The proof and measurement of association between two things.

By C. Spearman. 2015

6,996 citations recorded by the source corpus.

Subfields: Atomic and Molecular Physics, and Optics, Experimental and Cognitive Psychology, History and Philosophy of Science, Statistics and Probability

How researchers see this paper

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

In Atomic and Molecular Physics, and Optics

In Experimental and Cognitive Psychology

In History and Philosophy of Science

In Statistics and Probability

Citations in the corpus

Reframing Human-AI Collaboration for Generating Free-Text Explanations (2021, methodology): “Correlation between acceptability and other attributes To understand what factors are important for the overall “acceptability" judgement, we compute Spearman correlation (ρ; Spearman, 1987) between acceptability and all other attributes (Table 6).”
From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipeline (2024, neutral): “An outline of our data creation pipeline is shown in Figure 1.”
Bayesian Neural Word Embedding (2016, methodology): “ant improvement, when applying the similarities from Eqs. (17) and (18), instead of the cosine similarity. In order to compare between the BSG and SG methods, we compute for each method the Spearman (Spearman 1987) rank correlation coefficient with respect to the ground truth. The word analogy task ”
Interpretation of Neural Networks is Fragile (2017, methodology): “• Spearman’s rank order correlation: Because interpretation methods rank all of the features or training examples in order of importance, it is natural to use the rank correlation (Spearman 1904) to compare the similarity between interpretations.”
Unsupervised Learning of Sentence Embeddings Using Compositional n-Gram Features (2017, methodology): “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.”
Off-Policy Evaluation via Off-Policy Classification (2019, methodology): “We report two correlations, the coefﬁcient of determination R 2 of line of best ﬁt, and the Spearman rank correlation ⇠ [33].”
Measuring Model Biases in the Absence of Ground Truth (2021, methodology): “This approach is comparable to measuring correlations, although the common correlation metric of Spearman Rank [22] operates on assumptions that are not suitable for this task, such as linearity and monotonicity.”
Assessing Cross-Cultural Alignment between ChatGPT and Human Societies: An Empirical Study (2023, methodology): “Further, we utilize the Spearman correlation coefficient (Spearman, 1961) to assess the alignment between the cultural responses of ChatGPT and human societies.”
Emergent Ghosts of the Emotion Machine (2010, background): “It is also a product of classical test theory, which has been with us for some time now (Spearman, 1904a, 1904b).”
THE PdBI ARCSECOND WHIRLPOOL SURVEY (PAWS): ENVIRONMENTAL DEPENDENCE OF GIANT MOLECULAR CLOUD PROPERTIES IN M51 (2014, background): “A relation between size and velocity dispersion was identified in the early studies of Milky Way clouds (e.g., Solomon et al. 1979; Larson 1981; Dame et al. 1986).”
Item Response Theory—A Statistical Framework for Educational and Psychological Measurement (2021, neutral): “Classical test theory (CTT; Lord and Novick, 1968; Novick, 1966; Spearman, 1904b), also known as the true score theory, is another psychometric paradigm that was dominant before the prevalence of IRT.”
AutoLearn — Automated Feature Generation and Selection (2017, neutral): “• Pearson’s correlation [19] can only measure linear relationship whereas Spearmanś rho [20] and Kendallś tau [21] which both measure monotonic relationship, neither of them can capture all types of non-linear dependency between the two variables like distance correlation.”
Effective and General Evaluation for Instruction Conditioned Navigation using Dynamic Time Warping (2019, methodology): “Once scores are computed, we compare the rankings of each metric with the ones generated by humans, and calculate the average Spearman’s rank correlation [31].”
Judge the Judges: A Large-Scale Evaluation Study of Neural Language Models for Online Review Generation (2019, neutral): “To do this, we compute the correlations between H 1 , H 2 and each discriminative evaluator and correlations between H 1 , H 2 and the word-overlap evaluators, based on either their decisions on individual reviews, their scores of the generators (by Pearson’s coefﬁcient (Fieller et al., 1957)), and ”