Module stikpetP.effect_sizes.eff_size_cohen_f
Expand source code
from ..effect_sizes.eff_size_eta_sq import es_eta_sq
def es_cohen_f(nomField, scaleField, categories=None, useRanks=False):
'''
Cohen f
-------
An effect size measure for regression analysis or an ANOVA test. It gives roughly the proportion of variance explained by the categorical variable.
The Cohen f is often used with ANOVA, while Cohen f-squared with regression.
Parameters
----------
nomField : pandas series
data with categories
scaleField : pandas series
data with the scores
categories : list or dictionary, optional
the categories to use from catField
useRanks : boolean, optional
Use of ranks or original scores. Default is False
Returns
-------
f : float
the Cohen f value
Notes
-----
The formula used (Cohen, 1988, p. 284):
$$f = \\sqrt{\\frac{\\eta^2}{1-\\eta^2}}$$
Where \\(\\eta^2\\) is the value of eta-squared.
It can also be calculated using (Cohen, 1988, p. 371):
$$f = \\frac{\\sigma_{\\mu}}{\\sigma}$$
With:
$$\\sigma_{\\mu} = \\sqrt{\\frac{SS_b}{n}}$$
$$\\sigma = \\sqrt{\\frac{SS_w}{n}}$$
Where \\(SS_i\\) is the sum of squared differences, see the Fisher one-way ANOVA for details on how to calculate these.
The \\(f^2\\) can be found in Cohen (1988, p. 410).
References
----------
Cohen, J. (1988). *Statistical power analysis for the behavioral sciences* (2nd ed.). L. Erlbaum Associates.
Author
------
Made by P. Stikker
Companion website: https://PeterStatistics.com
YouTube channel: https://www.youtube.com/stikpet
Donations: https://www.patreon.com/bePatron?u=19398076
'''
eta2 = es_eta_sq(nomField, scaleField, categories, useRanks=useRanks)
f2 = eta2 / (1 - eta2)
return (f2)**0.5Functions
def es_cohen_f(nomField, scaleField, categories=None, useRanks=False)-
Cohen F
An effect size measure for regression analysis or an ANOVA test. It gives roughly the proportion of variance explained by the categorical variable.
The Cohen f is often used with ANOVA, while Cohen f-squared with regression.
Parameters
nomField:pandas series- data with categories
scaleField:pandas series- data with the scores
categories:listordictionary, optional- the categories to use from catField
useRanks:boolean, optional- Use of ranks or original scores. Default is False
Returns
f:float- the Cohen f value
Notes
The formula used (Cohen, 1988, p. 284): f = \sqrt{\frac{\eta^2}{1-\eta^2}}
Where \eta^2 is the value of eta-squared.
It can also be calculated using (Cohen, 1988, p. 371): f = \frac{\sigma_{\mu}}{\sigma} With: \sigma_{\mu} = \sqrt{\frac{SS_b}{n}} \sigma = \sqrt{\frac{SS_w}{n}}
Where SS_i is the sum of squared differences, see the Fisher one-way ANOVA for details on how to calculate these.
The f^2 can be found in Cohen (1988, p. 410).
References
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). L. Erlbaum Associates.
Author
Made by P. Stikker
Companion website: https://PeterStatistics.com
YouTube channel: https://www.youtube.com/stikpet
Donations: https://www.patreon.com/bePatron?u=19398076Expand source code
def es_cohen_f(nomField, scaleField, categories=None, useRanks=False): ''' Cohen f ------- An effect size measure for regression analysis or an ANOVA test. It gives roughly the proportion of variance explained by the categorical variable. The Cohen f is often used with ANOVA, while Cohen f-squared with regression. Parameters ---------- nomField : pandas series data with categories scaleField : pandas series data with the scores categories : list or dictionary, optional the categories to use from catField useRanks : boolean, optional Use of ranks or original scores. Default is False Returns ------- f : float the Cohen f value Notes ----- The formula used (Cohen, 1988, p. 284): $$f = \\sqrt{\\frac{\\eta^2}{1-\\eta^2}}$$ Where \\(\\eta^2\\) is the value of eta-squared. It can also be calculated using (Cohen, 1988, p. 371): $$f = \\frac{\\sigma_{\\mu}}{\\sigma}$$ With: $$\\sigma_{\\mu} = \\sqrt{\\frac{SS_b}{n}}$$ $$\\sigma = \\sqrt{\\frac{SS_w}{n}}$$ Where \\(SS_i\\) is the sum of squared differences, see the Fisher one-way ANOVA for details on how to calculate these. The \\(f^2\\) can be found in Cohen (1988, p. 410). References ---------- Cohen, J. (1988). *Statistical power analysis for the behavioral sciences* (2nd ed.). L. Erlbaum Associates. Author ------ Made by P. Stikker Companion website: https://PeterStatistics.com YouTube channel: https://www.youtube.com/stikpet Donations: https://www.patreon.com/bePatron?u=19398076 ''' eta2 = es_eta_sq(nomField, scaleField, categories, useRanks=useRanks) f2 = eta2 / (1 - eta2) return (f2)**0.5