SPM-2 data collection update
John Williams
Hi group,
As promised here is a summary of our data collection: so far we have about 30% of our total standardization data in. I am hopeful that we will pick up speed from here and get to 50% early in 2018. We will send another update after the first of the year.
Also, thanks to Kailey Bax, copied here: she wrote the summative and demographic syntax for all those forms !
John
John C. Williams, PhD
Senior Project Director
Licensed Clinical Psychologist
t 800.648.8857 or 424.201.8800
625 Alaska Avenue, Torrance, CA 90503
ateac...@aol.com
Is this for each form?
--
You received this message because you are subscribed to the Google Groups "SPM-2 Project Workspace" group.
To unsubscribe from this group and stop receiving emails from it, send an email to spm-2-project-wor...@googlegroups.com.
To post to this group, send email to spm-2-proje...@googlegroups.com.
Visit this group at https://groups.google.com/group/spm-2-project-workspace.
For more options, visit https://groups.google.com/d/optout.
John Williams
The 30% is an average across the entire standardization sample: some forms have more cases and some have less. We’ll look at selecting items on the forms in the new year as they hit the 50-65% mark.
As far as pitching in, just continuing to let folks know about the data collection opportunity is a big help.
Thanks!
John
ateac...@aol.com
John Williams
I’m not concerned – the fact that we have 12 forms that have to hit precise demographic targets makes this data collection a little slower than average. It’s a lot of cases: in terms of data collection, we’re not revising a test so much as we’re revising two tests, and creating three new ones !
In terms of results, we probably will not have a complete set of anything in the first quarter of 2018: the results are going to be in different phases for different forms. But frankly, reliability and validity characteristics really are a certainty for a test like this, and even if you wanted to present the results for all the SPM-2 forms , they would likely take up many more tables than you want to talk about. Maybe you could talk more about the procedures we used for the expansion. Why don’t we revisit this when we start selecting items in the new year ?
John Williams
Hi all,
Based on your recent email this may come as a surprise, but as I reported to you in November, there are no SPM-2 data sets that will be far enough along to produce numerical results for your talk. The reason remains the same: we have not yet collected enough data to select the items for the forms, and without forms to analyze, no results of that kind are possible. I thought we were all on the same page on this three months ago – if not, I apologize for not communicating this more clearly.
Some possible ideas for your talk,, if you still want to talk about psychometrics, you might try some of the following:
- Summarizing the psychometric properties of the SPM and SPM-P, which is itself a preview of the SPM-2 because we expect SPM-2 results to be more or less the same.
- Talking about SPM-2 expansion, particularly the process you used to develop the new items – how you adapted preschool items down to infants and toddler; child items up to teens; caregiver items to adult; parent report to self-report – perhaps presenting a table of comparable items across the age ranges.
- Discussing content validity: how SPM-2 item writing was based on the same theoretical principles as the prior version; how items were piloted and then selected for standardization based on each item’s psychometric properties and it’s place within the facet structure of the test as a whole.
- Talking about the different types of reliability and validity that we’re building into SPM-2, and why they’re important (basically Dave’s talk minus the tables). People would probably get a lot more out of a discussion of those principles than they would from slide after slide of tables.
Hopefully some of those ideas will work for you. In thinking about future talks of this kind, I would encourage all of you to not worry too much about presenting proof of psychometric soundness. This is a revision and expansion of a well-received flagship instrument produced by a commercial test publisher. No one expects the psychometric properties of subsequent versions to be anything other than perfectly adequate. Additionally, I see no need to produce a talk like Dave’s 2007 SPM presentation when SPM-2 is published. Given the size of SPM-2, a talk like that would run to at least 60 slides, all of that information is included in the manual, and it would leave most people either bored or confused. We should focus on how the test works, when and how to use it, and all the things you can do with the results.
Let me know if you have any questions about this…
John
John C. Williams, PhD
Senior Project Director
Licensed Clinical Psychologist
t 800.648.8857 or 424.201.8800
625 Alaska Avenue, Torrance, CA 90503
wps
unlocking potential
The information contained in or transmitted with this e-mail may be privileged and/or confidential. If you are not the intended recipient, you are advised that any dissemination or use of this communication is strictly prohibited.