# Examples of Statistics Used Inappropriately

Examplesof Statistics Used Inappropriately

InstitutionAffiliation

Examplesof Statistics Used Inappropriately

Inappropriateuse of statistics can be purposeful or accidental. There are cases inwhich individuals maliciously use statistics to influence opiniontowards their inclination which in most circumstances has always ledto adverse spiral effects(Wang, 2011).The misuse of statistics involves a gross violation of thefundamental principles which govern the discipline and is a muchbroader than being a tool for malice. Below are a few instances inwhich statistics have been used inappropriately.

ConfusingStatistical Significance with Practical Significance:In most scientific publications, statistical significance is oftenemployed as a metric to ascertain relevance of certain explanatoryvariables in explaining a response variable (Bickel&amp Doksum, 2015).However, in the interpretation of these outcomes, a majority ofpeople have often referred to the statistical significance of theindependent variables as if it was the practical (economic)significance. This is misleading since a variable may bestatistically significant, but not practically significant.

FalseCausality:Another common misuse of statistics entails using causality andcorrelation as though one implies the other. The correlation betweentwo variables, say A and B, does not necessarily mean that A causes Bor that B causes A. Correlation basically establishes a relationshipbetween variables, but further tests are often performed to establishdirection of causality.

Overgeneralization:In most circumstances, a statistic about a particular group isasserted among the members of a broader group amongst which thecharacteristics do not apply. For instance, if 9 out of 10 surveysshow that Iran is found to have nuclear weapons, one may assert thatthe Middle East is producing nuclear weapons.

Misreportingof estimated error:There are cases in which two research reports are often compared ashaving (exact) similar outcomes based on the estimated error withoutdue regard to confidence level, sample size and type of distribution.Other common mistakes made in the use of statistics includepseudoreplication, gamble’s fallacy and regression towards to themean.

References

Bickel,P. J., &amp Doksum, K. A. (2015). MathematicalStatistics: Basic Ideas and Selected Topics, volume I(Vol. 117). CRC Press.

Wang,C. (2011). Misuse of Statistics. In InternationalEncyclopedia of Statistical Science(pp. 821-824). Springer Berlin Heidelberg.