SCI - Correlation Lesson
Learning Targets:
- Define the concept of correlation, including how it indicates a relationship between two variables, and differentiate between positive and negative correlations.
- Discuss the concepts of illusory correlations and regression to the mean.
- Identify the features of experimental methods that allow researchers to determine cause-and-effect relationships.
Courtesy of the AP psychology course and exam description, effective fall 2024. (n.d.). Links to an external site.
Understanding Correlation in Research
Correlational research is a valuable method for detecting relationships between variables. Researchers use correlation to analyze data, make predictions, and identify factors for further study. However, it's crucial to remember that CORRELATION DOES NOT EQUAL CAUSATION.
When we talk about correlation, we are looking at how two variables are related to each other. The strength of this relationship is expressed numerically through a Pearson correlation coefficient, which ranges from -1.00 to +1.00. The closer the coefficient is to 1, whether positive or negative, the stronger the relationship. For instance, a correlation coefficient of -.91 indicates a stronger relationship than +.30.
Pearson correlations have two main components: the number, which indicates strength, and the sign, which indicates direction. In a positive correlation, both factors increase together, while a negative correlation shows that the factors move in opposite directions. It's essential to note that "positive" and "negative" refer to the relationship direction, not the value judgment of good or bad.
In statistics, the Pearson correlation coefficient, often denoted as r, measures the linear relationship between two continuous variables. It ranges from -1 to 1, where:
- 1 indicates a perfect positive linear relationship,
- -1 indicates a perfect negative linear relationship,
- 0 indicates no linear relationship.
The Pearson correlation is widely used in statistics to determine how strong the relationship between two variables is and in which direction it goes. It's commonly used in various fields such as research, social sciences, economics, and more to analyze and understand the association between variables.
Effect size is a crucial concept in understanding this relationship. It indicates the magnitude of the relationship between variables. A large effect size suggests a strong relationship, meaning that one variable can be explained or predicted by the other to a greater extent. The larger the effect size, the more influence one variable has on the other. Understanding the effect size helps us interpret the significance and impact of the relationship between variables in statistical analysis.
A scatterplot is a visual tool used to represent correlation. A perfect positive correlation means both factors increase together, a perfect negative correlation means they move in opposite directions, and no correlation suggests no relationship between variables. While perfect correlations are rare, understanding the direction of correlation is key.
Psychologists use correlational research to describe relationships between variables and experiment to establish cause-and-effect connections. Using statistical measures like correlation coefficients, researchers can predict how closely two variables vary and how well one predicts the other.
In summary, understanding correlation in research is about recognizing the strength and direction of relationships between variables, utilizing tools like scatterplots to visualize these connections, and being mindful of the distinction between correlation and causation in scientific study.
Identifying Positive and Negative Correlation
Review the activity below.
Understanding Correlation and Experimental Methods
Correlations help us identify relationships between two variables, preventing us from mistakenly assuming a connection that isn't there. Illusory correlations can create a false sense of control by linking personal actions to chance events, influenced by the statistical concept of regression toward the mean. This phenomenon explains why extreme outcomes return to a more average state over time. Recognizing regression can help prevent superstitious beliefs, as fluctuations in performance often naturally balance out without needing elaborate explanations. Experimental methods provide researchers with the tools to establish cause-and-effect relationships, offering valuable insights into the impact of various factors on outcomes.
[CC BY 4.0] UNLESS OTHERWISE NOTED | IMAGES: LICENSED AND USED ACCORDING TO TERMS OF SUBSCRIPTION