Statistics Discussion Posts: Correlation and Dependency and Statistical Parameter Covariance
Discussion Post 1
The post analyzes the variables, which are related to correlation and dependency. Having discussed the working hours per week and payment, the post refers to correlation since working 40 hours people could receive different salary, depending on their position and experience. Such statistical measure, when the change of one parameter does not necessary leads to the change of another, is called correlation. It is important to note that correlation is expressed by the possible association between two variables (Mukaka, 2012). The analysis of two other variables in the post, such as “hours worked” and “income earned”, can be related to another type of statistical relationship, dependency. When the change of one parameter leads to the same change of another parameter, such relations are called statistically dependent (Altman & Krzywinski, 2015). Therefore, the analysis of the variables provided in the post helps consider correlation and dependency. It should be noted that correlation does not necessarily mean dependency.
Discussion Post 2
The second post raises an issue of another important statistical parameter covariance. It is another type of relations between variables, which depicts the common change of two variables. In other words, covariance illustrates to what extend two variables can change together simultaneously (CTI Reviews, 2016). Schacter, Gilbert, Wegner, and Hood (2015) have indicated covariance as “a measure of how much two variables change together” (p. 66). The examples in the post measured the volume and cost of healthcare, having indicated that the use of covariance measurement can help define the problems and take measures to remove them. The author of the post has referred the use of covariance to measuring the quality of healthcare. However, CTI Reviews (2016) and Schacter et al. (2015) have noted that covariance statistical measure can be applied in many other fields, such as psychology, and where population should be measured.
References
Altman, N., & Krzywinski, M. (2015). Points of Significance: Association, correlation and causation. Nature Methods, 12, 899–900.
CTI Reviews. (2016). Epidemiology, biostatistics and preventive medicine. Cram101 Textbook Reviews.
Mukaka, M. M. (2012). A guide to appropriate use of Correlation coefficient in medical research. Malawi Medical Journal, 24(3), 69–71.
Schacter, D., Gilbert, D., Wegner, D., & Hood, B. (2015). Psychology: Second European edition. London: Palgrave …
