Dear Chanan,
"cooccurrence" as used in the ATLAS.ti model exploits the spatial relations
of data segments ("quotations") e.g., those overlapping, following,
embedding one another. Cooccurence of "events" does only map into this
scheme as long as the linear sequence of your textual (or video, audio)
indicators represent the time line of events, as with many narrative texts.
Of course this is not necessarily the case with other data. To accurately
model "co-occurrence of events" you would need to more explicitly represent
an "event", foremost including the date and time it happened. You could
manually "redate" the quotations representing the event episode at a later
stage. However, to be able to exploit such representations when querying
the data, a better aproach would be to assign time points or intervals
represented by codes and code families to the quotations.
Such codes could be termed: "2004-12-20" or "2004-12-10_20" or "2004-12" ,
the latter being an abstract code connected to all 2004-12-xx codes via the
"isa" relation. Alternatively (as recursive nesting is not neccessary) a
code family could be used to represent the range.
This would allow queries like:
"All events happening on December 20, 2004" or - using the Query Tool's SUB
operator - "All events happening in December, 2004"

To manage such time codes you would also introduce a "type" family "Times
and Intervals". But this is for house-keeping not for modeling!-)

Take this as a first raw appraoch.

Regards
- Thomas


At 07:13 27.12.2004, you wrote:
>Dear All,
>I wonder how you can incorporate the time dimension within the Atlas coding
>scheme in order to be able to look for co-occurring events? Does the Atlas
>support "Nesting levels" analysis? (looking automatically at events that
>occur at the same time frame) or to preform a "Leg sequential" analysis?
>Chanan
>************************************************* ********
>Elchanan Gazit
>LIRT Lab, Institute for Interdisciplinary Applications of Computer Science
>The Caesarea Rothschild Institute, University of Haifa
>Mount Carmel, Haifa 31905, Israel
>Office: Education building, room 565
>E-mail: gazit@cri.haifa.ac.il
>http://www.tau.ac.il/~gazit/
>Phone (work) +972+4+8288371/2
>(cell) +972+545+277800
>E-mail (home): vrlearn1@netvision.net.il
>Telfax (home): 972+3+5405974
>************************************************* *********
>The Quotation of the Month:
>
>Play is not a means for the child to have fun. Rather, it serves a
>fundamental role in the child's development, because it creates a zone of
>proximal development in which the child always behaves beyond his average
>age, above his daily behaviour - Lev Vygotsky (1978)
>
>
>-----Original Message-----
>From: ATLAS.ti The Knowledge Workbench - User Forum
>[mailto:ATLAS-TI@LISTSERV.DFN.DE] On Behalf Of MicheleR
>Sent: Monday, December 27, 2004 3:01 AM
>To: ATLAS-TI@LISTSERV.DFN.DE
>Subject: Re: Lising codes that co-occur
>
>If you are looking simply for codes that occur in the same PDs, you can use
>Codes > Output > Primary Docs Table. This gives a chart showing codes down
>the left and PDs across the top; by scanning down the columns you could spot
>codes that occur in the same PD.
>
>If you are looking for codes that co-occur in the sense of being inside one
>another, the easiest way is to go into the query tool. Look at the "codes"
>pane (lower left, says "Codes:" at the top) where all the codes are listed.
>Select any code, right click on it, and choose either "Embedding codes"
>which will show you other codes that contain this one, or "Embedded codes"
>which will show you the codes that are embedded within this one. Very
>handy!! The only problem is there isn't a quick way to generate a list, nor
>could I find a way to save the information -- you will have to go down your
>whole list, taking notes as you go, I guess.
>
>Michele
>
>
>----- Original Message -----
>From: "Marian Reiff" <mr82@COLUMBIA.EDU>
>To: <ATLAS-TI@LISTSERV.DFN.DE>
>Sent: Thursday, December 23, 2004 3:38 PM
>Subject: Lising codes that co-occur
>
>
> > I am having trouble getting a list of CODES (or families) that occur
> > in conjunction with another code.
> > I tried queries and networks but I can get only a list of quotations.
> > I am still using Atlas 4 (I'm ashamed to say!) but hope it can be done
> > somehow. Has anyone done this successfully?
> > Thanks.
> > Marian.

__________________________________________________ _____________________
_Computers, like every technology, are a vehicle for the transformation
of tradition_ (Winograd & Flores, 1987)
ATLAS.ti Scientific Software Development GmbH - Berlin - www.atlasti.com
Dipl.-Psych. Dipl.-Inform. Thomas Muhr (CEO) - Muhr@ATLASti.de