NPS version 2.x makes substantial changes, some of which may disrupt or alter the interpretation of existing code based on NPS. Most prominently:
Intervals for the Net Promoter Score are calculated using a new method,
with superior performance. See ?nps_test
for details.
snake_case
No more dots. For example, nps.se
is now nps_se
. For the
first 2.x release, these will be depricated (they will work for
now, but with a warning
). It is strongly suggested that
you move to using the new function names as soon as possible.
Are now available, meaning that you can set options for how NPS
should work once, and it will stick with them for the rest of the
session. For example, suppose that your Net Promoter study uses a
5pt Likert scale, with 1-3 coded as Detractors, 4 as Passives, and 5
as Detractors. Instead of having to supply an argument to the
breaks
parameter each time you wish to use an NPS function, you
can now set options(nps.breaks = c(1,3,4,5))
at the start of
your session, and all calculations and tests will now be based on your
category defintions.
NA
no longer always omitted by defaultWhile certain parts of R
return NA
if there's an
NA
in the input, others (such as tests) tend to use
na.omit
by default. v2.x of NPS now follows this convention;
nps
will return NA
if there's any NA
s in
the input, whereas nps_test
will perform na.omit
and
emit a warning.
The two most common ways to describe 'Net Promoter responses' are as vectors of either Likert responses, e.g.:
c(0, 0, 0, 7, 7, 10))
or as aggregate counts of the categories, e.g., for the same distribution:
c(Detractors = 3, Passives = 2, Promoters = 1))
NPS v2.x delineates functions for these data types more clearly, with
those for the latter style always ending in _
. All v1.x
functions have equivalents, but those with the old naming convention
have been deprecated.
Brendan Rocks foss@brendanrocks.com
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.