Sensitivity of AJUS to tolerance parameter

Last week I introduced Galtung’s AJUS system and its implementation in R. As noted, the tolerance parameter is not a trivial matter, but we can use the provided function ajusCheck to explore how different tolerance parameters affect the outcome. Basically we throw a number of tolerance parameters at the ajus function, and see what happens.

In this example, the default values give type “S” (i.e. multi-peaked). Next, we visually inspect the distribution using ajusPlot:

ajus_

So maybe the last increase is not significant, so we can try a different tolerance value (e.g. 0.2). We can continue this, or just try them all (e.g. between 0.1 and 2):

library(agrmt) # install from R-Forge
# Data:
V <- c(0,0,1,2,1.5,1.6)
# AJUS
ajus(V) # using default tolerance = 0.1
ajusPlot(V) # visual inspection
ajus(V, tolerance=0.2) # exploring
ajusCheck(V, t=seq(0.1,2,0.1)) # let's try them all

$tolerance
[1] 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9
[20] 2.0

$type
[1] “S” “A” “A” “A” “J” “J” “J” “J” “J” “J” “J” “J” “J” “J” “J” “F” “F” “F” “F”
[20] “F”

Which is it? Usually the substantive meaning of our numbers is the guide but sometimes we deal with constructs that are more difficult to assess. Using ajusCheck we can easily check whether we should spend hours thinking about whether 0.2 or 0.3 is more appropriate – or whether we get the same result.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s