The problem lay here in a conflict of interest in the processing of the data (i.e., someone
who stood to gain or lose significantly depending on whether the data reflected positively or
negatively on the effectiveness of the drug had been involved in preparing the data). In addition,
the entire series of papers reporting the results of that clinical study was retracted on the basis that
there were problems over the reliability of the data. This was an issue of research misconduct.
Whenever a conflict of interest or research misconduct arises in the acquisition or processing of
data, the accuracy of the clinical trial and the effectiveness of the drug is called into question.
Researchers need to be aware of mishandling data that can lead to this kind of major problem.

Conclusion

Data integrity is at the core of research. Thus researchers are obligated to ensure that they collect
data in a professional manner. To achieve this goal, researchers are required to be mindful and
diligent both in the acquisition and the management of their data. Because data serves as the
building blocks that allow research fields to advance, researchers have the obligation to share
their data responsibly with colleagues and others.