Error in Conclusion Statistically

Error in Conclusion Statistically

Student’s Name

Institution Affiliation

February 13, 2018

DQ1: Error in Conclusion Statistically about Cigarettes

The statistical error made to conclude that cigarettes cause the pulse rate increase is due to the wrong analysis. Undoubtedly, the statistical technique of linear correlation analysis is always used in measuring causation, significance, degree, or direction of co-variation of more than one variable and how they relate to other variables (Sasvári, 2013). As a result, the conclusion afore-mentioned does not validate the situation as the linear correlation stipulated can be used to show the relationship, which is stated otherwise. Additionally, when analyzing the correlation between the two variables, that is, pulse rate and cigarettes, linear correlation provides the strength of the relationship. Notably, when examining two or more variable, one should know correlation not merely implying causal relations. Thus, where there is causal relation as the one depicted above, i.e., as the number of cigarettes it increases the pulse rates, smoking cigarettes ceases to be a causal factor of pulse rates, but instead, it induces pulse rates. This means pulse rate is an effect of smoking cigarettes.

Therefore, it is clear that in the statistical analysis of the above linear correlation one cannot make a valid conclusion whether there exists such a correlation between cigarettes and pulse rates. Moreover, a direct linear correlation is always used to depict just the strength of two variables and not a causal relationship between the variables (Joglekar, 2003). Thus, a conclusion from the above scenario should be one that shows that smoking cigarettes and pulse is related, but none of the variables should cause the other. While a linear correlation may also be used in ascertaining the nature or extent of a relationship, it should not be used to make assumed conclusion as there may be other factors causing the relationship.

Reference

Joglekar, A. (2003). Statistical methods for six sigma. Hoboken, NJ: Wiley-Interscience.

Sasvári, Z. (2013). Multivariate Characteristic and Correlation Functions. Berlin: De Gruyter.