Chi-Square and Correlation- Classmate Response (1): Topic 4 DQ 1

Chi-Square and Correlation- Classmate Response (1): Topic 4 DQ 1



QUESTION-Correlation is a common statistic to measure a general linear relationship between two variables. Explain why correlation does not equal causation.


Classmate (Wanda)-. Response to the question-

Correlation involves a statistical procedure that tests the relationship between quantitative and categorical variables. It describes the level of relatedness between the two variables. The correlation can be positive or a negative, strong or weak (Yadav, 2018). While correlation does explain that there is a relationship or pattern between the two variables, it does not show the nature of that relationship. If the two variables are related, they are correlated. While there may be a level of correlation, a cause-and effect relationship may exist but does not have to exist. Pearson’s correlation coefficient is a statistical analysis tool that helps to quantify that relatedness between two variables (Corty, 2016).

When there is causation, it shows that there is a cause-and-effect relationship between the two variables, that one event caused another to occur. The relationship can also be ambiguous in the direction of cause. This is considered the “chicken and egg” problem, trying to figure which came first. For example, it is found that those with irritable bowel syndrome (IBS) have different gut bacteria compared to healthy or those without IBS. Is it the IBS that causes the different gut bacteria or is it the different gut bacteria that caused the IBS (Chen, 2021)?

There are times when human nature assumes causation due to correlation. In a study by Bleske-Rechek, Morrison, & Heidtke (2015) they sought to examine the degree to which people in the general community draw causal inferences from hypothetical descriptions of experimental and non-experimental research on human behavior. What they found was that people drew causal inferences from non-causal data while drawing inferences that fit with their intuitive notions, regardless of the findings presented (Bleske-Rechek, Morrison, & Heidtke, 2015). This is why caution must be taken when presenting information to the public. It is the non-scientific mind that will draw conclusions about correlation and causation based on their own personal experiences.


Bleske-Rechek, A., Morrison, K. M., & Heidtke, L. D. (2015). Causal inference from descriptions of experimental and non-experimental research: Public understanding of correlation-versus-causation. Journal of General Psychology142(1), 48–70.

Chen, D. (2021). When correlation does not imply causation: Why your gut microbes may not (yet) be a silver bullet to all your problems. Retrieved from

Corty, E. (2016). Using and interpreting statistics: A practical text for behavioral, social and health sciences. New York, NY. Worth Publishers

Yadav S. (2018).  Correlation analysis in biological studies. J Pract Cardiovasc Sci [serial online. Retrieved from:







Chi-Square and Correlation: Classmate Response

As you stated, correlation is a statistical measure that is commonly used in expressing how two variables are linearly related. Correlation describes the simple relationship between variables without making a statement about cause and effect. Correlation refers to any statistical relationship, whether the relationship is casual or not. According to Schober et al. (2018), correlation is applied in the context of two continuous variables and is usually referred to as Pearson product-moment correlation. Pearson correlation coefficient is commonly used for data with a bivariate normal distribution. For ordinal data or for continuous data that is nonnormally distributed, spearman rank correlation is utilized to measure the monotonic association. The term is used to describe the degree to which variables move-in coordination. A positive correlation is when the two variables involved move in a similar direction. Negative correlation, on the other hand, refers to when two variables move in a different direction.

Causation implies that a change in one variable results in a change in the other variable. Causation describes a cause-effect relationship. Causation mainly has three conditions that include temporal precedence, covariation, and control for third variables. Correlation, however, does not equal or imply causation. A strong correlation can be interpreted as causality, but it can be due to other reasons such as random chance. Random chance is where variables may appear connected without any underlying relationship. A lurking variable may also result in variables appearing to have a strong relationship than the actual relationship. This is common in observational data where correlation does not confirm causation. It is, however, possible to infer causation from correlation through the use of directed acyclic graphs that provides the visual representation of the existing causal assumptions (Rohrer, 2018).


Rohrer, J. M. (2018). Thinking Clearly About Correlations and Causation: Graphical Causal Models for Observational Data. Advances in Methods and Practices in Psychological Science1(1), 27-42.

Schober, P., Boer, C., & Schwarte, L. A. (2018). Correlation Coefficients: Appropriate Use and Interpretation. Anesthesia & Analgesia126(5), 1763-1768.







Get 20% off your first purchase