In the interest of being a participatory citizen of the scientific community, here I offer some data analysis computer code that I have written for my own use, and that might be of use to others.
The Autism Diagnostic Observation Schedule (ADOS) is a standardized assessment of social interaction and communication used in diagnostic evaluations of individuals with a possible autism spectrum disorder (ASD).
As described by the publisher (WPS):
“This semi-structured assessment can be used to evaluate almost anyone suspected of having autism — from toddlers to adults, from children with no speech to adults who are verbally fluent. The ADOS consists of various activities that allow you to observe social and communication behaviors related to the diagnosis of pervasive developmental disorders. These activities provide interesting, standard contexts in which interaction can occur.”
This GitHub repository contains code to score the ADOS (catalog #W-365 at WPS) using IBM’s SPSS statistical program, generating final ASD classifications for individual cases from raw ADOS item scores.
There are two versions of this code, in separate files:
1. ADOS-SPSS-syntax.sps — used to score ADOS modules 1-4 with the original scoring algorithm
2. ADOS-SPSS-syntax-revised-algorithm.sps — used to score ADOS modules 1-3 with the revised scoring algorithm (see Gotham et al, J Autism Dev Disord. 2007 Apr;37(4):613-27). At the time of writing this code, revised classification cutoffs for module 4 were not available.
For input, this code takes in the SPSS-formatted output from a REDCap database export.
For output, this code generates a final variable (i.e. ados_revisclassif_calc for the revised algorithm) with ASD classifications for each case in the dataset.
This code will calculate a classification based on individual item scores, but also allows for a manual classification (already stored in the database), and will check the two against each other for concordance.
This code expects some variables (i.e. age_calc_yrs_ados, etc), which your database may not have; please consider this code as the starting point on which you can build. Most likely, you will need to modify this code to work for your particular dataset.
This code was written and tested using SPSS version 19. If you find an error, please let me know. Please use this code at your own risk. I can’t guarantee that I’ll be able to help you troubleshoot if a problem comes up.
A version of this code has been used in analysis of data that has been published after peer-review (i.e. here).
License: This is free and unencumbered software released into the public domain. Full license here. For more information, please refer to unlicense.org.
