STATISTICAL CORRUPTIONS AND DISTORTIONS

From time to time, statistics are calculated and flashed around without any thought about their level of accuracy. Sometimes, too, they are intentionally misused to support a particular viewpoint or action.

Some statistics, for example, are cited without knowing the individual biases that went into the development of the statistics or the institutional influences that may have heavily influenced the results. Some statistics about a narrow population are used to make inferences about a whole population. The assumption or intent is that a statistic can be generalized. Some statistics may have been calculated incorrectly, thereby providing inaccurate results.

The benefits of identifying the potential statistical corruptions and distortions are twofold. First, it halts propagating bad information that can lead to faulty conclusions in a study. Two, it leads to more accurate decisions because a decision-maker knows the weaknesses in the data being used.

for Identifying Statistical Corruptions and Distortions

• Determine the purpose of the data that will be used.
• Identify the developers of the data.
• Contact the developers of the data, if possible, to obtain the assumptions and statistical approach used.
• If corruptions and distortions were used, determine the disposition of the statistic (e.g., use it with caveats, discard it).