Description Usage Arguments Value Author(s) Examples
Detects guilty bots in Mturk data based on Qualtrics lat/long coordinates
1 | guilty_bot(lat, long, dat, more_than=10)
|
lat |
Column name (character) of survey latitude coordinate, typically generated via Qualtrics |
long |
Column name (character) of survey longitude coordinate, typically generated via Qualtrics |
dat |
data.frame() object, housing survey responses, probably downloaded from Qualtrics |
more_than |
more_than = numeric number, default = 10 |
Returns dataframe() object of all rows fitting "more_than" criterion. Users should then examine lat/long overlaps. Turk IDs, IP addresses, and demographics may all be different, but 15 responses from the exact same lat/long coordinates seems unlikely. Users can then examine these cases for botness with various validity checks.
Loren Collingwood <loren.collingwood@ucr.edu>
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 | ######################
# guilty_bot Example #
######################
n <- 100
lat <- c( rep("42.999967", n), rnorm(n*4, mean = 44))
long <- c( rep("-80.444825", n), rnorm(n*4, mean = -83))
toy1 <- rnorm(n*5, 20)
toy2 <- rnorm(n*5, 30)
toy3 <- rnorm(n*5, 10)
toy_turk_id <- paste("faux_id:", round(runif(n*5, 0, 2),6), sep="")
df <- data.frame(lat, long, toy1, toy2, toy3, toy_turk_id, stringsAsFactors = F)
# Prints out problem coordinates and returns original dataset but only bots
no_soup_for_you <- guilty_bot(lat = "lat", long = "long", dat = df)
# Keep only non-bot data
keep <- !df$toy_turk_id %in% no_soup_for_you$toy_turk_id
df_bot_free <- df[keep,]; uniqueN(df_bot_free$toy_turk_id)
head(df_bot_free)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.