The article discusses the proof and measurement of association between two things, focusing on correlation methods in psychology. Spearman outlines the importance of quantifying relationships between variables, emphasizing the need for a numerical symbol to measure correspondence. He explains that correlation is a mathematical fact, but its value depends on the likelihood of further cases showing similar patterns. Accidental deviations, or "variable errors," are common and must be accounted for to ensure accurate results. Spearman introduces the "product moments" method, which calculates correlation by comparing deviations from the mean. He also discusses the "rank" method, which is useful when quantitative measurement is not possible. This method reduces accidental error and is particularly effective for short series of data. However, it may not always provide the same level of precision as the product moments method. The article also covers "cross multiples" and other auxiliary methods for calculating correlation, especially when direct quantitative measurement is not feasible. Spearman highlights the importance of considering the number of cases in an experiment, as a larger number reduces the probable error and increases the reliability of results. He emphasizes that while the product moments method is generally preferred, the rank method is often more practical and effective for many psychological studies. Spearman concludes that the rank method is particularly valuable for short series of data, as it provides a reliable measure of correlation. He also notes that while the rank method may not always yield the same level of precision as the product moments method, it is often more practical and effective in psychological research. The article underscores the importance of accurately measuring and interpreting correlations to understand the relationships between different psychological attributes.The article discusses the proof and measurement of association between two things, focusing on correlation methods in psychology. Spearman outlines the importance of quantifying relationships between variables, emphasizing the need for a numerical symbol to measure correspondence. He explains that correlation is a mathematical fact, but its value depends on the likelihood of further cases showing similar patterns. Accidental deviations, or "variable errors," are common and must be accounted for to ensure accurate results. Spearman introduces the "product moments" method, which calculates correlation by comparing deviations from the mean. He also discusses the "rank" method, which is useful when quantitative measurement is not possible. This method reduces accidental error and is particularly effective for short series of data. However, it may not always provide the same level of precision as the product moments method. The article also covers "cross multiples" and other auxiliary methods for calculating correlation, especially when direct quantitative measurement is not feasible. Spearman highlights the importance of considering the number of cases in an experiment, as a larger number reduces the probable error and increases the reliability of results. He emphasizes that while the product moments method is generally preferred, the rank method is often more practical and effective for many psychological studies. Spearman concludes that the rank method is particularly valuable for short series of data, as it provides a reliable measure of correlation. He also notes that while the rank method may not always yield the same level of precision as the product moments method, it is often more practical and effective in psychological research. The article underscores the importance of accurately measuring and interpreting correlations to understand the relationships between different psychological attributes.