Data Fabrication vs. Data Falsification
By Dr. Zubair Khalid, DVM, MS, PhD ·
Quick Answer
- Fabrication means inventing data or results that never existed, while falsification means manipulating or altering real research materials, data, or processes.
- The practical first step when reviewing research records is to compare raw data against reported results to identify whether missing or altered values indicate fabrication or falsification.
- Both forms constitute research misconduct under official policies, but distinguishing them matters for accurate reporting, investigation scope, and corrective action.
Understanding Research Misconduct Definitions
Research integrity depends on clear definitions that allow institutions, journals, and funding agencies to respond consistently when allegations arise. The Committee on Publication Ethics Core Practices establish expectations for handling misconduct allegations, including the need for clear policies and procedures that address fabrication and falsification as distinct categories of research misconduct. These practices guide journals and institutions in responding to allegations with fairness and consistency.
The National Library of Medicine provides access to authoritative biomedical books and research method references that describe the standards of responsible research conduct. These resources help researchers understand the expectations for honest data collection, analysis, and reporting that form the foundation of scientific work.
Research misconduct is commonly defined through three primary categories. Fabrication involves making up data or results and recording or reporting them as if they were real. Falsification involves manipulating research materials, equipment, or processes, or changing or omitting data or results such that the research is not accurately represented in the research record. Plagiarism, the third category, involves appropriating another person's ideas, processes, results, or words without giving appropriate credit.
The distinction between fabrication and falsification is also practical. Investigators, institutional review boards, and funding agencies need to know which type of misconduct occurred to determine the scope of the problem, identify affected publications, and implement appropriate corrective actions. Researchers who understand the difference can more accurately report concerns and participate effectively in investigations.
Core Definitions and Boundaries
Fabrication Defined
Fabrication is the invention of data or results that were never collected or observed. A researcher who fabricates data creates entries in laboratory notebooks, spreadsheets, or published figures that do not correspond to any actual experiment, measurement, or observation. The fabricated data may be entirely invented or may be constructed to fit a desired outcome.
Examples of fabrication include creating patient records for clinical trials that never enrolled those patients, inventing laboratory measurements for samples that were never analyzed, or generating Western blot images using image editing software instead of actual experimental results. Fabrication can occur at any stage of the research process, from initial data collection through final publication.
The key characteristic of fabrication is that the data have no basis in reality. The researcher has not performed the work described, and the reported results cannot be traced to any actual experiment or observation. This makes fabrication particularly serious because it introduces false information into the scientific record that may mislead other researchers and waste resources.
Falsification Defined
Falsification is the manipulation of real data or results to change their meaning or outcome. Unlike fabrication, falsification begins with actual data that was collected through genuine experiments or observations. The researcher then alters, omits, or manipulates that data to produce a different result or to make the data support a particular conclusion.
Common forms of falsification include deleting data points that do not support the desired conclusion, changing numerical values to improve statistical significance, altering images to enhance or remove bands or signals, and omitting data that contradicts the research hypothesis. Falsification can also involve manipulating equipment or processes to produce data that appears to support the intended result.
The key distinction is that falsification involves real data that has been changed. The researcher has performed the work, but the reported results do not accurately represent what was actually observed.
Overlap and Distinction
While fabrication and falsification are distinct categories, they can overlap in practice. A researcher might fabricate some data points and falsify others within the same study. The investigation must identify each instance and classify it appropriately.
The practical distinction matters for several reasons. First, the response to an allegation may differ depending on whether data was invented or altered. Second, the scope of the investigation may differ, as fabricated data may require checking whether any experiments were performed at all. Third, the impact on the scientific record may differ, as falsified data may be more difficult to detect because it is based on real observations.
Research Misconduct Policies and Frameworks
Committee on Publication Ethics Core Practices
The Committee on Publication Ethics (COPE) Core Practices provide a framework for journals and publishers to handle allegations of misconduct. These practices address authorship, peer review, data, conflicts of interest, and misconduct. The misconduct section specifically addresses how journals should respond to allegations of fabrication, falsification, and plagiarism.
COPE Core Practices emphasize that journals should have clear policies for handling allegations of misconduct and should follow a fair and consistent process. This includes giving the accused researcher an opportunity to respond, conducting a thorough investigation, and taking appropriate corrective action when misconduct is confirmed.
The COPE framework is relevant to researchers because it establishes the expectations that journals have for handling allegations. Researchers who understand these expectations can better prepare for the possibility of an allegation and can respond appropriately if one occurs.
NIH Grant Policy and Research Integrity
The National Institutes of Health (NIH) Grants and Funding pages provide official policy context for research integrity in the context of federally funded research. NIH expects that researchers who receive funding will conduct their work with integrity and will report accurate data and results.
NIH grant policy includes requirements for responsible conduct of research training, which covers topics including fabrication, falsification, and plagiarism. Researchers who receive NIH funding are expected to understand these definitions and to conduct their work accordingly.
The NIH also has procedures for handling allegations of research misconduct involving NIH-funded research. These procedures are designed to protect the integrity of the research record and to ensure that public funds are used appropriately.
Data Management and Sharing Policy
The NIH Data Management and Sharing Policy establishes expectations for how research data should be managed and shared. This policy requires researchers to plan for data management and sharing, including how data will be preserved and made available to other researchers.
The Data Management and Sharing Policy is relevant to fabrication and falsification because it establishes the expectation that research data will be available for verification. When data is shared, other researchers can examine it and identify potential fabrication or falsification. This creates a deterrent to misconduct and provides a mechanism for detecting it.
The policy also emphasizes the importance of data integrity and the need for researchers to maintain accurate records of their research. This includes maintaining original data, documenting any changes to data, and ensuring that data is preserved for the appropriate period.
At a Glance
| Feature | Fabrication | Falsification |
|---|---|---|
| Data origin | Data or results are invented, no actual experiment or observation occurred | Data originates from real experiments or observations |
| Primary action | Creating false entries, records, or figures from nothing | Altering, omitting, or manipulating existing data or materials |
| Detection focus | Verify whether experiments were performed and records exist | Compare original records against reported results for discrepancies |
| Example | Recording measurements for samples never analyzed | Deleting outlier data points to achieve statistical significance |
| Investigation scope | May require checking whether any work was done at all | May require detailed audit of data handling and analysis methods |
Identifying Fabrication and Falsification
Common Signs of Fabrication
Fabricated data often has characteristics that can be identified through careful review. These signs are not definitive proof of fabrication, but they may warrant further investigation.
One common sign is data that is too consistent or too perfect. Real experimental data typically contains natural variation, and results that show no variation or that conform too closely to expected values may be suspect. For example, if a researcher reports that all 50 samples produced identical results with no variation, this may indicate that the data was invented.
Another sign is data that is inconsistent with known biological or physical principles. If reported data violates established scientific principles or is impossible given the experimental conditions, this may indicate fabrication. For example, if a researcher reports a measurement that exceeds the known limits of the equipment used, this may be suspect.
A third sign is data that is inconsistent with other data in the same study or with data from related studies. If the reported data does not match the data from other experiments in the same study, or if it does not match the data from other studies in the same field, this may indicate fabrication.
Identifying Signs of Falsification
Falsification can be more difficult to identify than fabrication because the data is based on real observations. However, there are signs that may indicate falsification.
One sign is data that has been selectively omitted. If the researcher reports only a subset of the data that was collected, and the omitted data does not support the conclusions, this may indicate falsification. The researcher may have omitted data points that were outliers or that contradicted the hypothesis.
Another sign is data that has been altered in ways that are inconsistent with the original records. If the original data records do not match the reported data, this may indicate falsification. The researcher may have changed values, adjusted measurements, or modified images.
A third sign is data that has been manipulated through statistical or analytical methods. If the researcher has used inappropriate statistical methods to achieve a desired result, or if the analysis has been adjusted to produce a particular outcome, this may indicate falsification.
Practical Workflow for Assessing Data Integrity
Step 1: Compare Original Records to Reported Results
The first step in assessing whether fabrication or falsification has occurred is to compare the original data records to the reported results. This requires access to the original data, which may include laboratory notebooks, electronic records, or other documentation.
The researcher should examine whether the reported data matches the original records. If the reported data does not match the original records, this may indicate falsification. If the original records do not exist, this may indicate fabrication.
Step 2: Examine Data Patterns and Consistency
The second step is to examine the data for patterns and consistency. The researcher should look for signs of fabrication, such as data that is too consistent or that violates known limits. The researcher should also look for signs of falsification, such as data that has been selectively omitted or altered.
Step 3: Review the Research Process
The third step is to review the research process to determine whether the data could have been generated as reported. The researcher should consider whether the equipment used was capable of producing the reported measurements, whether the experimental conditions were appropriate, and whether the data collection process was consistent with the reported methods.
Step 4: Seek Expert Review
If the researcher is uncertain whether fabrication or falsification has occurred, they should seek expert review. This may include consulting with a statistician, a methodologist, or a subject matter expert who can examine the data and provide an assessment.
Step 5: Escalate to Institutional Authorities
If the researcher believes that fabrication or falsification has occurred, they should escalate the concern to the appropriate institutional authorities. This may include the research integrity officer, the institutional review board, or the funding agency.
Options and Tradeoffs in Responding to Allegations
Institutional Investigation
When an allegation of fabrication or falsification is made, the institution typically conducts an investigation. The investigation may include a review of the original data, interviews with the researcher and other individuals, and an assessment of the evidence.
The investigation process has tradeoffs. A thorough investigation can take time and resources, but it is necessary to determine whether misconduct occurred. A less thorough investigation may be faster, but it may not identify all instances of misconduct.
Journal Response
Journals may also respond to allegations of fabrication or falsification. The journal may issue a correction, a retraction, or an expression of concern, depending on the nature and extent of the misconduct.
The journal response may be coordinated with the institutional investigation. The journal may wait for the institutional investigation to be completed before taking action, or it may take action based on its own assessment.
Funding Agency Response
Funding agencies may also respond to allegations of fabrication or falsification. The agency may suspend or terminate funding, require the researcher to return funds, or impose other sanctions.
The funding agency response may be coordinated with the institutional investigation. The agency may wait for the institutional investigation to be completed before taking action, or it may take action based on its own assessment.
Observations and Measurements
Documenting Data Integrity
Researchers should document the integrity of their data throughout the research process. This includes maintaining original data records, documenting any changes to data, and ensuring that data is reproducible.
The researcher should maintain a data management plan that describes how data will be collected, stored, and shared. The plan should also describe how data integrity will be maintained and how any changes to data will be documented.
Recording Data Management Decisions
Researchers should record their data management decisions, including how data was collected, how it was analyzed, and how it was stored. This documentation can be used to verify the integrity of the data and to respond to any allegations of misconduct.
The researcher should also record any changes that are made to the data, including the reason for the change and the date of the change. This documentation can be used to demonstrate that the data was not falsified.
Records and Measurements
Maintaining Original Data
The original data should be maintained in a secure location and should be available for review. The original data may include laboratory notebooks, raw data files, images, and other documentation.
The researcher should maintain the original data for the appropriate period, which may be specified by the institution or the funding agency. The researcher should also ensure that the original data is accessible to authorized individuals.
Documenting Data Changes
Any changes that are made to the data should be documented. This includes the reason for the change, the date of the change, and the individual who made the change. This documentation can be used to demonstrate that the data was not falsified.
Maintaining Data Reproducibility
The researcher should ensure that the data is reproducible. This means that another researcher should be able to reproduce the data using the same methods and conditions. The researcher should document the methods and conditions that were used to generate the data.
Common Failure Patterns
Failure to Maintain Original Data
One common failure pattern is the failure to maintain original data. If the original data is not preserved, it may be impossible to verify the integrity of the data. This can make it difficult to determine whether fabrication or falsification has occurred.
Failure to Document Data Changes
Another common failure is the failure to document data changes. If the researcher does not document the changes that are made to the data, it may be difficult to determine whether the data was falsified.
Failure to Report Concerns
A third common failure is the failure to report concerns about data integrity. If the researcher observes signs of fabrication or falsification but does not report them, the misconduct may continue and may have a greater impact on the scientific record.
Welfare and Safety Context
Protecting Research Participants
Fabrication and falsification can have serious consequences for research participants. If data is fabricated or falsified, the results may be used to make decisions that affect the health and safety of participants. For example, if a clinical trial reports fabricated data, the results may be used to approve a drug that is not safe or effective.
Protecting the Scientific Record
Fabrication and falsification also have serious consequences for the scientific record. If data is fabricated or falsified, the results may be used to guide future research, which may lead to wasted resources and missed opportunities.
Protecting the Public Trust
Fabrication and falsification can also have serious consequences for the public trust in science. If the public believes that research is not conducted with integrity, they may be less likely to trust the results of research and less likely to support research funding.
Limitations and Professional Escalation
Limitations of the Assessment
The assessment of whether fabrication or falsification has occurred is limited by the availability of the original data and the ability to verify the data. If the original data is not available, it may be difficult to determine whether the data was fabricated or falsified.
Professional Escalation Criteria
The researcher should escalate the concern to the appropriate institutional authorities if they believe that fabrication or falsification has occurred. The escalation should be based on the evidence and should be made in a timely manner.
The researcher should also escalate the concern if they are uncertain whether fabrication or falsification has occurred but they have observed signs that warrant further investigation. The escalation should be made to the appropriate institutional authority, such as the research integrity officer or the department chair.
Frequently Asked Questions
What is the difference between fabrication and falsification?
Fabrication is the invention of data or results that never existed. Falsification is the manipulation or alteration of real data or results. Fabrication creates data from nothing, while falsification changes data that was actually generated.
How can I identify fabrication in research data?
Fabrication can be identified by examining data for signs such as excessive consistency, data that violates known limits, or data that is inconsistent with other studies. The researcher should compare the reported data to the original records and to the experimental conditions.
How can I identify falsification in research data?
Falsification can be identified by comparing the reported data to the original records and looking for signs of alteration, such as omitted data points, changed values, or modified images. The researcher should also examine the statistical methods used to analyze the data.
What should I do if I suspect fabrication or falsification?
If you suspect fabrication or falsification, you should document your concerns and escalate them to the appropriate institutional authority. This may include the research integrity officer, the department chair, or the compliance officer.
What are the consequences of fabrication and falsification?
The consequences of fabrication and falsification can include retraction of publications, termination of funding, loss of employment, and damage to the researcher's reputation. The consequences may also include legal action in some cases.
How can I prevent fabrication and falsification in my research?
You can prevent fabrication and falsification by maintaining original data, documenting any changes to data, and ensuring that data is reproducible. You should also be aware of the signs of fabrication and falsification and report any concerns.
What is the role of the institution in responding to fabrication and falsification?
The institution is responsible for investigating allegations of fabrication and falsification and for taking appropriate action. The institution may conduct a preliminary review, a formal investigation, and may impose sanctions if misconduct is confirmed.
How do journals respond to fabrication and falsification?
Journals may respond to fabrication and falsification by issuing a correction, a retraction, or an expression of concern. The journal may also notify the institution and the funding agency of the misconduct.
Using the Evidence
| Source | Best use in this topic | Important limitation |
|---|---|---|
| Research Methods Resources | official guidance | Check the linked page for current local requirements |
| EQUATOR Network | official guidance | Check the linked page for current local requirements |
| Core Practices | official guidance | Check the linked page for current local requirements |
Related Articles
- Statistical Questions Examples: How to Write Them
- Data Management Plan Template: A Practical Guide for Researchers
- What Is Data Management? A Primer for Life Science Researchers
- Common Limitations in Research: Examples and How to Address Them
- Qualitative Data Examples: How to Recognize and Use Non-Numerical Evidence
References and Further Reading
- Research Methods Resources. National Library of Medicine.
- EQUATOR Network. EQUATOR Network.
- Core Practices. Committee on Publication Ethics.
- NIH Grants and Funding. National Institutes of Health.
- ORCID for Researchers. ORCID.
- Data Management and Sharing Policy. National Institutes of Health.
- NCBI Literature Resources. National Center for Biotechnology Information.
- PubMed. National Library of Medicine.
This article is educational and does not replace institutional policy, professional advice, or applicable safety and regulatory requirements.