forked from katherinemwood/lakens_effect_sizes
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathapp.R
275 lines (266 loc) · 15.4 KB
/
app.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
library(shiny)
library(shinythemes)
ui <- fluidPage(theme=shinytheme('flatly'),
titlePanel('Calculating Effect Sizes'),
tabsetPanel(type='tabs', id='tabset',
tabPanel('t Tests (Independent Samples)', value='indep',
h3('Results', align='center'),
fluidRow(
column(12, align='center',
tableOutput("ti")
)
),
br(),
h4('Enter whatever information you have',
'about the samples.'),
wellPanel(fluidRow(
column(3,
numericInput('n1', 'Group 1 Sample Size', 10)),
column(3,
numericInput('n2', 'Group 2 Sample Size', 10)),
column(3,
numericInput('nt', 'Total N', NA)),
column(3,
numericInput('itval', 't Value', 2.5175, step=.1))
),
fluidRow(
column(3,
numericInput('im1', 'Mean for Group 1', NA)),
column(3,
numericInput('im2', 'Mean for Group 2', NA)),
column(3,
numericInput('isd1', 'SD for Group 1', NA)),
column(3,
numericInput('isd2', 'SD for Group 2', NA))
),
fluidRow(align='center',
div(style="display:inline-block; padding-right:10px", actionButton('calc_ti', 'Calculate',
style="color: #fff; background-color: #ff5656; border-color: #ff5656")),
div(style="display:inline-block",actionButton('r_ti', 'Clear',
style="color: #fff; background-color: #337ab7; border-color: #337ab7"))
)
)
),
tabPanel('t Tests (Dependent or Correlated Samples)', value='dep',
h3('Results', align='center'),
fluidRow(
column(12, align='center',
tableOutput("td")
)
),
br(),
h4('Enter whatever information you have',
'about the samples.'),
wellPanel(
fluidRow(
column(3,
numericInput('pairs', 'Number of Pairs', 10)),
column(3,
numericInput('dm1', 'Mean for Group 1', NA)),
column(3,
numericInput('dm2', 'Mean for Group 2', NA)),
column(3,
numericInput('r', 'Correlation between Groups', NA))
),
fluidRow(
column(3,
numericInput('dtval', 'T Value', 4.7, step=.1)),
column(3,
numericInput('dsd1', 'SD for Group 1', NA)),
column(3,
numericInput('dsd2', 'SD for Group 2', NA)),
column(3, offset=.5, align='bottom',
div(style="display:inline-block; padding-top:25px; padding-right:10px",
actionButton('calc_td', 'Calculate',
style="color: #fff; background-color: #ff5656; border-color: #ff5656")),
div(style="display:inline-block; padding-top:25px",
actionButton('r_td', 'Clear',
style="color: #fff; background-color: #337ab7; border-color: #337ab7")))
)
)
),
tabPanel('F Tests',
sidebarLayout(
sidebarPanel(
id='ftests',
numericInput('fstat', 'F Statistic', 6.34, step=.1),
numericInput('dfh', 'Treatment Degrees of Freedom', 1),
numericInput('dfe', 'Residual Degrees of Freedom', 18),
div(style="display:inline-block;padding-right:10px", actionButton('calc_f', 'Calculate',
style="color: #fff; background-color: #ff5656; border-color: #ff5656;")),
div(style="display:inline-block;padding-right:10px", actionButton('r_f', 'Clear',
style="color: #fff; background-color: #337ab7; border-color: #337ab7"))
),
mainPanel(align='center',
h3('Results'),
tableOutput("f_es")
)
)
),
tabPanel('Correlations',
sidebarLayout(
sidebarPanel(
h5("Enter two of three values to calculate the third."),
numericInput('r1', 'Correlation 1', .5, step=.1),
numericInput('r2', 'Correlation 2', .42, step=.1),
numericInput('q', "Cohen's q", NA, step=.1),
div(style="display:inline-block;padding-right:10px", actionButton('calc_q', 'Calculate',
style="color: #fff; background-color: #ff5656; border-color: #ff5656;")),
div(style="display:inline-block;padding-right:10px", actionButton('r_q', 'Clear',
style="color: #fff; background-color: #337ab7; border-color: #337ab7"))
),
mainPanel(align='center',
h3('Results'),
tableOutput("c_q")
)
)
)
),
hr(),
fluidRow(
column(width=11, offset=.5, h5("This is adapted from Daniel Lakens's original spreadsheets.",
"To cite, please see:",
a('Lakens, 2013.',
href='http://journal.frontiersin.org/article/10.3389/fpsyg.2013.00863/full'),
'and the accompanying', a('OSF page.', href='http://openscienceframework.org/project/ixGcd',
target="_blank"))),
column(width=11, offset=.5, h5('Application created by Katherine Wood', a('(@kathmwood)', href="https://twitter.com/kathmwood",
target="_blank")))
)
)
server <- function(input, output, session) {
get_ti <- function(n1, n2, nt, im1, im2, isd1, isd2, itval) {
if (sum(!is.na(c(n1, n2))) == 0 & is.na(nt)){
print('Sample size required.')
}
if (sum(is.na(c(im1, im2, isd1, isd2, n1, n2))) == 0){
df <- n1 + n2 - 2
mdiff <- abs(im1 - im2)
sdiff <- sqrt((((n1-1)*isd1^2) + ((n2-1)*isd2^2))/df)
sediff <- sqrt(((isd1^2)/n1) + ((isd2^2)/n2))
t <- mdiff/(sdiff*sqrt((1/n1) + (1/n2)))
p <- pt(t, n1+n2-2, lower.tail=FALSE)*2
cl <- pnorm(abs(mdiff/sqrt(isd1^2+isd2^2)))
ds <- mdiff/sqrt((((n1-1)*isd1^2) + ((n2-1)*isd2^2))/df)
d <- mdiff/sqrt((((n1-1)*isd1^2) + ((n2-1)*isd2^2))/(n1+n2))
hgs <- ds*(1-(3/(4*df - 1)))
ci_width <- qt(.025, df, lower.tail=FALSE)*sediff
td <- data.frame('Mean Difference'=mdiff, '95% CI (lower)'=mdiff-ci_width,
'95% CI (upper)'=mdiff+ci_width, '\U1D461 Value'=t, '\U1D45D Value'=p,
"Cohen's ds"=ds, "Cohen's d"=d, 'Hedges gs'=hgs,
'CL Effect Size'=cl, check.names=FALSE)
} else if (sum(is.na(c(n1, n2, itval))) == 0) {
ds <- (itval*sqrt(n1+n2))/sqrt(n1*n2)
data.frame("Cohen's ds"=ds,
'\U1D45D Value' = pt(itval, n1+n2-2, lower.tail=FALSE)*2,
'Hedges gs'=ds*(1-(3/(4*(n1+n2) - 9))),
'CL Effect Size'=pnorm(ds/sqrt(2)), check.names=FALSE)
} else if (sum(is.na(c(nt, itval))) == 0) {
ds <- (2*itval)/sqrt(nt)
data.frame("Cohen's ds"=ds,
'\U1D45D value' = pt(itval, nt-2, lower.tail=FALSE)*2,
'Hedges gs'=ds*(1-(3/(4*nt - 9))),
'CL Effect Size'=pnorm(ds/sqrt(2)), check.names=FALSE)
} else {
print('Sorry, not enough information.')
}
}
get_td <- function(pairs, dm1, dm2, dsd1, dsd2, r, dtval) {
if (is.na(pairs)){
print('Sample size required.')
}
if (sum(is.na(c(r, dm1, dm2, dsd1, dsd2, pairs))) == 0){
df <- pairs - 1
mdiff <- abs(dm1 - dm2)
sdiff <- sqrt(dsd1^2 + dsd2^2 - (2*r*dsd1*dsd2))
sediff <- sqrt(((dsd1^2)/pairs) + ((dsd2^2)/pairs) - (2*r*(dsd1/sqrt(pairs))*(dsd2/sqrt(pairs))))
t <- mdiff/(sdiff/sqrt(pairs))
p <- pt(t, pairs-1, lower.tail=FALSE)*2
cl <- pnorm(mdiff/sdiff)
ci_width <- qt(.025, df, lower.tail=FALSE)*sediff
dz <- mdiff/sdiff
drm <- dz*sqrt(2*(1-r))
dav <- mdiff/sqrt((dsd1^2+dsd2^2)/2)
hgav <- dav*(1 - (3/(4*df - 1)))
hgrm <- drm*(1 - (3/(4*df - 1)))
corr_factor <- mdiff/sqrt((df*(dsd1^2 + dsd2^2))/2*df)
rec <- ifelse(abs(corr_factor - drm) < abs(corr_factor - dav),
'Hedges grm', 'Hedges gav')
data.frame('Mean difference'=mdiff, '95% CI (lower)'=mdiff-ci_width,
'95% CI (upper)'=mdiff+ci_width, '\U1D461 value'=t,
'\U1D45D value'=p, "Cohen's dz"=dz, "Cohen's drm"=drm, 'Hedges grm'=hgrm,
"Cohen's dav"=dav, 'Hedges gav'=hgav, 'Rec. ES'=rec,
'CL Effect Size'=cl,check.names=FALSE)
} else if (sum(is.na(c(dtval, pairs))) == 0){
dz <- dtval/sqrt(pairs)
data.frame('\U1D45D value'=pt(dtval, pairs-1, lower.tail=FALSE)*2,
"Cohen's dz"=dz, 'CL Effect Size' = pnorm(dz), check.names=FALSE)
} else {
print('Sorry, not enough information.')
}
}
get_f <- function(fstat, dfh, dfe) {
data.frame('Partial \U03B7\U00B2'=(fstat*dfh)/(fstat*dfh + dfe),
'Partial \U03C9\U00B2'=(fstat-1)/(fstat+((dfe+1)/dfh)),
'Partial \U03B5\U00B2'=(fstat-1)/(fstat+(dfe/dfh)),
'\U1D45D value'=pf(fstat, dfh, dfe, lower.tail=FALSE), check.names=FALSE)
}
get_q <- function(r1, r2, q) {
corrs <- c(r1, r2)
if (is.na(q) | sum(is.na(c(r1, r2, q))) == 0) {
q <- .5*log((1+r1)/(1-r1)) - .5*log((1+r2)/(1-r2))
} else {
r1 <- corrs[!is.na(corrs)]
r2 <- (exp(2*q+log((1+r1)/(1-r1))) - 1)/(exp(2*q+log((1+r1)/(1-r1))) + 1)
}
data.frame('r\U2081'=r1, 'r\U2082'=r2, "Cohen's \U1D45E"=q, check.names=FALSE)
}
reset <- function(session, fields, value=NA) {
sapply(fields, function(field) updateNumericInput(session, field, value=value))
}
dataframes <- reactiveValues(c_q=NULL, ti=NULL, td=NULL, f_es=NULL,
lcq=0, lcti=0, lctd=0, lcf=0,
rcq=0, rcti=0, rctd=0, rcf=0)
observe({
if (input$calc_ti > dataframes$lcti | input$calc_ti==0) {
dataframes$lcti <- input$calc_ti
dataframes$ti <- isolate(get_ti(input$n1, input$n2, input$nt, input$im1, input$im2,
input$isd1, input$isd2, input$itval))
}
if (input$r_ti > dataframes$rcti) {
dataframes$rcti <- input$r_ti
reset(session, c('n1', 'n2', 'nt', 'itval', 'im1', 'im2', 'isd1', 'isd2'))
}
if (input$calc_q > dataframes$lcq | input$calc_q == 0) {
dataframes$lcq <- input$calc_q
dataframes$c_q <- isolate(get_q(input$r1, input$r2, input$q))
}
if(input$r_q > dataframes$rcq) {
dataframes$rcq <- input$r_q
reset(session, c('r1', 'r2', 'q'))
}
if (input$calc_f > dataframes$lcf | input$calc_f == 0) {
dataframes$lcf <- input$calc_f
dataframes$f_es <- isolate(get_f(input$fstat, input$dfh, input$dfe))
}
if (input$r_f > dataframes$rcf) {
dataframes$rcf <- input$r_f
reset(session, c('fstat', 'dfh', 'dfe'))
}
if (input$calc_td > dataframes$lctd | input$calc_td==0) {
dataframes$lctd <- input$calc_td
dataframes$td <- isolate(get_td(input$pairs, input$dm1, input$dm2, input$dsd1,
input$dsd2, input$r, input$dtval))
}
if (input$r_td > dataframes$rctd) {
dataframes$rctd <- input$r_td
reset(session, c('pairs', 'dm1', 'dm2', 'r', 'dtval', 'dsd1', 'dsd2'))
}
})
output$ti <- renderTable(dataframes$ti, striped=TRUE, digits=5, bordered=TRUE)
output$c_q <- renderTable(dataframes$c_q, striped=TRUE, digits=5, bordered=TRUE)
output$f_es <- renderTable(dataframes$f_es, striped=TRUE, digits=5, bordered=TRUE)
output$td <- renderTable(dataframes$td, striped=TRUE, digits=5, bordered=TRUE)
}
# Run the application
shinyApp(ui = ui, server = server)