Hi,

I'm a PhD student, who's used the IAT to look at attitudes to mental vs physical illnesses. I created the IAT by modifying the insects/flowers one. I've had 280 participants complete it, but now am having problems analysing the data in SPSS. When I try to run the SPSS syntax, an error message appears.

(I also haven't figured out yet how to get data from one participant onto one line in SPSS- my data is over several columns and rows at the moment, and I'm only looking at data from one participant at a time).

I am a psychologist, and my PhD is in medical education, so I have limited in-depth SPSS knowledge (basic stats are about it). I've read a lot of posts on the forum but didn't understand much of the computery talk!

Any help would be much appreciated, as I am getting very stressed out with the IAT at the moment!

Thanks

Gemma  

I see several issues here:

(1) IAT template and matching SPSS syntax

The first thing to clarify is which IAT template exactly you used.
There are a bunch of them around:

- Tony
Greenwald's (somewhat outdated) template available from http://faculty.washington.edu/agg/iat_materials.htm
along with a matching SPSS
syntax file.

- The
Inquisit 3 templates available from http://www.millisecond.com/download/samples/v3/IAT/default.aspx
along with SPSS command syntax.

- The
older Inquisit 2 IAT templates available from http://www.millisecond.com/download/samples/v2/default.aspx
along with a variety of SPSS syntax files for the different templates.

Note that the SPSS files are specifically tailored to their
respective IAT template, they cannot be used interchangeably (at least
not straight out of the box). Also, the various templates differ
considerably in terms of the data they put out (e.g. the Inquisit 3
templates automatically compute the D score and write it to the data
file whereas the others do not).

Also, if you've made any
substantial changes to the procedure as implemented by the original
template (e.g. reduced the number of blocks), there's a fair chance that
the respective SPSS syntax file won't work anymore / would have to be
adapted for the modified procedure.

BTW, here's an excellent
IAT / D score / Inquisit overview written by Sean: http://www.millisecond.com/forums/Topic3444.aspx.

(2) Dealing with SPSS error messages

While I'm personally not a huge fan of the software, the error
messages spit out by SPSS are usually pretty informative. So you might
want to take a closer look at said error message(s), look up additional
information on the reported errors in the SPSS documentation or on the
internet and (if all of this doesn't help) post the specific error
messages here or in one of the many SPSS related forums (official SPSS
support or user-driven places like http://www.spssforum.com).

Regarding (1) and (2), this thread might offer some hints: http://www.millisecond.com/community/forums/t/1034.aspx.

(3) Dealing with reaction time data / Aggregating reaction time
data

Ron Dotsch has an excellent, easy to follow tutorial available right
here (and it even uses IAT data as an example): http://web.me.com/rdotsch/tutorials/inquisit2spss/index.html.
As a general introduction to preparing RT data for analysis in SPSS I
would like to recommend working through the tutorial article by Lacroix
& Giguère
(2006) which is available here along with a sample data set: http://www.tqmp.org/doc/vol2-1/vol2-1.htm.

(4) Computery talk

Well, if you're doing computery research you'll have to cope with at
least some computery talk...;-) Can't have one without the other -- that
would be like wanting to eat a tasty steak without (someone) having to
kill a cow first. No wonder I'm on a vegetarian diet...

Just my 0.02$. Thanks for reading and good luck!

Best wishes from a fellow Inquisit user,

~Dave

Sorry, forgot one more thing. If you're unsure about the statistical analysis you need to conduct or just can't get your head around the data preparation stuff: Get someone experienced with RT data analysis at your faculty to help you -- that's not a shame. It really isn't.

~Dave

Hi Dave

I looked into your advice above and to Sean's post. I downloaded the syntax but it's all gibberish.

Is there a step by step guide on using the syntax on SPSS, and importing the IAT data into it?

So, I have the IAT data. How do I import it to SPSS?

Sorry, I'm not a researcher. I am a clinician doing research, so I'm limited on computer programming slang.

I would appreciate any guidance here.

regards

Miguel

I'd recommend you work through a decent beginner's guide to SPSS, either online or in book-form.

Regards,

~Dave

Hi Dave

it's not with SPSS, I have managed to open the IAT.dat file on SPSS, and I have the variables there, but where does the IAT syntax come to play? I don't get it!

What use is the syntax for analysis?

Cheers for your help.

Miguel

The SPSS syntax aggregates the raw data and computes summary measures, which in turn may be submitted to further data analysis (cf. Lacroix & Giguere, 2006, for a general introduction). If you're only interested in the IAT D-score, you don't necessarily need the SPSS syntax, you may just use the score as computed by Inquisit and contained in the original data file(s).

Regards,

~Dave

oh.. brilliantly put Dave. Makes sense.

Thanks for that.

Miguel

Hello,

I have 80 cases in a data file using the Self-Esteem IAT script I downloaded from the Inquisit task library. I'm trying to start my analysis, and I've been reading the threads on the forum about how to proceed. The problem is that when I try to run the syntax (downloaded from Inquisit), I get a number of error messages about file definition, value labels and failure to open the file. I followed the instructions at the beginning of the syntax to re-name my data file IAT.dat, but that's all I did before attempting to run the syntax. Are there other steps that I need to take with the file before I re-name it? I'm sorry for the elementary question. I too am a PhD student with only basic stats knowledge, learning as I go along with this project. I'd appreciate the help.

Best,

Kathryn

Current SPSS versions may require you to enter full file paths. As for error messages, you need to follow up on them via dedicated SPSS resources (i.e, the SPSS documentation, etc.). It might be worthwhile to first get more familiar with SPSS in general as well as its syntax facilities in particular. There should be  no shortage of books, online tutorials, etc. on these topics.

Regards,

~Dave

Hi Dave,

Thank you for your quick reply. If I somehow identify a better file path, is the data, as generated by Inquisit, in the correct format for spss and the Inquisit syntax file, or does it need to be adjusted or manipulated in some way first? In other words, have people using previous versions of SPSS been able to just rename the Inquisit data file and run the syntax?

Thanks,

Kathryn

SPSS's requirement for full file paths has nothing to do with the data
file format. SPSS will read the data file as output by Inquisit just
fine, no additional preprocessing or format changes are needed.

Thanks, Dave. I changed the file path and successfully got the data into SPSS.

Hello,

I'm trying to determine whether to exclude several slow responders from my data set (who have some responses >10,000 ms). How is the aveltncy variable calculated? Does it take into account the latencies on all trials, including practice? When I run means on the cases the numbers aren't the same as the value for this variable.

Thanks very much.

Kathryn

Just look at the SPSS syntax:

* Use data for Blocks 3,5,7, and 9 (i.e., TEST = 1 or 2) .
SELECT IF (TEST=1 OR TEST=2) .

[...]

* Record criteria for potential use in subject discards .
* This count can later be used as the basis for subject discard .
AGGREGATE OUTFILE = 'CRITERIA.SAV'
 / BREAK = SUBJECT ORDER
 / PCT_300 = PGT(FLAG_300,0)
 / PCT_400 = PGT(FLAG_400,0)
 / PCT_10K = PGT(FLAG_10K,0)
 / AVELTNCY ERRORPCT = MEAN (LATENCY ERROR)
 / NTRIALS = N .

* Drop trials slower than 10000 ms for LATENCY .
SELECT IF (LATENCY LE 10000) .

[...]

* AVELTNCY = overall average latency .
* [Note that 'aveltncy' was computed _BEFORE_ discard of latencies above 10000 ms] .

Regards,

~Dave