Bivariate Statistics

Examples for Bivariate Statistics

When bivariate statistics is employed to examine a relationship between two variables, bivariate data is used. Bivariate data consists of data collected from a sample on two different variables. The goal of bivariate statistics is to explore how two different variables relate to or differ from each other. For instance, if communication researchers want to examine the relationship between communication anxiety and performance in a public speaking [Page 97]class, they need to work with bivariate data. In this example, communication anxiety can be treated as an independent variable. Independent variable is the variable that can be controlled and manipulated in the study. Suppose that communication anxiety influences overall performance in the class. In this case, performance scores will depend on the level of communication anxiety experienced by students. In other words, performance is the dependent variable as it is controlled or influenced by an external factor or the independent variable (e.g., communication anxiety).

Another example illustrating the use of bivariate statistics can be found in the association of self-esteem with communication apprehension. Suppose that a group of researchers wants to investigate the relationship between self-esteem and communication apprehension. They wants to see if there is any change in the self-esteem scores of individuals, when communication apprehension scores change. In this example, bivariate statistics is used to explore how self-esteem is associated with communication apprehension.

Types of Bivariate Statistics

There are different types of bivariate statistics. Researchers decide on the type of bivariate analysis to use in accordance with the levels of measurement (i.e., nominal, ordinal, interval, and ratio). For instance, to examine the relationship between two nominal variables (e.g., age and preference for a communication channel), researchers need to use a different bivariate analysis than when they explore the relationship between two ratio variables (e.g., self-efficacy and communication competence).

One of the common types of bivariate statistics that is widely used in communication research is correlation. Correlation is a measure of the strength of an association between two variables. This bivariate statistical analysis involves creating a scatter graph and calculating a Pearson’s correlation coefficient. The coefficient is represented by r and provides researchers with the value representing the strength of the association between two variables. The value of r ranges from −1 to +1. Suppose that researchers examine the correlation between self-efficacy and communication competence, and calculate the r as .85. If the correlation coefficient is statistically significant, researchers can conclude that there is a strong positive correlation between self-efficacy and communication competence. While the sign of r represents the direction of the correlation, the size of r shows the strength of the correlation. A positive value of r indicates that two variables are positively correlated, and an increase in one variable will lead to an increase in the other variable. A negative value for r shows that two variables are negatively correlated, and an increase in one variable will result in a decrease in the other variable. It is important to note that correlation does not imply causality between two variables. In other words, it does not inform the researcher about whether one variable causes another. It simply suggests that two variables are related, and this relationship may be caused by one of the variables involved in the analysis or another third variable that is not tested in the analysis.

...