Clinical versus statistical prediction: A theoretical analysis and review of the evidence.

By E. L. Kelly. 1955

How researchers see this paper

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

Citations in the corpus

Fair Prediction with Disparate Impact: A Study of Bias in Recidivism Prediction Instruments (2016, background): “…we would like to note that there is a large body of literature showing that data-driven risk assessment instruments tend to be more accurate than professional human judgements [13, 14], and investigating whether human-driven decisions are themselves prone to exhibiting racial bias [15, 16].”
Evidence‐Based Assessment as an Integrative Model for Applying Psychological Science to Guide the Voyage of Treatment (2017, background): “Benchmarking the base rates for the most frequent problems is the preamble, creating a shortlist of hypotheses that will be worth considering precisely because they are commonplace (Meehl, 1954).”
When do people rely on algorithms (2016, background): “Theoretical Implications The results introduce an interesting contrast to the widespread assumption that people are averse to algorithms (Bazerman, 1985; Dawes, 1979; Dawes, Faust, & Meehl, 1989; Kleinmuntz, 1990; Kleinmuntz & Schkade, 1993; Meehl, 1954; Meehl, 1957).”
Confidence and Professional Judgment in Assessing Children’s Risk of Abuse (2010, background): “…that carefully validated actuarial models outperform clinical judgment in estimating future risk (Dawes, 1994; Grove & Meehl, 1996), clinical judgment is nevertheless understood to remain critical to the complex process of predicting risk (Meehl, 1954; Schwalbe, 2008; Van de Luitgaarden, 2009).”
Making sense of recommendations (2019, background): “Meehl, P. E. (1954). Clinical versus statistical prediction: A theoretical analysis and a review of the evidence. US: University of Minnesota Press. Nielsen Company (2015). Global trust in advertising survey.”
Forecasting: theory and practice (2020, background): “Studies examining individual’s trust in model versus expert forecasts show that individuals often distrust algorithms (Meehl, 2013; Burton et al., 2020) and place higher trust on human advice (Diab et al., 2011; Eastwood et al., 2012).”
Organizational Behavior and Human Decision Processes (2019, unknown): “The first scholarly reference to psychological distrust of algorithms may belong to Meehl (1954).”
The Threat of Algocracy: Reality, Resistance and Accommodation (2016, background): “It has already been proven that algorithmic systems are better at making predictions than human experts in certain fields (Bishop & Trout 2002; Meehl 1996; Grove and Meehl 1996).”
The replicability and utility of three personality types (2002, result): “We know that statistical prediction is superior to clinical prediction (Meehl, 1954), and the present results suggest that, using statistical methods, FFM factors are substantially better predictors than ARC types, but in the day-to-day work of the clinician it is possible that, because of evolved…”
Hybrid Intelligence (2019, background): “Even in very simple actuarial models, they outperform human experts in making predictions under uncertainty (Meehl 1954).”
A Five-Factor Theory Perspective (2002, background): “It is unclear how individuals make a composite judgment of a large body of people (most of whom they have never met), and we have long known that statistical predictions are more accurate than clinical judgments (Meehl, 1954) and do not necessarily agree with them.”
Heuristics and Biases: Clinical versus Actuarial Judgment (2002, unknown): “as this one” (Meehl, 1986, p. 373). Possible exceptions. If fair comparisons consistently favor the actuarial method, one may then reverse the impetus of inquiry and ask whether there are certain circumstances in which the clinical judge might beat the actuary. Might the clinician attain superiority”
Reasons, Values, Stakeholders: A Philosophical Framework for Explainable Artificial Intelligence (2021, background): “As another alternative, we could appeal to a reasoning style that incorporates the two reasoning approaches to decision making, both a statistically optimal one and an intuitive one, to arrive at a decision outcome [82, 83, 84].”
Geographic Profiling: The Fast, Frugal, and Accurate Way (2004, methodology): “The demonstration of human judgemental deficiencies has been seen by many as justification for the development of actuarial techniques (Edwards, 1972; Hastie & Dawes, 200l; Meehl, 1954).”
A case study of algorithm-assisted decision making in child maltreatment hotline screening decisions (2018, background): “A subject where there is greater consensus concerns the relative accuracy of so-called “clinical” versus “actuarial” judgment.1 Some of the earliest research comparing human predictions to those of statistical models goes back to the pioneering work of Meehl (1954).”
Creative Development as Acquired Expertise: Theoretical Issues and an Empirical Test☆ (2000, background): “Such trial-and-error learning would be rendered difficult by inherent limits to human information processing (Faust, 1984; Fiske & Taylor, 1991; Kahneman, Slovic, & Tversky, 1982; Meehl, 1954).”