Different outputs for freq and freq +d3 commands

[Leonid Spektor]

Anne-Christin,

    Thanks for the sample file and exact instructions of how to replicate
the problem. I was able to find the bug. It occurred only when +d2 or +d3
option was used, i.e. statfreq program was involved. I have fixed this and
new CLAN is on childes web site.

Leonid.


On 15-02-08 05:09, "Anne-Christin Tannhäuser" <act at khi.is> wrote:

> 
> Hello Leonid,
> 
> here the information you asked for:
> 
> command for calculation in CLAN output window:
> freq +s at nouns. +t*SBJ
> 
> command for output in file:
> freq +s at nouns. +t*SBJ @ +d3
> and then
> statfreq stat.out.cex +f +d
> 
> I use a PC and Windows XP.
> Our data does not have a complete %mor line and therefore we used a
> list of nouns to count their occurence in each file.
> I email you the language data files, the noun list, plus the 2
> different outputs I got.
> 
> Thank you so much in advance,
> Anne-Christin
> 
> 
> 
> On Feb 6, 7:22 pm, Leonid Spektor <spek... at andrew.cmu.edu> wrote:
>> Anne-Christin,
>> 
>>     Could you please give us command lines you use and email sample files
>> that demonstrate this problem. Also, tell us which Computer (PC, Mac PPC,
>> Mac Intel) and Operating system (Window 2000, XP, Vista, OSX 10.x) you are
>> using.
>> 
>> Thanks,
>> Leonid.
>> 
>> On 06-02-08 11:25, "Anne-Christin Tannhäuser" <a... at khi.is> wrote:
>> 
>> 
>> 
>> 
>> 
>>> I have a remark to make concerning the freq and freq +d3 (giving out
>>> the stats, only) commands: In my analysis seems to be a constant flaw
>>> concerning the freq +d3 output, i.e. all type counts are one item
>>> higher than indicated by the freq analysis. This becomes especially
>>> obvious, when using word lists to check for the occurence of certain
>>> items and some data files contain none or only a few of those
>>> particular items on the list. Then the TypeTokenRatio of the freq +d3
>>> output will eventually become higher than 1.
>>> I double and tripple checked the outputs and always ended up with the
>>> same mistake. Has anybody encountered the same problem?
>> 
>>> Kind Regards,
>>> Anne-Christin
>> 
>>> PS: Thank you Brian McWhinney for commenting on adjusting the header
>>> lines of older CHAT files to the latest format.- Hide quoted text -
>> 
>> - Show quoted text -
> > 
> 



--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "chibolts" group.
To post to this group, send email to chibolts at googlegroups.com
To unsubscribe from this group, send email to chibolts-unsubscribe at googlegroups.com
For more options, visit this group at http://groups.google.com/group/chibolts?hl=en
-~----------~----~----~----~------~----~------~--~---