ezANOVA() fails for unbalanced completely between-Ss design

[or duek]

Hi,
I encountered this problem also.
https://github.com/mike-lawrence/ez/issues/1
As I'm not an expert, can anyone tell me how to implement the fix?
thank you very much.

[Mike Lawrence]

See the message on the ez github home page
(https://github.com/mike-lawrence/ez) for instructions on how to
download and load the latest ez source.

[or duek]

thank you very much for that.
But...now I get a different error message:

Error in order(str_count(to_return$ANOVA$Effect, ":")) :
could not find function "str_count"

both if I run ezPLOT or ezANOVA.

[ROLO]

Hi Mike,

I found that I needed to add the ssl.verifypeer option to your
download instructions:

getURL( 'https://raw.github.com/mike-lawrence/ez/master/R/ezDev.R',
ssl.verifypeer=FALSE )

Maybe you can update them?

And another question (I hope I am not hijacking this thread): when do
you plan to release a new version of the package via CRAN? Having a
stable, released package would make advocating to use ez to my (rather
new to R) colleagues much easier for me.

Best regards and thanks for the nice work with ez,
Roelof

On Oct 23, 5:32 pm, Mike Lawrence <Mike.Lawre...@dal.ca> wrote:

[or duek]

It seems to have that option in the github page.

Do you also suffer from the same problem as I in the ezPlot function?

[ROLO]

That option was added this week, after my suggestion, thanks Mike.

And I did not suffer from the ezPlot problem, but from this ezANOVA
problem:
https://github.com/mike-lawrence/ez/issues/1

which has been solved in the development version.

[or duek]

I really don't know why do I get this strange error in ezPLOT (using
the latest development version)

Warning: Data is unbalanced (unequal N per group). Make sure you
specified a well-considered value for the type argument to ezANOVA().

Error in order(str_count(to_return$ANOVA$Effect, ":")) :
could not find function "str_count"

Any help would be appreciated.

[Mike Lawrence]

Sorry for not replying earlier, I was down all week with a rather
potent cold and only sporadically doing email.

Your error suggests that somehow ezDev either failed to install or
failed to load the stringr package. Try doing this yourself manually:

install.packages('stringr')
library(stringr)
library(RCurl); eval(expr =
parse(text=getURL('https://raw.github.com/mike-lawrence/ez/master/R/ezDev.R',ssl.verifypeer=FALSE)));
ezDev()

[or duek]

thank you very much !!
That did the trick.

Hope you'll feel better.

On Oct 29, 11:18 am, Mike Lawrence <Mike.Lawre...@dal.ca> wrote:
> Sorry for not replying earlier, I was down all week with a rather
> potent cold and only sporadically doing email.
>
> Your error suggests that somehow ezDev either failed to install or
> failed to load the stringr package. Try doing this yourself manually:
>
> install.packages('stringr')
> library(stringr)
> library(RCurl); eval(expr =

> parse(text=getURL('https://raw.github.com/mike-lawrence/ez/master/R/ezDev.R',ssl.verifyp...