P_eta_sq: Bayesian Estimation of "Partial Eta-Squared" in Fixed-Effects...
In rnorouzian/BayesianforL2: Introduction to Bayesian Thinking for L2 Researchers

Description Usage Arguments Value Author(s) References See Also Examples

Provides complete Posterior as well as its various summary statistics for "Partial Eta-Squared" in Fixed-Effects AN(C)OVA Designs.

1	P_eta_sq(f, N, df1, df2, alpha, beta)

`f`	Observed F-value from a main or an interaction effect.
`N`	Total design sample size.
`df1`	Degress of freedom for the effect in question aka "Numerator".
`df2`	Degrees of freedom "Error" aka "Denominator", "Within" or "Residual".
`alpha`	shape1 parameter of a "Beta Prior".
`beta`	shape2 parameter of a "Beta Prior".

Provides graphical as well as full textual description of posterior distribution of "Partial Eta-Squared".

Reza Norouzian <rnorouzian@gmail.com>

Norouzian, R. & Plonsky, L. (2017). Eta- and partial eta-squared in L2 research: A cautionary review and guide to more appropriate usage. Second Language Research, pp.1-15. doi: 10.1177/0267658316684904

Use function Beta_ID provided in this package to find an appropriate "Beta Prior" in your specific domain of resaerch.

Norouzian R., & L. Plonsky (2017) available at: http://journals.sagepub.com/doi/abs/10.1177/0267658316684904

This supplemetary document also briefly explains when Partial Eta Squared should be used: https://osf.io/7acye/

# P_eta_sq(f = 10, N = 60, df1 = 2, df2 = 57, alpha = 2.55, beta = 3.42)

# Note that, the "P_eta_sq(...)" when applied to ONE-WAY designs, is basically
# the same as "ETA-SQUARED" but when "P_eta_sq(...)" is applied to factorial
# designs it produces "PARTIAL ETA-SQUARED".

rnorouzian/BayesianforL2 documentation built on May 29, 2019, 8:37 a.m.