Description Usage Arguments Value Author(s) See Also Examples
A function to remove outliers of a given data frame. Extreme values can be excluded based on a minimal/maximal values (thresholds) and/or based on a given standard deviation value. Return a data frame of filtered data as well as the number and relative percentage of data filtered.
1 |
data |
A data frame in the long format (one row per record). |
fpass |
A vector length two with the minimal and maximal accepted value to use for a first filtering of the reaction times. To filter only the lowest or highest values, indicate NA as value. For instance c(100, NA) will only remove RT < 100ms. Defaults to NULL. |
target |
A string indicating the column names of the data to filter. Defaults to 'RT'. |
sdv |
A number indicating how many standard deviations should be used to filter the data. Defaults to 3. |
tokeep |
A vector of column names to indicate in the filtering section which conditions were used in the processing, typically subjects and/or experimental condition. Default to NULL. |
Return a list with a data frame of filtered data and a data frame of number of data excluded and its relative percentage per condition.
Guillaume T. Vallet gtvallet@gmail.com, University of de Montreal (Canada);
Benoit A. Riou riouba@gmail.com, Lyon2 University (France)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 | # Generate fake data with a subject number in the first colum,
# a fake experimental condition in the second column and fake
# reaction times in the third column
df = data.frame(Subj=1, Cond="Test", RT=rnorm(25, mean=550, sd=50))
# Adding extreme values
df[5,3] <- df[5,3]+300
df[25,3] <- df[5,3]+500
df[19,3] <- df[19,3]-350
df[12,3] <- 55
df[15,3] <- 2340
# Filter with low and high thresolds and with 3 standard deviations
outliers(df, target='RT', fpass=c(100,1000), sdv=3)
# Filter with only a low thresold with 2 standard deviations
outliers(df, target='RT', fpass=c(100, NA), sdv=2)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.