Significance, Effect Sizes, Stepwise Methods, and Other Issues: Strong Arguments Move the Field

By B. Thompson. 2001

240 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/ae58219d3c4dcbc0b78571a4c3f83af7684d03da

How researchers see this paper

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

In Artificial Intelligence

In Statistics and Probability

Citations in the corpus

Utilising learning environment assessments to improve teaching practices among in-service teachers undertaking a distance-education programme (2009, methodology): “To examine the magnitudes of these actual-preferred differences, as well as their statistical significance (as recommended by Thompson, 1998; 2001), effect sizes were calculated in terms of the differences in means divided by the pooled standard deviation.”
Using Lasso for Predictor Selection and to Assuage Overfitting: A Method Long Overlooked in Behavioral Sciences (2015, methodology): “…in regression models makes the assumption that the predictors are fixed entities, but methods like stepwise regression repeatedly add and subtract predictors to the model without accounting for this adaptive nature of the model (Lockhart, Taylor, Tibshirani, & Tibshirani, 2014; Thompson, 2001).”
STUDY QUALITY IN SLA (2013, unknown): “…testing (NHST), a practice that has been condemned for decades by quantitative methodologists (e.g., Cohen, 1994 ; Lykken, 1968 ; Schmidt, 1996 ; Thompson, 2001 ), cautioned against sporadically by L2 researchers (Crookes, 1991 ; Larson-Hall, 2010 ; Lazaraton, 1991 ; Nassaji, 2012 ; Norris &…”
A History of Effect Size Indices (2002, methodology): “Furthermore, as Thompson (2001) recently noted regarding Cohen’s criteria, “If people interpreted effect sizes with the same rigidity that α = .05 has been used in statistical testing, we would merely be being stupid in another metric” (pp. 82-83).”
Un sistema de clasificación de los diseños de investigación en psicología (2013, background): “Se han esgrimido muchas razones para preferir la selección jerárquica a la selección paso a paso (Snyder, 1991, Thompson, 1995, 2001), la principal de las cuales es que con la selección jerárquica es posible incorporar en las primeras etapas del proceso las variables extrañas que el investigador…”
Effect size, confidence interval and statistical significance: a practical guide for biologists (2007, methodology): “We emphasize the point made by Thompson (2001) who stated that if we use these fixed benchmarks with the same rigidity that a ¼ 0.05 is used in NHST, we are just being stupid in another metric.”
How to Estimate and Interpret Various Effect Sizes (2004, background): “However, empirical studies of reporting practices indicate that in a surprising number of studies, authors report effects without clearly indicating exactly which effect is being presented (Kieffer, Reese, & Thompson, 2001; Vacha-Haase et al., 2000).”
How Big Is “Big”? Interpreting Effect Sizes in L2 Research (2014, background): “The academic controversy surrounding statistical significance, where practical significance is often offered as an important alternative or supplement, dates back to the first half of the 20th century (see Cohen, 1994; Thompson, 2001) and includes occasional critiques of p values within L2 research ”
You Think You’ve Got Trivials? (2003, background): “Their point is well taken, despite the apparent recanting of this form of persuasion by Thompson (2002), who cautioned “headcounts of views are not perfect indicators of truth” (p. 85).”
Evaluating the Quality of Evidence from Correlational Research for Evidence-Based Practice (2005, methodology): “Confidence intervals can be computed for (a) reliability coefficients (Fan & Thompson, 2001); (b) sample statistics (e.g., M, r); and (c) effect sizes (Thompson, 2002b).”
Effect-Size Measures and Meta-Analytic Thinking in Counseling Psychology Research (2006, methodology): “In such a case, it would be important to examine both standardized weights and structure coefficients to evaluate each variable’s contribution to the effect size (Courville & Thompson, 2001; Henson, 2002).”