using python to automate data translation process from eprime to SPSS

[David McFarlane]

Kyle,

Not sure I entirely understand your request, but I gather it has 
something vaquely to do with wanting to automate the processing of 
batches of single data files.  I realize E-DataAid does not include a 
mechanism to handle *batches* of .edat files, but you might try the 
free AutoIt (http://www.autoitscript.com/site/autoit/ ) as a 
scripting tool to automate such a batch process using E-DataAid.  If 
you do that, please write back!

Addressing your more specific question, AFAIK there is no 
documentation of E-Prime's data .txt format.  But really, the format 
is quite readable, and anyone with a modicum of skill could figure it 
out just by generating a judicious set of examples and reading 
them.  Students here have done it from time to time (goodness knows 
why), and I could do it myself, but I have never found a good reason to bother.

-----
David McFarlane
E-Prime training 
online:  http://psychology.msu.edu/Workshops_Courses/eprime.aspx
Twitter:  @EPrimeMaster (twitter.com/EPrimeMaster)


At 8/19/2012 05:55 AM Sunday, Peter Quain wrote:
>Don't know why you'd want to go to this effort to save a single menu 
>driven operation (save as), but at least as a start it seems  ... 
>That asterisks are Section delimiters (e.g., *** Header Start *** 
>and *** Header Ends *** ... or *** LogFrame Start *** and *** 
>LogFrame End ***); within all Sections each line records a variable 
>name and value; that the colon (:) delimits variable name and value 
>per line within Sections; and that Level numbers in the colon 
>delimited format occur on lines between asterisk delimited section 
>End and Start statements. The Experiment structure exists as a 
>hierarchy of n levels: (descending ...) Session (Experiment) / Block 
>/ Trial/ ...  withSubTrial elements 1 ...n ... or something like 
>that. SubTrial elements are logged attributes from a trial sequence, 
>and in my sample file all occur as children of Level 3. Now it is a 
>matter of constructing an algorithm which transposes this (well, the 
>correct ...) file structure appropriately to dispense with useless 
>data and collect single variables of interest in single columns, not 
>multiple rows across sections.
>
>It is perhaps possible that you could accomplish this within SPSS 
>using (as a starting point) String Function syntax embedded in 
>Python script looping structures. The raw .txt file will import into 
>SPSS as a single string variable if you define no delimiter, and the 
>text qualifier as 'Nothing'. So in fact it is easy to circumvent 
>edataaid in terms of getting the edat .txt format data file into 
>SPSS, it is as you say the parsing which presents the exercise. 
>Maybe you would be looking for SubTrial elements logged after a 
>Level 3 Section was identified, and prior to a *** LogFrame End*** 
>statement. I think though that it might be equally possible to parse 
>the file using Python libraries alone (as has obviously been done by 
>mysterious python forum person(s) with their tab-del output) ... I 
>suppose you would look for string functions to parse the file, array 
>commands to format the data, and write(?) commands to set the file 
>to disk. (?) Of course, there are many ways to skin a cat. But in 
>any way it is correctly capturing the data elements sequentially 
>from each Level 3 (in my example) child that will perhaps win the 
>beginning of the day at least.
>
>Best
>Peter
>
>
>
>At 01:27 PM 19/08/2012, you wrote:
>>Thank you for your reply Peter but that is not exactly what I was 
>>looking for and I probably did not clarify enough.  I know how to 
>>use EdataAid, I just want a way to circumvent it.  If you look in 
>>the folder where your edat files are stored there is also a .txt 
>>file with the data information.  I was wondering if I could somehow 
>>translate this file into something useable and skip edataid all 
>>together.  I am trying to automate the process so I can do it with 
>>a large amount of data all at once and ensure that it is all done 
>>the same way without having to worry about human error (like when 
>>someone accidentally forgets to uncheck the unicode box or clicks 
>>on the wrong thing).  It looks like the txt file contains the data 
>>from the experiment but it looks all jumbled up and in one 
>>column.  Someone on the python forums said that they had created 
>>script that would parse out the data from that text file and make 
>>it usable.  Unfortunately they did not really elaborate how to do 
>>that.  Thank you again for your reply Peter and I hope I helped 
>>clear things up a bit.
>>-Kyle
>>
>>On Saturday, August 18, 2012 7:33:05 PM UTC-4, Kyle wrote:
>>Hello,
>>I am currently struggling with trying to automate my data 
>>translation process from eprime to SPSS.  I am not using emerge 
>>because I need to create individual spss data files for each of my 
>>subjects.  I have written sytax that automates the translation from 
>>the edat output (tab-separated) to an SPSS data file but I need to 
>>figure out a way to circumvent e-dataAid so I can automate the 
>>whole process and reduce the chance of any errors during the 
>>process.  I was planning on using python so that I could have the 
>>user select the files that they want to convert to SPSS and then it 
>>does it automatically.  From what I gathered from python help 
>>forums is that I need to parse the raw data txt document into a 
>>tab-separated file.  Is there a way to do this or documentation how 
>>to do this somewhere?  The workflow would be something like: raw 
>>data txt file --> tab separated --> spss syntax (it reads from a 
>>list file I created in spss) --> spss data file.  I think the main 
>>spot I am stuck at is parsing the data from the raw file so I can 
>>circumvent edataAid.  From there it seems like it might be somewhat 
>>straightforward.  Thank you in advance for any help you can 
>>provide.  If you have any tips for the python part of this process 
>>that would be appreciated as well, but I know the group is for 
>>e-prime and do not expect help with that.
>>Thank you,
>>-Kyle

-- 
You received this message because you are subscribed to the Google Groups "E-Prime" group.
To post to this group, send email to e-prime at googlegroups.com.
To unsubscribe from this group, send email to e-prime+unsubscribe at googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.