Using visual-analogue scales in E-prime

[Tony Zuccolotto]

Dear Kai,

You should be able to achieve this by just using a Label object in the
procedure and a Goto statement in Inline script.

Put the Label above the object that collects your input and then after
the object check in script (using an IF statement) if the subject has
made the 'final' response.  If the 'final' response was not made then
just process the key as normal and then call Goto Label1 (replace Label1
with whatever you called your label).

The TrialProc would have the following object on it
	<Label1>	' designates where to jump back to
	<Stimulus>	' collects the response
	<Inline>	' tests for final response, updates interface
and jumps

Assuming you were using the '3' key on the SRBOX as an indication of the
'final' response the inline script you would need would look sometime
like
	If Stimulus.RESP <> "3" Then
		' move arrow and draw at new location
		....

		Goto Label1
	End If

Because you jump back to the top of the procedure logging to the data
file never occurs.   Thus after the final response comes in you just let
the Procedure end normally and only that 1 trial should be logged that
contains the final response you are interested in.

If you want the 'total' response time then you should save the
Stimulus.OnsetTime of the very first trial in a global variable and then
you can compute a total RT as follows.

	Dim totalRT as Long
	totalRT = Stimulus.RTTime - g_nOnsetTimeOfFirstStimulus
	c.SetAttrib "TotalRT", totalRT

Hope that helps,
Tony

Anthony P. Zuccolotto
Vice President of Operations
Psychology Software Tools, Inc.
2050 Ardmore Boulevard
Suite 200
Pittsburgh, PA 15221-4610
Phone     412-271-5040
FAX       412-271-7077
Email     anthony.zuccolotto at pstnet.com
Internet  http://www.pstnet.com/


-----Original Message-----
From: Kai [mailto:kai.osterberg at ymed.lu.se]
Sent: Monday, January 07, 2002 11:12 AM
To: eprime at mail.talkbank.org
Subject: Using visual-analogue scales in E-prime


Dear e-prime users,

Using e-prime, we are trying to program a 20-item questionnaire with
visual-analogue scales (VAS) for the responses. This will require the
movement of an arrow along a line (from 0 to100 percent) in response to
the question; e. g., the stimulus "Do you feel tired ?" would be
presented at the top of the screen, and below would be presented a 0-100
percent response line (with labeled endpoints) and a moveable arrow. The
subject would move the arrow back and forth along the line (preferably
by using the SRBox keys no 1 and 5) in order to fine-tune its position
until it is perceived as represntative for his/her current state. The
subject would then confirm/enter this response by pressing a separate
key, and the next question would appear (and so on).

Problem: We are able to achieve the movement of an arrow along a line,
but only by setting up each small step of arrow movement (i. e. each
key-press) as a separate reaction time (RT) stimulus-response (S-R). But
this leads to serious problems: If the scale resolution is set to 2
percent (0-100% = 50 steps), already moving along the scale once leads
to 50 RT S-R:s. Together with some margin for additional movements, the
number of RT:s must reasonably be set to at least 300 for each question!
This would of course make a 20-item questionnaire extremely large (not
to mention the enormous amount of unwanted data obtained). Obviously
this approach is incorrect. We have not succeeded to solve the riddle of
how make the arrow move back and forth along the VAS-scale in response
to keyboard inputs without making e-prime believe that each arrow
movement represents a RT S-R event. The only two response data of
interest to us is the position of the arrow when the subject makes his
FINAL confirmatory key-press and the decision time.

We would welcome any suggestion of how to properly program this type of
VAS-question within e-prime.

Thanks,

Kai Österberg, PhD
Dept. Occupational and Environmental Medicine
Lund University
SE-22185 Sweden

kai.osterberg at ymed.lu.se
phone +46-46-177292