Statistical Rules of Thumb

By L. S. Nelson. 2003

416 citations recorded by the source corpus.

Subfields: Geometry and Topology, Signal Processing

View Paper: https://www.semanticscholar.org/paper/e6f138b898f5e872a512dafe8a92900ff4819f5f

How researchers see this paper

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

In Geometry and Topology

In Signal Processing

Citations in the corpus

Are there any consistent predictors of invasion success? (2008, background): “This last recommendation is tempered by the typical data limitations (ten observations per variable is a useful rule of thumb—see van Belle 2002 ), and the need to provide an easily interpretable output.”
Practical guide to controlled experiments on the web: listen to your customers not to the hippo (2007, background): “Our experience is that strong interactions are rare in practice (33), and we believe this concern is overrated.”
IMPROVING THE PRECISION OF ESTIMATES OF THE FREQUENCY OF RARE EVENTS (2005, background): “If the event is very rare (,10 events per covariate incorporated in the logistic regression; Van Belle 2002), logistic regression may not be useful for modeling the probability of very rare events.”
The problem of pseudoreplication in neuroscientific studies: is it affecting your analysis? (2010, background): “Thus when there is a positive correlation, null hypotheses will be rejected too often, and this is the reason that violating the independence assumption can be more serious than violating the normality or equal variances assumption [8].”
Differential Privacy for Clinical Trial Data: Preliminary Evaluations (2009, methodology): “Our general modifications of the algorithm from [11] are given in Algorithm 1.”
The impact of Levene’s test of equality of variances on statistical theory and practice (2009, background): “It is well known (van Belle, 2002) that dependence can have a major effect on the distribution of many standard statistics.”
Initial Scale Development: Sample Size for Pilot Studies (2010, background): “For similar reasons, Hill (1998) suggested 10 to 30 participants for pilots in survey research. van Belle (2002) suggested that researchers “use at least 12 observations in constructing a confidence interval” (p. 11).”