THE RANDOMIZATION THEORY OF' EXPERIMENTAL INFERENCE*

By O. Kempthorne. 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

The Econometrics of Randomized Experiments (2016, background): “…variety of literatures, including the statistical literature on the analysis and design of experiments, e.g., Wu and Hamada (2009), Cox and Reid (2000), Altman (1991), Cook and DeMets (2008), Kempthorne (1952, 1955), Cochran and Cox (1957), Davies (1954), and Hinkelman and Kempthorne (2005, 2008).”
Permutation tests for multi-factorial analysis of variance (2003, background): “Exchangeability can be ensured through the random allocation of treatments to units in experimental design (e.g., Fisher, 1935; Kempthorne, 1955; Scheffé, 1959) or must be assumed for observational studies (e.g., Kempthorne, 1966).”
Covariance Adjustment in Randomized Experiments and Observational Studies (2002, background): “…are never models for data; rather, they are approximations to randomization distributions—they are good approximations to the extent that they reproduce randomization inferences with reduced computational effort (e.g., Welch, 1937; Wilk, 1955; Cox, 1956, 1958; Kempthorne, 1955; Robinson, 1973a, b).”
Panel experiments and dynamic causal effects: A finite population perspective (2020, methodology): “Our approach builds on the potential outcomes formulation of causal inference (Neyman, 1923; Kempthorne, 1955; Cox, 1958; Rubin, 1974) and takes a purely design-based perspective on uncertainty, allowing us to be agnostic to the outcomes model and avoid assumptions about hypothetical super-populatio”
The permutation testing approach: a review (2010, unknown): “(Kempthorne, 1955).”
Permutation tests for univariate or multivariate analysis of variance and regression (2001, background): “Kempthorne (1955) and Edgington (1995) have especially popularized Pitman’s idea that statistical inferences from randomization tests may not be regarded as extending beyond the sample itself, except when the extra assumption of random sampling from a population holds.”
Regression-adjusted average treatment effect estimates in stratified randomized experiments (2019, methodology): “above results were obtained under a randomization-based inference framework [19, 20, 36].”
Distance‐Based Tests for Homogeneity of Multivariate Dispersions (2006, background): “It has been suggested that the normal-theory test gains its validity by virtue of its agreement with the permutation test (Fisher, 1936; Kempthorne, 1955; Edgington, 1995).”
A paradox from randomization-based causal inference (2014, background): “Although rooted in randomization theory (Kempthorne, 1955; Hinkelmann and Kempthorne, 2007), the analysis of factorial experiments is dominated by linear and generalized linear models, with factorial effects often defined as model parameters.”
Rerandomization to improve covariate balance in experiments (2012, unknown): “However, randomization-based inference is still valid [Anscombe (1948a); Kempthorne (1955); Brillinger, Jones and Tukey (1978); Tukey (1993); Rosenberger and Lachin (2002); Moulton (2004)], because the rerandomization can be accounted for during analysis.”
Robust, accurate confidence intervals with a weak instrument: quarter of birth and education (2005, methodology): “In randomization inference, as developed by Fisher (1935) and later researchers (e.g. Pitman (1937), Welch (1937), Kempthorne (1955), Wilk (1955), Cox (1958), Robinson (1973), Tukey (1985), Gail et al. (1988) and Cox and Reid (2000)), a quantity whose value is determined by the random choice of P…”
Causal inference from 2K factorial designs by using potential outcomes (2015, methodology): “The first formal notation for potential outcomes was introduced by Neyman (1990) for randomization-based inference in randomized experiments and subsequently used by several researchers including Kempthorne (1955), Wilk (1955), Wilk and Kempthorne (1956, 1957) and Cox (1958) for causal inference fro”
Randomization in the Design of Experiments (2009, background): “In particular, the mean and the covariance matrix of the random variables determine the second-order properties of the design and, in the special case of randomized blocks, leads to equations (4) and (5) (Kempthorne, 1952, 1955; Hinkelmann & Kempthorne, 2008).”