Description Usage Arguments Details Value Note Author(s)
Converts most common test statistics into most common (signed) effect sizes.
1 2 3 4 5 |
df1 |
Degrees of freedom |
df2 |
NULL or degrees of freedom of the denominator for the f-distribution. |
N |
Number of data points used in calculation of test-statistic. |
n1 |
Number of data points in sample 1. |
n2 |
Number of data points in sample 2. |
esType |
Type of test statistic. One of: "t", "lm.t", "f", "lm.f", "r", "X2", "Z", "lm.Z" |
CIcalc |
If |
CL |
Confidence Limit (default: .95). |
rID |
Correlation among predictor values in a linear model. |
q |
Number of predictors in the model. |
alternative |
Alternative hypothesis (default = "two"). |
keepSign |
Return effect size with sign of test statistic? (default = TRUE). |
keepSignNames |
Which effect sizes should keep the sign if |
st |
Value(s) of a test statistic. |
The procedure to calculate a variety of effect sizes is as follows:
If CIcalc == FALSE
, package::compute.es
will be used to convert the test statistic to a large number of effect size estimates. The confidence intervals around the effect size estimates will be based meta-analytic estimates of effect size variance (e.g., for type - "t": tes).
If CIcalc == TRUE
, package::MBESS
will be used to calculate the confidence interval for the test statistic based on its noncentral distribution (e.g., for type - "t": conf.limits.nct). Subsequently the test statistic, as well as its lower and upper confidence limit will each be passed to compute.es
seperately.
If keepSign == TRUE
the sign of the test statistic will be copied to all the effect sizes in keepSignNames
.
The effect sizes calculated by compute.es
corresponding to the test statistic(s), with either meta-analytic, or, exact CI.
The prefix "lm" is currently disregarded, but will be implemented in future versions to indicate the test statistic is in fact a fixed factor in a linear model.
Fred Hasselman (inspired by RP:P function any2r
by CHJ Hartgerink)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.