Dunn Test by A. Dinno user written package - unknown function ulength() r(133)

[Giuliano Di Tommaso]

Hey I have an error response in using the package, may I ask your help?
(one distribution Elasticity4anni, Isource2 as groupvar)
(I report more than one result because I tried to remove options)

. dunntest Elasticity4anni, by(Isource2) ma(none)

Kruskal-Wallis equality-of-populations rank test

+-----------------------------+
| Isource2 | Obs | Rank Sum |
|----------+------+-----------|
| DP | 1000 | 607828.00 |
| MI | 1000 | 2.50e+06 |
| NBIIp | 1000 | 1.39e+06 |
+-----------------------------+

chi-squared = 2408.491 with 2 d.f.
probability = 0.0001

chi-squared with ties = 2408.491 with 2 d.f.
probability = 0.0001


Dunn's Pairwise Comparison of Elasticity4anni by Isource2
(No adjustment)
Col Mean-|
Row Mean | DP MI
---------+----------------------
MI | -4.88e+01
| 0.0000
|
NBIIp | -2.03e+01 28.553381
| 0.0000 0.0000
unknown function ulength()
r(133);

end of do-file

r(133);

. do "C:\Users\Julian\AppData\Local\Temp\STD0000000 0.tm p"

. dunntest Elasticity4anni, by(Isource2) ma(none) nokwallis


Dunn's Pairwise Comparison of Elasticity4anni by Isource2
(No adjustment)
Col Mean-|
Row Mean | DP MI
---------+----------------------
MI | -4.88e+01
| 0.0000
|
NBIIp | -2.03e+01 28.553381
| 0.0000 0.0000
unknown function ulength()
r(133);

end of do-file




Thank you

[Alexis Dinno]

Dear Giuliano,

If you read the help file for dunntest (or conovertest, which is always more powerful than Dunn's test, and to be preferred), you will find the following:

Alexis Dinno
Portland State University
[email protected]

Please contact me with any questions, bug reports or suggestions for improvement. Fixing bugs will be facilitated
by sending along (1) a copy of the data (de-labeled or anonymized is fine) in Stata .dta file format, (2) a copy of
the command used and (3) a copy of the exact output of the command.

Which is a route of communication likely to provide a more rapid and accurate response than asking a public forum on the internet.

As to your question: the most recent update broke dunntest for versions of Stata earlier than 14. This has been corrected, and your can update dunntest by typing adoupdate, update within Stata.

[Giuliano Di Tommaso]

Thank you