Computerized Profiling

Home CP Information Downloads Support Archive Dialects References Search Contact

Page last modified: April 23, 2008

Support

Frequently Asked Questions

Click questions to jump to answers and discussion below (questions are repeated and extended as well).

Answers and Discussion

How do I run CP under Windows 3.x?

To run the program from Windows, create a Program Item that launches the program. To do this you enter:

Description: Computerized Profiling
Command Line: cprof
Working Directory: c:\cp [assuming that you stored the program in a subdirectory named "cp"]

Then, using the PIF Editor, create a CPROF.PIF file. In that file, select "Windowed" under "Display Usage". When the CP program item is double clicked, it launches the program in a window. You can click the Control-menu box (the one in the left-hand corner) and select "Fonts". The default font size is 8 X 12 but this can be enlarged to 10 X 18 or 12 X 16. If you click the "Save Settings on Exit" box then CP will always appear in the larger font. But, if for any reason you don't want it that size, just type Alt-Enter and you'll get the standard text display back.

to top

How do I run CP under Windows 95?

To run the program from Windows 95, click on the Start button, select Settings, then select Task Bar. When the Taskbar Properties window appears, select Start Menu Programs. Click Add, click Browse, and then locate the folder (subdirectory) where you installed CP and double click on the file CPROF.EXE. This creates a shortcut to the program, which you may then place on the Start Menu, on the desktop, or anywhere else you want it.

to top

[While running an automatic analysis in the LARSP+ module] one of my students using the program has gotten the message "corpus file contains contextual notes marking grammatical constituents. Do you want to suspend the auto word level analysis?" What type of markings is it referring to?

What this message means is that the program encountered a contextual note beginning with #, which is the character set aside for marking grammatical constituents in a CORPUS file. (There's a CORPUS help file called "Coding Constituents" that explains this.) When a SALT file is converted to CORPUS, any slash codes (e.g., /ing, /ed) are converted into # contextual notes (e.g., # ing:4). Since the slash codes in SALT are presumably correct, there is no need for the LARSP parser to look at word level (line 4). In fact, if it does, it may introduce some correct codes. So...the rule to follow is: if the CORPUS file was created from a slashed SALT file, answer the prompt, yes.

to top

I'm writing to ask about how one of the calculations at the end of the LARSP+ chart is generated. Specifically, I'm interested in how the denominator or #turns for either partner is determined when the mean # sent.\turn is calculated. Does it come from CAP data, or is it the number of utterances which come after a partner utterance?

No, it doesn't come from CAP. A turn is simply defined as a change in speaker, ignoring any Null Responses that may be coded in the transcript. For example:

P blah blah blah P turns =1 
T blah blah blah T turns =1 
P blah blah blah P turns =2 
P blah blah blah 
P blah blah blah 
T blah blah blah T turns =2 
P null 
T blah blah blah 
P blah blah blah P turns =3 

P sent/turn = 5/3 = 1.67 
T sent/turn = 3/2 = 1.5 

to top

I have a question as to what Corpus does with partially intelligible utterances eg "Mommy XXX". Would that get counted as intelligible or not in the utterance count? Then if I ran an IPSYN would such an utterance get counted in the analysis or not, or should I compensate by adding another utterance to get my 100 successive?

The CORPUS utterance count works the same as SALT, that is, to be counted as "complete and intelligible" an utterance can have no Xs in it. To put it another way, partially intelligible utterances (no matter how partial) are counted as unintelligible. IPSyn, on the other hand, looks at the LARSP codes, not the CORPUS file, to determine what is unintelligible and therefore should be excluded. The question, then, is how to get the CORPUS utterance count to line up with the LARSP codes. The answer is that in Preferences, option I, LARSP+ Code Unintelligible Utterances should be set to "All Xs". What will happen then is that any utterance containing one or more Xs will be coded in LARSP as UT (unintelligible). Then the IPSyn module will automatically exclude all such utterances.

to top

Is there a way of re-viewing the PROPH Profile (.ph1) file in CP without having to go through the whole "tabulate" process?

Yes. At the Main Menu, select PROPH (8), Search File (5). Alternatively, at the Main Menu, select File Viewer/Editor (J), press F2, select the filename, select the .ph1 file. The File Viewer/Editor module can be used to edit the profile, or you can do this in a word processor, too.

to top

How do I cite CP in a paper?

The date of release and version number of CP are shown when you select "About CP" at the Main Menu. This information should be inserted into the following citation form:

Long, S.H., Fey, M.E., and Channell, R.W. (XXXX). Computerized Profiling (Version X.XX). Cleveland, OH: Case Western Reserve University.

to top

Can you please send me the full reference for Fey (1986) so I can look it up. I also want to know where can I find the whole bibliographical list for CP.

If you go to the Main Menu and select CP Help Menu, then CP Program Features, then References, you'll get the full citations for Fey (1986) and every other work cited in the documentation.

to top

Was any manual ever published for CP, something like the MacWhinney (1995) for the CHILDES program?

The manual for CP is incorporated into the program and does not exist as a separate document.

to top

When I open CP now I get the whole screen which is fine, but how can I get it to a partial screen when I need to drag and drop a file? I am using Windows95B.

To have CP appear in a window rather than full screen, press Alt-Enter. To configure the program so that it always opens in a window, go to the shortcut you use to launch the program, right-click on it, select Properties, click on the Screen tab, and then under Usage click the radio button that says Window.

to top

I downloaded the program and started to look at the tutorials. It must be me but each tutorial I opened sent me to another one to be done first. Help, what am I doing wrong? We'd certainly like to see if this would be useful for us but haven't cracked the code of how to use it. I'd appreciate your input.

Start with the CORPUS tutorial, as this explains how to create or import a transcript file and "tag" it using the syntax parser. (Even if you're not analyzing the speaker's syntax, the program uses this parsing information.) When you've completed CORPUS, you can try any of the other tutorials. Also, in case you didn't notice them, at the web site are a set of downloadable instructional movies that show/narrate the steps for different parts of the program. The movie files are quite large, so if you're downloading through a modem, they may take a few minutes and you'll probably want to delete them after you've watched and learned the procedure. Otherwise, they'll start to hog space on your hard drive.

to top

I recently downloaded CP and attempted to use it to analyze a language sample. I was successful in entering the sample and reached the point at which the program checks it for grammatical errors (apostrophies, periods, etc...). When the CP reaches the last line of my sample and I answer the last grammer question (about making a proper name all one word)with a "Y", the screen goes blank and it tells me "unknown error". press a key and puts me back into the menu screen without creating a corpus file. I don't think I ever get to the point at which I am supposed to "tag" the sample - and I'm not sure exactly what that means.

Have you tried doing the CORPUS tutorial and, if so, did it work correctly? If you haven't done it, do so (at the Main Menu, select Tutorials Menu, then select CORPUS, and you can print out the instructions). If the tutorial works correctly, then we know the program is properly installed. In that case, what will work best if you email me, either as an attachment or simply pasted into an email message, the transcript you're attempting to analyze. It's possible that you've stumbled across a bug that has heretofore escaped our notice. But, first things first. Try the tutorial and then, if necessary, send me your transcript file.

to top

Are you aware of any technical problems with running CP under Virtual PC on Macintosh computers? I am attempting to install the program on such a system; after unzipping the files, I get to the initial install screen where you select the program defaults and when I hit a key to make a change it causes the program to exit citing an illegal operation. Actually the program crashes regardless of which key you press. If you have any support information, it could help me to troubleshoot the installation.

[from Ron Channell]
Steven Long forwarded your message to me. I am a Mac-using person with little experience with Windows, but I am succesfully running CP 9 under VPC 3.0 for Windows 98 on my Mac (Powerbook G3, 64Meg, OS9). It installed okay for me, though I didn't try to make any customizations right off, so perhaps I'm not going to be able to duplicate the problem you reported.

I had downloaded the sections from Steven's server on the Mac side using Internet Explorer. Next, I moved the 3 sections into my shared folder (which VPC lets you set up; mine is called F:shared) Then I started the VPC program and moved the 3 files from my shared folder to my C: drive before running them. (That was a bit of a nusiance, because Windows wanted to make shortcuts instead of actually moving the files.) With the 3 files on the C: drive window, I then ran the installer. Like I said, I didn't make any changes in terms of deleting modules, but I was able to go further and browse around. I have since shut down Windows and re-opened CP and created/analyzed a file, etc. without problems.

to top

I cannot seem to enter utterances into the program. The program says something like, file does not exist. I can't get past selecting the choice for creating DSS file. Please tell me what I am doing wrong or how to enter my utterances.

What you need to do is to begin with the CORPUS module. (You may want to do the CORPUS tutorial. At the Main Menu, select Tutorial Menu, then select CORPUS.) CORPUS is the module used to create or import a transcript text file and then "tag" that file for use by the various analysis modules in CP (such as DSS). The "Create a DSS file" option is used to generate a set of morphosyntactic codes for the transcript created by CORPUS. Make sense? Like all software, there's a procedure to follow. Once you've been through the procedure the first time, things get much simpler. Remember that, in addition to the tutorials, there are available at the web site a set of downloadable instructional movies that show and narrate the steps for various parts of the program.

to top

I have attempted to download the CP program from your website, but my computer does not recognize the type of file that is downloaded. We have Windows on the computer. Is there a certain type of read-me file that we need on our computer? Is there some other procedure in downloading than clicking on the downloading icon (in three parts)?

Usually a web browser will download the file once you click on the link. If yours isn't doing this, right-click on the link and then select "Save Target As" (that's what Internet Explorer says - if you're using a different browser, it may say something slightly different). This should allow you to download the three files to your computer. Take a look at the downloading instructions if you want to see how the entire process should work and let me know if you run into further problems.

to top

I have just installed CP version 9.2.6 and am unable to get started with this tool. I have tried to create a LARSP file and there is not a SAMPLE file installed (per the tutorial). The software would not let me create a file (via CREATE A LARSP FILE). I received an error message "No Files with the Path "C:\CPDATA\*.COR. Even with your sample file copied into C:\CPDATA I received the same error message until I copied SAMPLE.TXT to SAMPLE.COR.

What you need to do is to begin with the CORPUS tutorial. CORPUS is the module used to create or import a transcript text file and then "tag" that file for use by the various analysis modules in CP (such as LARSP). The "Create a LARSP file" option is used to generate a set of morphosyntactic codes for the transcript created by CORPUS. Like all software, there's a procedure to follow. Once you've been through the procedure the first time, things get much simpler. Remember that, in addition to the tutorials, there are available at the web site a set of downloadable instructional movies that show and narrate the steps for various parts of the program.

to top

I get through to the program setup then the setup screen does not appear. So I can't configure the program. Is there is anything special I can do?

Go to downloading instructions and scroll down to "Installation Problems". Follow the procedure described there. If that doesn't solve your problem, let me know.

to top

A few years ago I purchased a computerized profiling manual and disk at an ASHA convention. Is this the same program that I purchased which is currently sold by the Psychological Corporation?

CP is the same program *formerly* sold by TPC but in a much updated version. The changes made in versions produced since the one you purchased are summarized here. Now, of course, CP is a free program.

to top

I plan to introduce CP to my students in a course this semester. I am wondering if I can load it on a server and have the students access it that way or is it best to load it on each individual PC.

CP was not designed for use on a server. It appears to work on some but not all systems. The best advice I can give you is to install it on the server and try accessing the program *simultaneously* from more than one machine. If it doesn't work, then (obviously) take it off the server and install it on each machine individually.

to top

Do you have anything that could be used or adapted for hard of hearing and Deaf children? Testing protocols are few and far between for these kids. I would like to look into creating test measurements for this specific population.

CP will be useful with this population to the extent that oral language sampling and analysis or phonological analysis of a speech sample is useful. Certainly it is the case that Deaf or HOH children exhibit phonological, syntactic, and other language behaviors that differ considerably from, say, children with SLI or mental retardation. CP is not set up to deal specifically with one population any more than another. My suggestion is that you take a look at the web page and the description/examples of the program and what it produces. If you then have any questions about how certain analyses might be applied to your population, I'd be happy to try to answer them.

to top

Is it possible to add any additional codes to the pragmatic section (eye gaze, touching, etc.)?

Not to the CAP module. However, you can create and use an alternate set of LARSP error codes. Read the help file and watch the movie on Edit LARSP Error Codes. What you would do, then, is enter your additional pragmatic codes as LARSP "errors", tabulate the file, and then only pay attention to the Error section of the profile. You wouldn't actually do a LARSP analysis; you'd just be using the LARSP module as a way of entering/tabulating/and searching for your pragmatic codes. CP stores CAP and LARSP data in the same, so it's possible to display all these codes on the screen at the same time.

If the above sounds complicated, it's because you're tricking the program to do something that it wasn't optimally designed to do. But the procedure itself is fairly simple, once you get it down. Feel free to call me, if that will help clarify things. If you have a lot of data to analyze, you should make sure CP will do the job efficiently for you.

to top

We have imported our SALT files and have completed the transcription problems section. When we code, none of the words we selected as interjections are enclosed in parenthesis...We are asked if we want to include oh as an interjection and include in parenthesis. We respond with Y for yes, but it doesn't show up when we code. It does convert carrot/s to carrots, and go/3s to goes. Are we misunderstanding the question about interjections?

The purpose of the "transcription problems" module, in most respects, is to optimize a transcript for grammatical analysis. Thus, it checks for items such as interjections because these raise problems for morphosyntactic coding. For example, if a child says

oh, I can do that. 

should "oh" count as a morpheme? If it does, and you then proceed to LARSP the transcript, how is "oh" coded? I would argue that most forms described as interjections (oh, wow, jeez, etc.) should be mazed in the transcript. All that the "transcription problems" module is doing is calling these forms to your attention and asking if you'd like to maze them, i.e., enclose them in parentheses. Now, what CP does with forms that are mazed in the transcript text file is (1) remove them from the CORPUS file (which is the file that gets analyzed by the LARSP, DSS, IPSyn, CAP, etc. modules) but preserve them in a contextual note that appears above the utterance line. Thus, the line above would become

* (oh), I can do that. 
I CAN DO THAT. 

Note that you don't *have* to maze interjections, in which case they will appear in the CORPUS file. For that matter, you don't *have* to use the "transcription problems" module--you have the option of skipping it and going directly to file tagging.

to top

How does CP compare to SALT? Why create a new program if SALT is available?

SALT works with a user-coded transcript and all the codes (marking bound morphemes, clause or phrase types, grammatical errors, or whatever) must be decided upon and typed into the transcript by the user. CP, on the other hand, uses a set of dictionaries and linguistic parsing algorithms to automatically code the transcript. Depending on what kind of information is being sought, these codes must then be reviewed and some or all of the mistakes committed by the program corrected. CP is designed to yield a number of analyses that provide a developmental perspective and that are already familiar to many clinicians and researchers (e.g., DSS, LARSP, IPSyn, PRISM-L). CP also performs analyses both on orthographic and phonetic data, that is, it is useful for both "language sampling" and "phonological analysis", as these terms are commonly used by clinicians. Finally, CP is designed to work WITH SALT. It can be used, for example, to generate a coded SALT transcript and it can also import SALT-formatted transcripts so that they can be analyzed by CP without retyping.

to top

I'm having difficulty importing my SALT file in the CORPUS module. I've printed out the help pages but I'm still having difficulty. Our file name does not appear on the screen, when I press the TAB key and the file location, but it gives an error and defaults me to the main menu. Do you have any suggestions for me?

First, you should set CP's data location to the folder that contains your SALT file. To do this, go to the Main Menu, press B, press Tab, then enter the pathname for that folder, e.g., c:\cpdata\research. Then go to the CORPUS module, select Import Text File, SALT format. If your SALT file has an extension of .slt, it will appear on the screen and you can select it. If it has a different extension, press Tab, then enter the 3-character extension. Your file should then appear on the screen and you can select it.

to top

I seem to be having a problem loading the SIL phonetic fonts. I downloaded the silip12a.exe file, double clicked it, answered "yes" to all the questions, but don't see the font in the font folder or in MS Word 97. I've restarted the computer several times, still I don't find it anywhere. What am I doing wrong?

The font names are SILDoulosIPA, SILManuscriptIPA, and SILSophiaIPA. Since Windows applications display fonts in alphabetical order, you'll need to scroll down to S. If you still don't see the fonts, take a look at the questions and answers page at the SIL web site (http://www.sil.org/computing/fonts/ipafaq.html).

to top

Is CP appropriate for accent reduction?

Yes, to the extent that linguistic analysis is helpful. For example, the PROPH modules can be used to simplify the task of analyzing phonetic transcription data. If you administer an articulation test of some kind and carry out analyses on it, then I think you'll find that CP can save you time.

to top

This is a basic question, I think. How do you save a proph file so that I can print it from a computer that doesn't have the program? What I need to print is the analysis page. Can you save it and open it in wordpad or word?

The complete PROPH profile is in the file with the extension PH1. It can be opened by any word processing program. The issue is changing the document font so that the graphics and phonetic characters print properly. The BEST solution is to download and use the Word macro that automatically converts the PROPH profile into a Word document with true IPA characters. The macro, and the Encore phonetic font it requires, are available at the CP downloads page. The alternative is simply to open the .PH1 file in Word and then change the font for the entire document to "MS Line Draw".

to top

I have been attempting to open PROPH profiles in Word97 using the macros and although when it converts everything looks ok on the screen when I go to print it I have little box characters all where the boxes should be and throughout the document. I notice in the instructions that to use the macro open the Proph document you want to convert and word will ask you how you want to import it. When I do it Word just automatically opens and does not allow me to specify Text Only. How am I able to get a paper print out of the conversion or is this not possible?

The problem appears to be with your printer driver, which doesn't support the box drawing characters used in the converted PROPH profile. There is a fix available. It's described in the Microsoft support document at

http://support.microsoft.com/support/kb/articles/Q159/4/18.ASP?LNG=ENG&SA=ALLKB&FR=0

Depending on how facile you are with the inner workings of Windows, you may want to enlist the support of a technically-inclined person to help you make the changes outlined in the document. Alternatively, you could save the PROPH profiles as Word files after the macro has converted them and then take them to another machine attached to a different printer to make your hard copies. The Microsoft document lists the printers that are affected by this driver problem. So, in theory at least, if you can use another PC connected to a brand of printer that's not on that list, you should be able to print out the profiles without a problem.

to top

 

Copyright © 2008 Steven Long. All rights reserved.