Millisecond Forums » Millisecond Forums » Inquisit 3 » Compute separate D measures for two IATs that work with "include" file

Compute separate D measures for two IATs that work with "include" file

Sonja — Mon, 02 May 2016 14:35:09 GMT

Dear all,

I would like to run two IATs per person in sequence and thus wrote an "Include" data file which is working just fine. Except for one thing: I realized that since I kept all the <values> and <expressions> data in the "include" data file, Inquisit only computes one D measure for the two IATs. However, I need a D measure for each of the two IATs. I have tried to delete the <values> and <expressions> commands from the "Include" file and to write these commands preceded by "iat1" and "iat2", respectively (just like you need to do with many other commands), but Inquisit does not run this script properly. Can anyone help me on this? I would truly appreciate this!

Thanks and regards

Sonja

RE: Compute separate D measures for two IATs that work with "include" file

Dave — Mon, 02 May 2016 14:35:09 GMT

Yes. At the end of IAT #1 store the D-score in a separate <values> entry. Do the same at the end of IAT #2. Log both values to the summary data file.

RE: Compute separate D measures for two IATs that work with "include" file

Sanneve — Mon, 02 May 2016 13:19:50 GMT

Hi!

I'm also combining two IATs in one experiment. I was wondering if it is possible to show an overall D-score of IAT1 and an overall D-score of IAT2 in the summarydata file?

Thanks in advance.

Regards,
Sanne

RE: Compute separate D measures for two IATs that work with "include" file

Dave — Sat, 14 Apr 2012 11:16:34 GMT

Working fine for me. I recommend removing and re-installing the web control as per https://www.millisecond.com/support/uninstalljava.aspx.

RE: Compute separate D measures for two IATs that work with "include" file

Jimmy Page — Sat, 14 Apr 2012 10:44:51 GMT

Dave,

Thanks for the clarification. I have finished the script. It works fine when I run on my desktop, but it crashes 98% of the time when I run on the web. I have tried to run on different computers and browsers, but it starts and then takes me to the script's last page. I have cleaned the cache, updated my browser, and done everything else you can think of. Any help would be greatly appreciated.

Launch page: http://research.millisecond.com/bwalker/college.web

Files: https://www.dropbox.com/sh/te39y7aw9mdkx4m/vhhJxaM3Q7

RE: Compute separate D measures for two IATs that work with "include" file

Dave — Sun, 08 Apr 2012 11:10:36 GMT

[quote]So, my question is this: Should I add the onblocked element anyways?[/quote]

If you want to have separate Ds for the two IATs, yes. See the first reply in this thread. The point is that you obviously must reset the various sums, etc. used to compute D to zero after the 1st IAT finishes and before the 2nd IAT starts.

[quote]However, I cannot tell whether the two scores are independent from each other. Is there a way to find out just by looking at the data file?[/quote]

Check if the relevant expressions recorded in the data file start at 0 at the for the 2nd IAT.

[quote]Should I change "values.magnitude" to "values.iat1magnitude"? I did change, but I got an error.[/quote]

Why would you want to do that? (The instructions in the respective documentation topic do not suggest to do so either, as far as I'm aware).

Regards,

~Dave

RE: Compute separate D measures for two IATs that work with "include" file

Jimmy Page — Sun, 08 Apr 2012 10:23:19 GMT

Hi Dave,

I have a similar script (please see files attached), but I did not delete the summary section. So, my question is this: Should I add the onblocked element anyways? I run my script and it seems that it did record two separate d scores. However, I cannot tell whether the two scores are independent from each other. Is there a way to find out just by looking at the data file?

Question 2: I did follow the instructions found here to combine multiple scripts using the include method. Should I change "values.magnitude" to "values.iat1magnitude"? I did change, but I got an error.

Thanks a lot for your help.

J.Page

RE: Compute separate D measures for two IATs that work with "include" file

Dave — Tue, 21 Jun 2011 05:07:15 GMT

[quote]Does a tutorial or anything exist how to program the d-score algorythm in inquisit? I would like to understand a little bit better what I use :-)[/quote]

Read Greenwald et al. (2003) for all the details on the D scoring algorithm. Then look at the respective <values> and <expressions> in the Inquisit script. It's rather basic stuff, really. Also see https://www.millisecond.com/forums/Topic3444.aspx#3444 and https://www.millisecond.com/community/forums/p/928/3204.aspx#3204.

RE: Compute separate D measures for two IATs that work with "include" file

inquisitor — Tue, 21 Jun 2011 05:02:21 GMT

Thanks a lot!

I followed your advice and created a Block "clear":

<block clear>
/ onblockend = [values.sum1a=0; values.sum2a=0; values.sum1b=0; values.sum2b=0;
    values.n1a=0; values.n2a=0; values.n1b=0; values.n2b=0;
    values.ss1a=0; values.ss2a=0; values.ss1b=0; values.ss2b=0;
    values.magnitude="unknown"]
</block>

I think the data recording works as it should now, because there are a few clear lines in the data sheet between the first and the second iat (e.g. attributepractice and expressions.db). They were not without the clear-block.

Does a tutorial or anything exist how to program the d-score algorythm in inquisit? I would like to understand a little bit better what I use :-)

Thanks again!

Arne

RE: Compute separate D measures for two IATs that work with "include" file

Dave — Tue, 21 Jun 2011 03:20:35 GMT

[quote]

I just deleted the summary-trials and -blocks, and left everything as it has been in the original script. For each ST-IAT I changed a) the trial-labels b) the block-labels (except the attributepractice-block) c) the item-labels and d) the variables

I changed nothing in the values- or expressions-elements.

Did I do anything correctly? I want to start my main study next week and it would be terrible if the d-score-recording would not work.

[/quote]

You need to reset a bunch of values *between* the two IATs, i.e., after the 1st one finished and before the 2nd one starts. See my previous post in this thread. Since you deleted the summary block (which would have been suitable for this purpose, because it sits right between the two IATs), you need to introduce some kind of 'reset' block and/or trial element and run it between the two procs.

Regards,

~Dave

RE: Compute separate D measures for two IATs that work with "include" file

inquisitor — Tue, 21 Jun 2011 02:40:48 GMT

Hey,

I have encountered a similar problem, that's why I don't want to create a new topic.

In my study, I'm also running two ST-IATs in one script and it seems that everything works fine. But since I don't present the d-scores as feedback to the participants of my study, I'm not able to test if the recorded d-scores are the right ones.

I changed nothing in the values- or expressions-elements.

Did I do anything correctly? I want to start my main study next week and it would be terrible if the d-score-recording would not work.

Thanks in advance and regards

Arne

RE: Compute separate D measures for two IATs that work with "include" file

Sonja — Mon, 13 Dec 2010 22:12:34 GMT

Hi Dave,

thanks so much, it works! [:D]

Regards,

Sonja

RE: Compute separate D measures for two IATs that work with "include" file

Dave — Mon, 13 Dec 2010 15:27:43 GMT

Add

/ onblockend = [values.sum1a=0; values.sum2a=0; values.sum1b=0; values.sum2b=0;
    values.n1a=0; values.n2a=0; values.n1b=0; values.n2b=0;
    values.ss1a=0; values.ss2a=0; values.ss1b=0; values.ss2b=0;
    values.magnitude="unknown"]

to <block summary> and you should be all set.

~Dave