Millisecond Forums

Hi Dave, I have only one last question.

The IAT is making compatible and incompatible matching but not in the same order every time. My question is, that this is calculated in the last d score, or i have to make some changes with the reversed ones.

Thanks in advance,

Martin

The order of compatible / incompatible blocks is counterbalanced between subjects based on the numerical subject id -- see the <variables> element for details. This does not affect the computation of d in any way.

Hey Dave,

I've read in another thread that having a D score that is higher than 1.0 is rare, however, I seem to have 4 instances of it in my data. 

Any ideas why this might be?

Thanks

Rare does not mean impossible. D may vary between -2 and +2 in theory. Other than that, I have nothing to offer on this subject, I'm afraid. Might be your particular subject matter, etc. Those are theoretical and methodological questions, so you should be better equipped than anybody else to answer them (since you know your particular area of research best).

Cool. I just wanted to check that it wasn't a technical issue. 

Thanks 

Dear Sean,

Can I just check that from having read the above (very helpful) description and having translated that to the Single Target IAT that I have understood the expression.d score for the Single Target IAT correctly:

A positive score would mean there was an association between Target A and Attribute A

A negative score would mean there was an association between Target A and Attribute B

So with you necktie example. Lets say Target A was neckties, Attribute A was pleasant and Attribute B was unpleasant. A positive score would mean the participant liked neckties, a negative score would mean that they did not like neckties? With my study Target A is 'disabled', attribute A 'pleasant' and attribute b 'unpleasant'. So a positive score would mean that participants associated 'disabled' with 'pleasant' (thus positive implicit attitudes), and a negative score would mean that participants associated 'disabled' with 'unpleasant' (thus negative implicit attitudes) right?

Many thanks,

Michelle

Yes.

Hi,

I've used one of your IAT scripts as my implicit attitudes' instrument, and then, to analyse my data, I also used the SPSS syntax that you also provide in your website. However, I've been struggling with it, since I am not an experienced user, and would like to clear some questions.

Before importing my .dat file to SPSS, I introduced some alterations to it because:

1) I had to discard two participants for not meeting my eligibility criteria (to avoid descriptive statistics based in an erroneous N). and,
2) I made a mistake when introducing one participant's ID before running Inquisit... leaving me with two participants with the ner. 48 ... So I recoded this participant to the next available even slot (n.er 52 - it's a small sample), since IAT counterbalanced participants according to that property (odd or even ID number).

After this I tried to import my data to SPSS and, in the last step from the importing procedure, I've chosen to paste the syntax (and I've indicated the respective file path in it too). I got one Output window, a .sav data file based on my IAT's raw data, and an aggregated file, with all the relevant data necessary to calculate D index (and the actual index)/ per participant.

My results only revealed negative D_biep values. And now... my doubts and the problems found...

1 - The doubt.

- to verify the correctness of the data obtained, I calculated the D_Biep index of some of the participants by hand. I noticed that for participants with an even number ID, the values obtained - although the same in module to the ones indicated in SPSS - were positive instead of negative; and that didn't make sense given that the mean latencies and respective differences indicated negative implicit attitudes toward my main target of interest... I then remembered that maybe there's some "built-in recoding" in IAT's syntax (but I couldn't really find it there - I'm a rookie in this so...). I just need you to confirm (or not) this idea:

In what concerns even numbered ID's (order 2), the block order is "incompatible first", instead of "compatible first" so, in order the data to make sense, the syntax recodes the blocks' order, as if they had been ran in order 1.

According to the algorithm used:
Mean latency Incompatibletest 1 (Block 5 when "compatible first") - Mean latency Compatible test 1 (Block 3 when "compatible first");

So, when in order 2,

Mean latency Incompatibletest 1 (Block 3 when "incompatible first") - Mean latency Compatible test 1 (Block 5 when "incompatible first").
And this is how we get a negative value for this operation…

Is this right? If so, what are the following code lines for:




2 - And now, the problems (mixed up with some other doubts)…

As already stated above, after I ran the SPSS syntax I obtained one Output window, a data file based on my IAT's raw data, and an aggregated file, with all the relevant data necessary to calculate D index (and the actual index)/ per participant.

I haven’t altered anything in SPSS syntax besides my file name, and haven´t altered anything in IAT script besides deleting the summary block from the script's <expt> elements' /blocks attributes (with your help), so that the participants didn’t receive feedback about their performance.

After analysing the output file, it revealed several errors… One of which prevented the respective variables flagging the <300ms, <400ms and >10 000ms trials from appearing in the respective .sav file(for every trial), although their descriptive statistics appear in the mentioned output file (I attached a file with it, in order to make myself clearer (the errors are highlighted in yellow).

What has happened? Why are these errors occurring? I’ve tried to overcome this “N times” :p and haven’t reached any conclusion… Is it due to the “holes” in my sample (there aren’t participants ner 32 and 46, for they didn’t meet my eligibility criteria and I deleted them before running the script), the "recoding" of the doubled 48 participant, or any other alteration that I have made in my raw data file?

Please help… I don’t know what else to do!

Thanx,
Ana

Re. #1 the first thing you need to check is *how* your IAT script assingns the different block orders: (1) Based on subjectnumber (odd vs even) OR (2) based on groupnumber (odd vs even).

Adjust the SPSS syntax accordingly.

If you study the SPSS syntax you will find that there's a variable computed that reflects the block order (called ORDER). Verify that variable is computed and used correctly. If it is not, the *sign* of the D-score will be wrong.

As for the specific syntax snippet you asked about, you will find this covered in this very topic https://www.millisecond.com/forums/Topic3444.aspx :
" A positive score indicates an association of targetA with attributeA  and targetB with attributeB.  A negative score indicates an association of targetA with attributeB and targetB with attributeA. Translating the score into a preference or attitude thus depends on how you have assigned your real world categories to these 4 groups. You can determine the mappings by looking at the topmost section of your IAT script, where you'll see the <item> definitions for each category."

If your mappings are not consistent with the above, the scoring must be reversed. That's what that syntax does if enabled.


Re. #2: The error message is very explicit.

* Save data to re-use for computing other measures .
SAVE OUTFILE = 'temp.sav' .
>Error # 5332 in column 16. Text: temp.sav
>The specified file or directory is read-only and cannot be written to.
>The file will not be saved. Save the file with another name or to a different
>location or change the access permissions first.
>Execution of this command stops.

You are trying to save 'temp.sav' to a location on your computer where you do not have write permissions. This will be the case for e.g. SPSS's program directory (which is its default working directory). Edit the SPSS syntax and supply a file path where you do have write permissions.


The same applies to all the remaining files written by the syntax, too. You *must* provide full, proper file paths to a location where you may write to under your user account (e.g. your Desktop folder, your "My Files" directory, etc.).

Hi Dave. Thank you very much for your reply!

Re. # 1 - Where can I see how the IAT script assigns the different block orders (I used the Picture IAT Millisecond provides for download)? And, more importantly, how do I adjust the SPSS syntax accordingly? :s

I already knew about what had been written in this topic in the link  https://www.millisecond.com/forums/Topic3444.aspx. I am sorry, I forgot to give you the information regarding mi IAT script.

There's a copy of my final .sav aggregated data file (with D_Biep) attached (so that I can better illustrate what I said previously in the other reply).

 My IAT script includes:

<item attributeAlabel>
/1 = "Good"
</item>
(…)
<item attributeBlabel>
/1 = "Bad"
</item>
(…)
<item targetAlabel>
/1 = "Persons with Disabilities"
</item>
(…)
<item targetBlabel>
/1 = "Persons without Disabilities"
</item>
(…)

_______________

 According to my SPSS syntax,

/ ORDER
    1 'compatible first'
    2 'incompatible first'.

So,
Order 1 = “Persons with Disabilities” (Target A) + “Good” (Attribute A)| “Persons without Disabilities” (Target B) + “Bad” (Attribute B)
Order 2 = “Persons with Disabilities” (Target A) + “Bad” (Attribute B) | “Persons without Disabilities” (Target B) + “Good” (Attribute A)


The rest of the syntax concerning order goes like this:

VALUE LABELS blocknum
 1 'Target practice'
 2 'Attribute practice'
 3 'First pairing practice'
 5 'First pairing test'
 6 'Reversed target practice'
 7 'Second pairing practice'
 9 'Second pairing test' .

IF (MOD(subject,2) = 1) ORDER = 1 .
IF (MOD(subject,2) = 0) ORDER = 2 .

COMPUTE PAIRING = 0.
IF ((ORDER=1) and (blocknum=3|blocknum=5)) PAIRING = 1.
IF ((ORDER=1) and (blocknum=7|blocknum=9)) PAIRING = 2.
IF ((ORDER=2) and (blocknum=7|blocknum=9)) PAIRING = 1.
IF ((ORDER=2) and (blocknum=3|blocknum=5)) PAIRING = 2.

COMPUTE TEST = 0.
IF (blocknum=3|blocknum=7) TEST = 1.
IF (blocknum=5|blocknum=9) TEST = 2.

VALUE LABELS
   TEST 0 'single-task practice' 1 '1st combined block' 2 '2nd combined block'
 / correct 0 'error' 1 'correct' .

VARIABLE LABELS
   correct  "0=error, 1=correct"
 / blocknum "block number"
 / trialnum "trial number"
 / ORDER "order of combined tasks"
 / PAIRING "paired categories"
 / TEST "1st or 2nd combined block".

VALUE LABELS
   PAIRING
    0 'single task practice'
    1 'compatible'
    2 'incompatible'
 / ORDER
    1 'compatible first'
    2 'incompatible first'.
COMPUTE error = 100*(1 - correct).


Can you tell me, based in this new information if ORDER was computed and used accordingly?

If it is, the negative scores indicate that my entire sample holds preference for people without disabilities (therefore, holds also negative implicit attitudes toward persons with disabilities), right?


Re. #2: Error messages

Can it be something like this?: 

C:\Users\Ana\Desktop\temp.sav
C:\Users\Ana\Desktop\criteria.sav


Thank you again for your help!

Ana