Interpretation of Experimental Data

Essential questions 

What constitutes experimental data processing?

How unlawful data manipulation and scientific fraud are defined?

Data collection is one of the key steps in experimental research. It is a process of gathering, recording and systemic analysis of the experimental outcomes that can be used to address research questions, test hypothesis, or compare the results.

Once experimental data are collected, they must be processed, organized, analyzed and presented in such form that reveals any relationship that was established in the research experiment. In some cases, the raw data are meaningful and presented (examples of such are genomic, transcriptomic, proteomic data sets, etc.). Most often, the raw data are tabulated, graphed, and/or statistically analyzed. 

Data processing often involves clean up or removal of data points that are obviously erroneous due to the mistakes in experimental procedure, collection, reporting, or because they are not relevant to the purpose of study as they are product of effects not sufficiently accounted for by the hypothesis that was tested.

Exploratory data analysis can involve graphic or tabular representation that can help reveal main trends in data sets. It can also rely on descriptive statistics.

Inferential statistics (regression analysis) can be used to establish correlation between measured experimental values and establish the degree of their connection. 

Data interpretation provides a plausible explanation of the research outcomes. It should discern between correlations, causation, coincidence, and avoid bias regardless of its source. Data interpretation is based on comparison and contrasting experimental data sets, outcomes of statistical analysis, and produces predictions of future experiments.

Not all data interpretations are equally valid. Interpretation requires a logical scientific argument that explains the data. It is largely based on individual expertise of scientist, and available body of knowledge that can be used to make inferences, suggestions, or hypotheses about the meaning of experimental data. As any other human endeavor, data interpretation occasionally can be erroneous but, in most cases, those are honest errors.

Unlawful data manipulations constitute scientific fraud. These include falsification and fabrication. Falsification is the omission or deliberate changing data to support hypothesis that is tested in experimental setting. It can be attained not only through direct data manipulations but also through manipulation of research processes, instrumentation, and materials. 

Fabrication is the deliberate creation or addition of data that were never obtained in experimental settings and were not part of the acquired experimental data sets. It can also take form of research outcomes that are based on incomplete data sets but were claimed to have been based on complete experimental data sets. This is misleading and distorting the true outcomes of research. 

Plagiarism the experimental research happens when a researcher presents results obtained by other researchers as his/her your own. While this form of scientific misconduct is not distorting the results or outcomes of research experiments, it is dangerous as it undermines the ethical foundation of science. Researchers can have access to unpublished data during the peer review process, and in such instances, they should be guided by the principle of data confidentiality and not use them until the data are published.

Federal regulations that were developed by the Office of Science and Technology Policy, identify scientific misconduct as fabrication, falsification, or plagiarism of research results [1]. Honest error or differences of opinion regarding laboratory data and their interpretation do not include research misconduct.

Suggested readings

Science and Technology Policy, Federal Policy on Research Misconduct. Available at https://www.govinfo.gov/content/pkg/FR-2000-12-06/pdf/00-30852.pdf.

The Office of Research Integrity. U.S. Department of Health and Human Services.  https://ori.hhs.gov/

Resnik D. B. (2013). Data fabrication and falsification and empiricist philosophy of science. Science and engineering ethics, 20(2), 423–431. doi:10.1007/s11948-013-9466-z