The criticalESvalue package provide a set of functions to calculate
the critical effects size value for the common statistical model.
Currently, the package support the htest class (i.e., t.test and
cor.test), the lm class and the rma class from the metafor
package.
The package contains a set of functions to work with summary data and
the critical() method that takes objects of class htest, lm or
rma and provide enhanced printing and summary methods with information
about critical effects size.
You can install the development version of criticalESvalue like so:
# require(remotes)
remotes::install_github("psicostat/criticalESvalue")# loading the package
library(criticalESvalue)# t-test (welch)
x <- rnorm(30, 0.5, 1)
y <- rnorm(30, 0, 1)
ttest <- t.test(x, y)
critical(ttest)
#>
#> Welch Two Sample t-test
#>
#> data: x and y
#> t = 1.8, df = 57.7, p-value = 0.077
#> alternative hypothesis: true difference in means is not equal to 0
#> 95 percent confidence interval:
#> -0.055794 1.065374
#> sample estimates:
#> mean of x mean of y
#> 0.36937 -0.13542
#>
#> |== Effect Size and Critical Value ==|
#> d = 0.46545 dc = ± 0.51689 bc = ± 0.56058
#> g = 0.45937 gc = ± 0.51014
# t-test (standard)
ttest <- t.test(x, y, var.equal = TRUE)
critical(ttest)
#>
#> Two Sample t-test
#>
#> data: x and y
#> t = 1.8, df = 58, p-value = 0.077
#> alternative hypothesis: true difference in means is not equal to 0
#> 95 percent confidence interval:
#> -0.055739 1.065319
#> sample estimates:
#> mean of x mean of y
#> 0.36937 -0.13542
#>
#> |== Effect Size and Critical Value ==|
#> d = 0.46545 dc = ± 0.51684 bc = ± 0.56053
#> g = 0.4594 gc = ± 0.51012
# t-test (standard) with monodirectional hyp
ttest <- t.test(x, y, var.equal = TRUE, alternative = "less")
critical(ttest)
#>
#> Two Sample t-test
#>
#> data: x and y
#> t = 1.8, df = 58, p-value = 0.96
#> alternative hypothesis: true difference in means is less than 0
#> 95 percent confidence interval:
#> -Inf 0.97286
#> sample estimates:
#> mean of x mean of y
#> 0.36937 -0.13542
#>
#> |== Effect Size and Critical Value ==|
#> d = 0.46545 dc = -0.43159 bc = -0.46807
#> g = 0.4594 gc = -0.42598
# within the t-test object saved from critical we have all the new values
ttest <- critical(ttest)
str(ttest)
#> List of 15
#> $ statistic : Named num 1.8
#> ..- attr(*, "names")= chr "t"
#> $ parameter : Named num 58
#> ..- attr(*, "names")= chr "df"
#> $ p.value : num 0.962
#> $ conf.int : num [1:2] -Inf 0.973
#> ..- attr(*, "conf.level")= num 0.95
#> $ estimate : Named num [1:2] 0.369 -0.135
#> ..- attr(*, "names")= chr [1:2] "mean of x" "mean of y"
#> $ null.value : Named num 0
#> ..- attr(*, "names")= chr "difference in means"
#> $ stderr : num 0.28
#> $ alternative: chr "less"
#> $ method : chr " Two Sample t-test"
#> $ data.name : chr "x and y"
#> $ g : num 0.459
#> $ gc : num 0.426
#> $ d : num 0.465
#> $ bc : num 0.468
#> $ dc : num 0.432
#> - attr(*, "class")= chr [1:3] "critvalue" "ttest" "htest"We can check the results using:
ttest <- t.test(x, y)
ttest <- critical(ttest)
t <- ttest$bc/ttest$stderr # critical numerator / standard error
# should be 0.05 (or alpha)
(1 - pt(ttest$bc/ttest$stderr, ttest$parameter)) * 2
#> [1] 0.05# cor.test
ctest <- cor.test(x, y)
critical(ctest)
#>
#> Pearson's product-moment correlation
#>
#> data: x and y
#> t = 0.257, df = 28, p-value = 0.8
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> -0.31735 0.40172
#> sample estimates:
#> cor
#> 0.048461
#>
#> |== Critical Value ==|
#> |rc| = 0.36101# linear model (unstandardized)
z <- rnorm(30)
q <- rnorm(30)
dat <- data.frame(x, y, z, q)
fit <- lm(y ~ x + q + z, data = dat)
fit <- critical(fit)
fit
#>
#> Call:
#> lm(formula = y ~ x + q + z, data = dat)
#>
#> Coefficients:
#> (Intercept) x q z
#> -0.11680 0.06495 -0.13635 0.00332
#>
#>
#> Critical |Coefficients|
#>
#> (Intercept) x q z
#> 0.48599 0.43970 0.50174 0.39505
summary(fit)
#>
#> Call:
#> lm(formula = y ~ x + q + z, data = dat)
#>
#> Residuals:
#> Min 1Q Median 3Q Max
#> -1.8786 -0.9037 -0.0576 0.8473 2.3993
#>
#> Coefficients:
#> Estimate |Critical Estimate| Std. Error t value Pr(>|t|)
#> (Intercept) -0.11680 0.48599 0.23643 -0.49 0.63
#> x 0.06495 0.43970 0.21391 0.30 0.76
#> q -0.13635 0.50174 0.24409 -0.56 0.58
#> z 0.00332 0.39505 0.19219 0.02 0.99
#>
#> Residual standard error: 1.17 on 26 degrees of freedom
#> Multiple R-squared: 0.0162, Adjusted R-squared: -0.0973
#> F-statistic: 0.143 on 3 and 26 DF, p-value: 0.933library(metafor)
#> Loading required package: Matrix
#> Loading required package: metadat
#> Loading required package: numDeriv
#>
#> Loading the 'metafor' package (version 4.6-0). For an
#> introduction to the package please type: help(metafor)
dat <- escalc(measure="RR", ai=tpos, bi=tneg, ci=cpos, di=cneg, data=dat.bcg)
fit <- rma(yi, vi, mods=cbind(ablat, year), data=dat)
critical(fit)
#>
#> Mixed-Effects Model (k = 13; tau^2 estimator: REML)
#>
#> tau^2 (estimated amount of residual heterogeneity): 0.1108 (SE = 0.0845)
#> tau (square root of estimated tau^2 value): 0.3328
#> I^2 (residual heterogeneity / unaccounted variability): 71.98%
#> H^2 (unaccounted variability / sampling variability): 3.57
#> R^2 (amount of heterogeneity accounted for): 64.63%
#>
#> Test for Residual Heterogeneity:
#> QE(df = 10) = 28.3251, p-val = 0.0016
#>
#> Test of Moderators (coefficients 2:3):
#> QM(df = 2) = 12.2043, p-val = 0.0022
#>
#> Model Results:
#>
#> estimate se zval pval ci.lb ci.ub
#> intrcpt -3.5455 29.0959 -0.1219 0.9030 -60.5724 53.4814
#> ablat -0.0280 0.0102 -2.7371 0.0062 -0.0481 -0.0080 **
#> year 0.0019 0.0147 0.1299 0.8966 -0.0269 0.0307
#>
#> ---
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#>
#> |critical estimate|
#> intrcpt 62.85783
#> ablat 0.02211
#> year 0.03172x <- rnorm(30, 0.5, 1)
y <- rnorm(30, 0, 1)
ttest <- t.test(x, y)
m1 <- mean(x)
m2 <- mean(y)
sd1 <- sd(x)
sd2 <- sd(y)
n1 <- n2 <- 30
critical_t2s(m1, m2, sd1 = sd1, sd2 = sd2, n1 = n1, n2 = n2)
#> $d
#> [1] -0.023955
#>
#> $dc
#> [1] 0.51721
#>
#> $bc
#> [1] 0.52285
#>
#> $se
#> [1] 0.26102
#>
#> $df
#> [1] 56.144
#>
#> $g
#> [1] -0.023633
#>
#> $gc
#> [1] 0.51026
critical_t2s(t = ttest$statistic, se = ttest$stderr, n1 = n1, n2 = n2)
#> Warning in crit_from_t_t2s(t = t, n1 = n1, n2 = n2, se = se, conf.level =
#> conf.level, : When var.equal = FALSE the critical value calculated from t
#> assume sd1 = sd2!
#> $d
#> [1] -0.023955
#>
#> $dc
#> [1] 0.51684
#>
#> $bc
#> [1] 0.52248
#>
#> $se
#> [1] 0.26102
#>
#> $df
#> [1] 58
#>
#> $g
#> [1] -0.023643
#>
#> $gc
#> [1] 0.51012
critical_t2s(t = ttest$statistic, n1 = n1, n2 = n2)
#> Warning in crit_from_t_t2s(t = t, n1 = n1, n2 = n2, se = se, conf.level =
#> conf.level, : When var.equal = FALSE the critical value calculated from t
#> assume sd1 = sd2!
#> Warning in crit_from_t_t2s(t = t, n1 = n1, n2 = n2, se = se, conf.level =
#> conf.level, : When se = NULL bc cannot be computed, returning NA!
#> $d
#> [1] -0.023955
#>
#> $dc
#> [1] 0.51684
#>
#> $bc
#> [1] NA
#>
#> $se
#> NULL
#>
#> $df
#> [1] 58
#>
#> $g
#> [1] -0.023643
#>
#> $gc
#> [1] 0.51012