Correlational Statistics
Peter Miksza, Julia T. Shaw, Lauren Kapalka Richerme, Phillip M. Hash, Donald A. Hodges, Elizabeth Cassidy Parker · Music Education Research · 2023
Abstract This chapter explains how correlational statistics are used to summarize and describe associations between variables. Music teachers are often curious about how various phenomena relevant to music teaching and learning may be related to each other. Although identifying correlations among variables can yield important insights, it is important to remember that correlation does not equal causation. That is, just because two variables are related does not mean that one necessarily caused the other. Much can be learned about the nature of a relationship between interval- and/or ratio-level variables by creating a scatterplot. The most common statistical tool used for investigating correlation (i.e., the direction and degree of a relationship) between two interval- or ratio-level variables is the Pearson Product Moment Coefficient (Pearson coefficient or Pearson’s r). The coefficient can range from –1 to +1, with negative coefficients (i.e., less than 0) indicative of an inverse relationship and positive coefficients (i.e., greater than 0) indicative of a positive relationship. Common ranges are –.30 to 0 or 0 to .30, described as weak relationship; –.70 to –.30 or .30 to .70, described as moderate; and –1.0 to –.70 or .70 to 1.0, described as strong. Information is also given about the visualization and statistical analysis of relationships using ordinal and nominal data.