Correlation V.S Causation

Introduction

On of the problems linked with the assessment of research based on statistical analysis involves the determination of whether a true cause and effect connection between variables exists, or is it just a statistical relationship known as correlation. While there may be correlation between variables, there is not a causal relationship unless the variation in the independent variable or variables actually causes the variation in the dependent variable. Often correlation is misinterpreted as causation, as is the case in the examples presented in this essay. The first example is from a journal article that says watching TV increases a persons risk of heart disease and non-cancer related deaths. The second two examples are related to transportation, one saying that speeding causes car crashes and the other saying population in a traffic analysis zone (TAZ) causes trips produced. Although correlation is necessary it is not sufficient, it is important that a true causal relation exists before making conclusions.

Body (Example A)

The first study is presented in the journal, Circulation-Journal of the American Heart Association with the title, Television Viewing time and Mortality: The Australian Diabetes, Obesity and Lifestyle Study (AusDiab). The baseline data for the study was gathered between the years of 1999 and 2000. The locations for data collection were chosen based on Census Collector Districts in each of the Australian states and in the Northern Territory of Australia. In all 28,033 houses were approached in the selected clusters. Of these households, 8,800 adults greater than 25 years old were selected and surveyed for the study. Follow up time was either the date of death or November 16, 2006, whichever came first; there were a total of 284 deaths during the 6-year period. From the collected data statistical models were created for analysis. (Dunstan, Barr, & Shaw, 2010)

To make their models, the study did take into...