Randomised sub-lists

[David McFarlane]

BTW, in case you did not already know, here is how you set your own 
random seed in E-Prime:  Open the Experiment object properties and go 
to the Startup Info tab.  Add a Startup Info Parameter called 
"RandomSeed", making sure that it is Numeric.  Now when you start the 
run E-Prime will prompt you for the RandomSeed and use the entered 
value for the random seed.

Note that PST documents this as Knowledge Base article #882, "INFO: 
Ability to set random seed in E-Studio using RandomSeed Startup Info 
parameter".  In principle you might also figure it out yourself from 
studying the Main() code generated by E-Studio.

-- David McFarlane, Professional Faultfinder


At 4/22/2010 02:42 PM Thursday, David McFarlane wrote:
>H,
>
>I take it that you want randomized sequences that will continue 
>across separate experimental runs.  That's a tall order for any 
>system, as it would have to preserve and reuse information between 
>sessions.  As Ben says, you would first have to reuse the same 
>random seed between runs.  That alone would give each subject who 
>uses the same random seed the same sequence.  If you want to have 
>one subject run a random sequence of 60 items out of 360, and then 
>the next subject run the next 60 items from that same random 
>sequence, then you have to to a little more.  I think that I would 
>start with reusing the random seed as already posited, and in 
>additon use the Offset order for my List to start with the 61st item 
>from the random sequence, or 121st, etc.  Just a thought.
>
>You might also consider asking PST Web Support 
>(http://support.pstnet.com/e%2Dprime/support/login.asp ) for ideas, 
>they have come through for me on issues like this from time to 
>time.  For all I know the new StartupInfo file mechanism of EP2 
>could help you with this.
>
>-- David McFarlane, Professional Faultfinder
>
>
>At 4/21/2010 01:43 PM Wednesday, ben robinson wrote:
>>to do this you will probably need to manually assign a value for 
>>the Random Seed for each 6 participants you see.
>>the first 6 participants might be assigned a random seed of '1', 
>>the next six '2', etc.  if you set it up properly your list of 360 
>>stimuli will be shuffled according to the random seed value.
>>you could then determine which subset of 60 items, drawn 
>>pseudorandomly from the larger set of 360 items, to present to your 
>>subject based on Subject Number.  divide the the subject number by 
>>6, take the remainder which will always be a value from 0-5, and 
>>based on that value present one of six different subsets.
>>something like that.  sorry if it's not very clear...
>>
>>On Wed, Apr 21, 2010 at 11:53 AM, Hester 
>><<mailto:hester.duffy at gmail.com>hester.duffy at gmail.com> wrote:
>>Hi All,
>>
>>I'm trying to set up a study in which participants rate the stimuli
>>used in another experiment. I have 360 stimuli in all, but I only want
>>each participant to rate 60 of them, and I want to do this in such a
>>way that each of the stimuli gets used the same number of times.
>>Obviously I could just create 6 lists of 60 items each and use
>>different run files each time, but I'm wondering if there's a more
>>efficient and elegant way of doing it.
>>
>>I've tried setting the list to reset after a cycle of 360 samples, and
>>to exit after 60, but having run it three times, and then compared the
>>lists of stimuli used, I'm getting duplicates, so it's clearly not
>>doing what I want. Basically, I need it to give the first participant
>>a random list of 60 taken from the full list of 360; the second
>>participant should then get a random list of 60 taken from the
>>remaining 300, the next a random list of 60 from the remaining 240,
>>and so on, until the sixth participant has a randomly-ordered list
>>consisting of the remaining 60 items. It would then reset, so that the
>>seventh participant gets a random list of 60 from the original 360
>>(but not necessarily the same 60 as participant 1).
>>
>>Does anyone know whether E-Prime can do this, and if so, how?
>>
>>Many thanks,
>>
>>H

-- 
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 this group at http://groups.google.com/group/e-prime?hl=en.