Which Of The Following Are Not Research Data

Which of the Following Are Not Research Data? Understanding What Counts as Research Data

When scholars talk about “research data,” they often assume a shared understanding of what that term includes. Yet, in practice, the boundary between data that qualifies for research purposes and information that does not can be blurry. Knowing which of the following are not research data is essential for proper data management, compliance with funding agency policies, and maintaining the integrity of scholarly work. This article defines research data, outlines typical categories, clarifies what falls outside the definition, and offers practical guidance for researchers who need to sort their materials correctly.

Understanding Research Data: Definition and Importance

Research data refers to the recorded factual material commonly accepted in the scientific community as necessary to validate research findings. This definition, endorsed by organizations such as the OECD and many national funding bodies, emphasizes three core attributes:

Empirical basis – the material originates from observation, measurement, experiment, or simulation.
Recorded in a durable form – it exists as numbers, text, images, audio, video, or code that can be stored, retrieved, and examined later.
Used to support research conclusions – the data are directly linked to the hypotheses, analyses, or interpretations presented in a study.

Because research data underpin reproducibility, transparency, and accountability, distinguishing them from non‑data items helps institutions enforce data‑sharing mandates, protect intellectual property, and avoid ethical missteps.

Common Types of Research Data

Before identifying what is not research data, it helps to recognize the typical forms that are considered data in most disciplines:

Quantitative datasets – spreadsheets, databases, sensor readings, survey responses, genomic sequences.
Qualitative materials – interview transcripts, field notes, focus‑group recordings, open‑ended survey answers. - Experimental outputs – laboratory instrument logs, spectroscopy spectra, chromatography charts, simulation code and its output files.
Observational records – satellite imagery, astronomical photographs, ecological monitoring sheets, video recordings of behavior.
Derived products – processed tables, statistical models, annotated corpora, metadata that describe the original data.

All of these share the empirical, recorded, and research‑supporting characteristics outlined above.

What Is NOT Considered Research Data

Understanding the limits of the definition prevents over‑inclusion, which can clutter data repositories and create confusion about ownership and reuse rights. Below are common categories that do not qualify as research data, even though they may appear in a researcher’s workflow.

Raw Personal Opinions Without Documentation

Personal reflections, gut feelings, or undocumented hunches lack the recorded, verifiable component required for research data. For example, a researcher’s informal belief that “participants seemed more engaged in the morning” is not data unless it is captured in a structured format (e.g., a timestamped observation log with explicit criteria).

General Knowledge and Common Sense

Facts that are widely known and do not originate from a specific investigation—such as “water boils at 100 °C at sea level” or “humans have two eyes”—are not research data. They are background information that supports the rationale of a study but are not generated by the study itself.

Unprocessed Media Files Without Context

A raw video clip or audio file that lacks accompanying metadata (date, location, experimental conditions, consent forms) cannot be interpreted as research data. Without context, the file cannot be linked to a research question, making it unsuitable for validation or reuse.

Administrative Records Not Linked to Research

Documents such as purchase receipts, travel itineraries, or internal meeting minutes are administrative in nature. Unless they are explicitly collected to answer a research question (e.g., tracking funding expenditures for a budget‑analysis study), they remain outside the research‑data realm.

Hypothetical Scenarios Without Empirical Basis

Thought experiments, theoretical models that are never tested, or purely speculative narratives do not constitute data because they lack empirical observation or measurement. A mathematical proof that derives a new theorem is a contribution to knowledge, but the proof itself is not data; the data would be any empirical tests that verify the theorem’s predictions.

Published Literature Citations

References to previously published works are essential for situating new research, but the citation list itself is not research data. It is metadata about the scholarly conversation, not original evidence generated by the current study.

Software or Code That Is Only a Tool

While analysis scripts and software can be considered research data when they are part of the methodological record (e.g., a custom algorithm developed for the study), generic, off‑the‑shelf software used merely as a tool (like Microsoft Excel or SPSS) is not data. The distinction lies in whether the code embodies novel methodological contributions specific to the project.

Why the Distinction Matters

Data Management and Sharing Policies

Funding agencies such as the NIH, NSF, and the European Commission require researchers to deposit research data in approved repositories. Misclassifying non‑data items as research data can lead to unnecessary storage costs, violate privacy rules (if personal opinions are inadvertently shared), and complicate licensing agreements.

Ethical and Legal Considerations

Certain non‑data items—like raw personal opinions or unconsented media—may contain identifiable information. Treating them as research data without proper de‑identification or consent could breach ethical guidelines and data protection laws (e.g., GDPR, HIPAA).

Reproducibility and Transparency

Clear delineation helps reviewers and future researchers understand exactly what evidence supports the published claims. When non‑data items are mistakenly presented as data, it obscures the analytical pathway and undermines trust in the findings.

Practical Examples: Which of the Following Are Not Research Data? (Quiz Style)

To solidify the concept, consider the following items that might appear in a researcher’s folder. Indicate whether each is research data (RD) or not research data (NRD).

#	Item	Classification	Reasoning
1	A CSV file containing 10,000 survey responses with timestamps and anonymized IDs	RD	Empirical

Quiz Answers and Explanation

#	Item	Classification	Reasoning
1	A CSV file containing 10,000 survey responses with timestamps and anonymized IDs	RD	Empirical observations collected systematically for analysis.
2	A detailed methodology section describing the research design	NRD	Describes how data was collected, not the data itself.
3	A published paper that the current study references	NRD	A citation, not original evidence generated by the current research.
4	A custom Python script developed for data analysis	RD	Novel methodological tool created specifically for the study.
5	A journal article presenting a theoretical model	NRD	Theoretical framework, not empirical data.
6	A dataset from a previous study that’s being used as a reference	RD	If directly incorporated into the current analysis, it becomes part of the empirical basis.
7	A spreadsheet with raw data from a lab experiment	RD	Directly collected measurements or observations.
8	A video recording of an interview without consent	NRD	Unauthorized or unconsented media, not research data.
9	A statistical model’s output (e.g., regression coefficients)	RD	Generated results from analysis of empirical data.
10	A list of hypotheses generated before the study	NRD	Speculative ideas, not empirical evidence.

Conclusion
Distinguishing research data from non-data items is not merely a technicality; it is foundational to responsible research practice. By clarifying what constitutes valid empirical evidence, researchers uphold scientific integrity, ensure compliance with ethical and legal standards, and enhance the reproducibility of their work. Misclassifying non-data items as data risks wasting resources, violating privacy, and eroding public trust in research. As disciplines evolve and data types expand—from big data to AI-generated outputs—the need for precise definitions will only grow. Ultimately, research data must be empirical, actionable, and traceable to support meaningful scientific claims. This clarity ensures that the pursuit of knowledge remains grounded in evidence, not assumption.