Description

Functions to compute effect size measures for ANOVAs, such as Eta, Omega and Epsilon squared, and Cohen's f (or their partialled versions) for aov, aovlist and anova models. These indices represent an estimate of how much variance in the response variables is accounted for by the explanatory variable(s).

Effect sizes are computed using the sums of squares obtained from anova(model) which might not always be appropriate (Yeah... ANOVAs are hard...). It is suggested that ANOVA models be fit with afex package. See details.

Usage

Arguments

Details

For aov and aovlist models, the effect sizes are computed directly with Sums-of-Squares (for mlm / maov models, effect sizes are computed for each response separately). For all other model, the model is passed to anova(), and effect sizes are approximated via test statistic conversion (see F_to_eta2() for more details.)

Type of Sums of Squares

The sums of squares (or F statistics) used for the computation of the effect sizes is based on those returned by anova(model) (whatever those may be - for aov and aovlist these are type-1 sums of squares; for merMod these are type-3 sums of squares). Make sure these are the sums of squares you are interested in; You might want to pass the result of car::Anova(mode, type = 3), or use the afex package to fit ANOVA models.

It is generally recommended to fit models with contr.sum factor weights and centered covariates, for sensible results. See examples and the afex package.

Un-Biased Estimate of Eta

Both Omega and Epsilon are unbiased estimators of the population's Eta, which is especially important is small samples. But which to choose?

Though Omega is the more popular choice (Albers \& Lakens, 2018), Epsilon is analogous to adjusted R2 (Allen, 2017, p. 382), and has been found to be less biased (Carroll & Nordholm, 1975).

(Note that for ω_p^2 and ε_p^2 it is possible to compute a negative number; even though this doesn't make any practical sense, it is recommended to report the negative number and not a 0.)

Cohen's f

Cohen's f can take on values between zero, when the population means are all equal, and an indefinitely large number as standard deviation of means increases relative to the average standard deviation within each group.

When comparing two models in a sequential regression analysis, Cohen's f for R-square change is the ratio between the increase in R-square and the percent of unexplained variance.

Cohen has suggested that the values of 0.10, 0.25, and 0.40 represent small, medium, and large effect sizes, respectively.

Eta Squared from Posterior Predictive Distribution

For Bayesian models (fit with brms or rstanarm), eta_squared_posterior() simulates data from the posterior predictive distribution (ppd) and for each simulation the Eta Squared is computed for the model's fixed effects. This means that the returned values are the population level effect size as implied by the posterior model (and not the effect size in the sample data). See rstantools::posterior_predict() for more info.

Value

A data frame with the effect size(s) between 0-1 (Eta2, Epsilon2, Omega2, Cohens_f or Cohens_f2, possibly with the partial or generalized suffix), and their CIs (CI_low and CI_high).

For eta_squared_posterior(), a data frame containing the ppd of the Eta squared for each fixed effect, which can then be passed to bayestestR::describe_posterior() for summary stats.

A data frame containing the effect size values and their confidence intervals.

Confidence Intervals

Unless stated otherwise, confidence intervals are estimated using the Noncentrality parameter method; These methods searches for a the best non-central parameters (ncps) of the noncentral t-, F- or Chi-squared distribution for the desired tail-probabilities, and then convert these ncps to the corresponding effect sizes. (See full effectsize-CIs for more.)

CI Contains Zero

For positive only effect sizes (Eta squared, Cramer's V, etc.; Effect sizes associated with Chi-squared and F distributions), special care should be taken when interpreting CIs with a lower bound equal to 0, and even more care should be taken when the upper bound is equal to 0 (Steiger, 2004; Morey et al., 2016). For example:

References

• Albers, C., \& Lakens, D. (2018). When power analyses based on pilot data are biased: Inaccurate effect size estimators and follow-up bias. Journal of experimental social psychology, 74, 187-195.

• Allen, R. (2017). Statistics and Experimental Design for Psychologists: A Model Comparison Approach. World Scientific Publishing Company.

• Carroll, R. M., & Nordholm, L. A. (1975). Sampling Characteristics of Kelley's epsilon and Hays' omega. Educational and Psychological Measurement, 35(3), 541-554.

• Kelley, T. (1935) An unbiased correlation ratio measure. Proceedings of the National Academy of Sciences. 21(9). 554-559.

• Olejnik, S., & Algina, J. (2003). Generalized eta and omega squared statistics: measures of effect size for some common research designs. Psychological methods, 8(4), 434.

• Steiger, J. H. (2004). Beyond the F test: Effect size confidence intervals and tests of close fit in the analysis of variance and contrast analysis. Psychological Methods, 9, 164-182.

See Also

F_to_eta2()

Other effect size indices: cohens_d(), effectsize(), phi(), rank_biserial(), standardize_parameters()

Examples