Data: Tampering, Transformation, Merging
Data tampering is the practice of selectively reporting data incorrectly or creating false results. An example of this would be when data that disagree with the expected result are discarded when there is a proportion of the results that would confirm the hypothesis. While it can be important to remove outliers from results it is important that those results are truly outliers and not just information that disagrees with expected or wanted results. Another example would be when a data collector randomly generates a whole set of data out of a single measurement collected.
Data transformation is a recognised application of a formula to the data gained through a trial. This is often used to make the presentation of data clearer or easier to understand. For example, if measuring fuel efficiency for cars, it is natural to measure efficiency in the form of kilometres per litre. However, if you are assessing how much additional fuel would be required to increase the distance travelled it would be expressed as litres per kilometre. Applying an incorrect formula in this case, would affect the overall data.
Data merging is the act of combining data from multiple experiments in order to gain a better understanding of the situation. One of the most common forms of this is meta-analysis where a person or group compares results from several different experiments whose results have been published. It is important whilst doing this to carefully check that the experiments are the same or comparable. Any differences need to be scrutinised for possible hidden variables. An example might be the different species of mice in an animal test. Also, care has to be taken to make efforts to rule out any publication bias which might have occurred through combining inadequate sets of results.