Repeated Measures data in spreadsheets

[Darinka Andjelkovic]

Dear Brian and colleagues,

 

It seems I was not clear in my letter. 

 

This kind of order is necessary when applying General Linerar Model Repeated
Measures since same measures are within-subjects repeated.

It is actualy general and typical model of ordering data for any
within-subject factors design (like large quantity of data from
psychological tests where one participant has data for different
items-measures from test and retest, etc...).

 

I am aware of MORTABLE output. It is the same in ordering data as output of
FREQ and MLU, and I find it realy usefull.

But if one wants to compare the same measure (like morphems/utterances) for
the same child  from different age levels (repeated measures), then this
order is required for statistical reasons.

 

I send to Brian one of my output files as an example. 

 

The first sheet contains the table generated by CLAN MLU. 

The second sheet (named Target) has data that I sorted manualy.

I had to change the cells in column A in order to make clear that data are
coming from a set of files (different ages) for the same child.

And I added one row above all in order to make clear that columns contain
the same data (measure) coming from different age levels.

 

I know nothing about MEASURES and EVAL and I dont find any information of it
on the site, nor in the Manual (I use the last version).

 

I hope the file that I made will make it more clear.

 

Best wishes and thanks for your concern,

 

Darinka 

 

 

From: chibolts at googlegroups.com [mailto:chibolts at googlegroups.com] On Behalf
Of Brian MacWhinney
Sent: Saturday, May 05, 2012 6:05 PM
To: chibolts at googlegroups.com
Subject: Re: Repeated Measures data in spreadsheets

 

Dear Darinka,

 

    This format of subjects in rows and scores in columns is what we are now
using for the MEASURES, MORTABLE, and EVAL programs that output Excel-type
tables.  However, if you are creating various custom measures through
special purpose runs on data, it is difficult for me to see how CLAN can
guess the format of what you are creating.  Maybe you are saying that you
are creating a collection of separate files for each subject and then trying
to cut and paste data from these separate files into a single file.  But, in
that case, maybe you should have run your command on *.cha using *CHI as the
consistent identifier for your target child rather than "Peter" "Mary" and
"John".  

 

In any case, some of this can be done with what we already have and some
cannot.  It all depends on the details of how your data is coded, how
subjects appear across files, and so on.  In the end, sometimes it may just
be necessary to resort to cut and paste.  However, if there is some
consistent and typical method that we can implement, we would try.  But we
would need to have it very closely described.  Take a look at MEASURES,
MORTABLE, and EVAL to get some sense of how this might work.

 

Best regards,

 

-- Brian MacWhinney

 

On May 5, 2012, at 11:00 AM, Darinka Andjelkovic wrote:





Dear colleagues,

 

I have tried the command that Leonid recently made for Joyce Marzan for the
purpose of rotating CLAN spreadsheet output on its side, turning columns
into row and vice versa:

spreadsheet stat.frq.xls

and it works perfectly.

 

But my question is about  possibility to use this device for the preparation
of data file for further statistical analysis.

 

In most studies of language development one would look for one or more
measures (let's say number of utterances or morphemes) at different age
samples for the same child.

And, for the statistical reasons we have to treat this as a Repeated
Measures design for different age levels. So, working for example in SPSS,
we have to sort age samples into different columns like all other variables
(utterances, morphemes, etc.), while participants should be sorted in rows
like this:

 

                utter18m    utter19m   utter20m  morph18m   morph19m
morphs20m

Peter

Mary

John

 

I do not see any other way of preparation of Excel data into that kind of
sort,  except for manual cutting rows and pasting them into collumns by
means of PasteSpecial/Transpose.

It is rather slow, and very susseptable for making errors, especialy for a
large quantity of data.

 

How do you handle this?

Maybe a Macro would work, but I do not know how to make it. Is there any
available around, that I could use?

 

Any suggestion would be most appreciated.

 

Best wishes,

 

Darinka Andjelkovic

 

 

 

 

 

 

-- 
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.

 

-- 
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.

-- 
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.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/chibolts/attachments/20120506/3a2ff6ea/attachment.htm>