Effect sizes in single case research

Ask here about academic and research issues, like designing studies, recruiting participants, choosing statistics, submitting for publication, etc.
Post Reply
srlb2
Posts: 125
Joined: Mon Feb 27, 2012 6:43 pm

Effect sizes in single case research

Post by srlb2 » Wed Apr 23, 2014 3:55 pm

Hi all,

As many of you will probably know (I post far too many topics about this!) I'm doing a meta-analysis of single case research as part of my PhD. I've manage to overcome a lot of the issues that have come up about this kind of thing (I'm starting to understand why people don't do these types of studies often!) however I'm really stuck on how to calculate effect sizes in single case research designs. The more I research, the deeper I get (parametric/nonparametric, autocorrelation, regression models, confidence intervals.... :-| ) and I'm getting really confused and seem to be getting multiple headaches as a result! I usually consider myself to have a good grasp on statistics but this is so much more complicated and involved than anything I've ever come across before!

Has anyone had any experience in calculating effect sizes in single case research designs? Or alternatively, does anyone know of a decent text which explains this? Every article/book I find on the topic seems to assume a lot about the reader's knowledge and is impossible for me to understand without researching lots of other topics which take me further away from what I actually need to know!

I've also been in contact with the statistics consultancy desk at my university but they're closed for a month now which is not helpful!

Any advice or ideas would be appreciated!

A very stressed,
srlb2

User avatar
sarahg
Posts: 614
Joined: Sat Dec 10, 2011 1:46 am

Re: Effect sizes in single case research

Post by sarahg » Wed Apr 23, 2014 4:56 pm

Hello,

I personally do not know the answer but I often use a wide range of forums to gain statistical advice, which you might like to try. Perhaps provide them with an example that you are working with.

I use:
http://statsmodels.sourceforge.net/
and
http://www.talkstats.com/forumdisplay.p ... stics-Help

srlb2
Posts: 125
Joined: Mon Feb 27, 2012 6:43 pm

Re: Effect sizes in single case research

Post by srlb2 » Wed Apr 23, 2014 4:59 pm

Thanks Sarah - that's really helpful!

For some reason I hadn't thought to seek out statistics forums!

User avatar
sarahg
Posts: 614
Joined: Sat Dec 10, 2011 1:46 am

Re: Effect sizes in single case research

Post by sarahg » Wed Apr 23, 2014 5:15 pm

That's okay!

As a heads up, you get a better response if you provide something numerical for them to use!

User avatar
Borrowed Cone
Posts: 1269
Joined: Fri Nov 12, 2010 11:05 pm
Location: M25

Re: Effect sizes in single case research

Post by Borrowed Cone » Wed Apr 23, 2014 8:33 pm

no idea if this is appropriate, but reliable change index calculation seems to be popular in single case studies of change over time?

The Cone
"We can rebuild him. We have the technology. But I don't want to spend a lot of money..."

srlb2
Posts: 125
Joined: Mon Feb 27, 2012 6:43 pm

Post by srlb2 » Wed Apr 23, 2014 9:02 pm

Yeah I came across the reliable change index by accident (I haven't seen it used in single case research but that could just be something to do with the specific research I'm looking at. I've no doubt it is used as it seems like a useful measure). Its on my list of things to look at but as I hadn't seen it used very often I wasn't sure it would be appropriate. I have the paper containing it though so definitely worth a read - thank you!

Sent from my GT-I9505 using Tapatalk

User avatar
Borrowed Cone
Posts: 1269
Joined: Fri Nov 12, 2010 11:05 pm
Location: M25

Re: Effect sizes in single case research

Post by Borrowed Cone » Wed Apr 23, 2014 9:26 pm

No problem. I think it does require that you use standardised measures though... good luck!
"We can rebuild him. We have the technology. But I don't want to spend a lot of money..."

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest