The Proof and Measurement of Association between Two Things.

By M. Meyer

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Unsupervised Learning of Sentence Embeddings Using Compositional n-Gram Features (2017, neutral): “These similarity scores are compared to the gold-standard human judgements using Pearson’s r (Pearson, 1895) and Spearman’s ρ (Spearman, 1904) correlation scores.”
On the Effectiveness of Parameter-Efficient Fine-Tuning (2022, methodology): “To further show the relationship between the stability and the overall performance, we calculate Spearman’s rank correlation coeﬃcient (Spearman, 1904) for the two ranks.”
Estimates of the Standard Error of Measurement for Curriculum-Based Measures of Oral Reading Fluency (2007, unknown): “The concepts of SEM and CI derive from classical test score theory, which has been the foundation of most psychometric work for the past 100 years (Crocker & Algina, 1986; Guilford, 1954; Spearman, 1904, 1907).”
Multivariate Rank-Based Distribution-Free Nonparametric Testing Using Measure Transportation (2019, methodology): “…earliest approaches, for d1 = d2 = 1, was through the introduction of Pearson’s correlation (see, e.g., Pearson 1920), which was later modified into rank-based correlation measures such as Spearman’s rank correlation (see Spearman 1904) and Kendall’s τ (see Kendall 1938; Kendall and Gibbons 1990).”
A Task-Optimized Neural Network Replicates Human Auditory Behavior, Predicts Brain Responses, and Reveals a Cortical Processing Hierarchy. (2018, unknown): “We measured the amount of BOLD variance predicted in left-out sounds, correcting for both the reliability of the measured voxel response and the reliability of the predicted voxel response (Schoppe et al., 2016; Spearman, 1904).”
Judge the Judges: A Large-Scale Evaluation Study of Neural Language Models for Online Review Generation (2019, background): “…between H1, H2 and the text-overlap evaluators, based on either their decisions on individual reviews, their scores of the generators (by Pearson’s coefficient (Fieller et al., 1957)), and their rankings of the generators (by Spearman’s ρ (Spearman, 1904) and Kendall’s τ (Daniel et al., 1978)).”