When an error, or blunder, as human errors are called in science, strikes an experimental setup, the appropriate response from the experimental team ranges from ignoring it to voiding the results, depending on the blunder's severity. According to Wellesley College, human-caused errors should only be reported when they are a significant factor affecting the outcome of the experiment. Otherwise, such blunders may be safely disregarded in the final report.
Many different types of errors can creep into an experimental setup. Experiments are subject to errors in measurement, human errors and even random errors caused by unpredictable phenomena in the effect being measured. While mechanisms exist to control for imprecise measurements and random fluctuations in data sets, scientists who detect a blunder must decide whether the mistake is serious enough to alter the results of the experiment. If not, according to New Mexico State University, any report of the blunder should be omitted from the final draft of the paper.
If the blunder is serious, however, and affects the outcome of the experiment to a significant degree, it might be necessary to redo the experiment with more careful protocols in place. This helps to ensure the integrity of the reported results, according to Digipac.