Hi, when you say 'throw in more A's' do you mean repeat the baseline (A) every time before following up with a combination intervention e.g A BC A BD A CD where the A's represent different times another baseline was taken?
Hello, if I implement behavioral skills training (BST) intervention sequentially across 3 teachers (experimental control) and continued to assess performance according to a 10 step task analysis; in which case, I would have recorded Baseline, intervention (BST), and post-training data for each, can I safely say this is a multi-element design (so I think... please correct me if I'm wrong)? Also, in graphing the data for all 3 participants, I should have 3 different independent graphs, which would not be liked for comparison, unlike the Multiple Baseline Design. I'll appreciate your guidance. Thanks
Well you would have little to no experimental control at all. What you have described from what I can read is three AB designs. Baseline (A) and intervention (B) for three subjects. AB designs have basically zero internal validity. The multiple subjects is good - but the way you have described this study shows no control. You COULD overlap baseline measures for all three subjects concurrently then start your intervention at different times for each subject thereby producing a multiple baseline (see our multiple baseline part deux vid) to establish experimental control. This would be awesome. It is not a changing criterion design. With those you establish a baseline then switch conditions to intervention. During the intervention you choose a new level for a given behavior and when the subject reaches it with consistency (e.g., 3 out of 5 times or 4 times in a row, whatever) then you change the criteria to be closer toward your goal. We have some good vids for most if not all of these and Cooper covers them well too. If you have a Johnson and Pennypacker text they do an amazing job of presenting experimental design.
The effect of one intervention carries over into the next (b carries to c) It’s a problem if you don’t go back to baseline after each new intervention type. Because then you cannot tell which intervention is the one having the effect (did c change behavior or was it the effect of b carrying over to c?).
@@PsychCore Ahh, so going back to baseline is important so you dont get this problem! I see now! Thank you for putting it in to my terms!! And thanks for the videos!!!
Im in school for behavioral science and I've really enjoyed your videos. thank you!!!
Hi, when you say 'throw in more A's' do you mean repeat the baseline (A) every time before following up with a combination intervention e.g A BC A BD A CD where the A's represent different times another baseline was taken?
Pretty much!! There’s lots of potential carryover effects if you don’t balance everything out...
Hello, if I implement behavioral skills training (BST) intervention sequentially across 3 teachers (experimental control) and continued to assess performance according to a 10 step task analysis; in which case, I would have recorded Baseline, intervention (BST), and post-training data for each, can I safely say this is a multi-element design (so I think... please correct me if I'm wrong)? Also, in graphing the data for all 3 participants, I should have 3 different independent graphs, which would not be liked for comparison, unlike the Multiple Baseline Design. I'll appreciate your guidance. Thanks
Thinking about it, I reckon it's a changing criterion design.... again, please correct me if I'm wrong
Well you would have little to no experimental control at all. What you have described from what I can read is three AB designs. Baseline (A) and intervention (B) for three subjects. AB designs have basically zero internal validity.
The multiple subjects is good - but the way you have described this study shows no control.
You COULD overlap baseline measures for all three subjects concurrently then start your intervention at different times for each subject thereby producing a multiple baseline (see our multiple baseline part deux vid) to establish experimental control. This would be awesome.
It is not a changing criterion design. With those you establish a baseline then switch conditions to intervention. During the intervention you choose a new level for a given behavior and when the subject reaches it with consistency (e.g., 3 out of 5 times or 4 times in a row, whatever) then you change the criteria to be closer toward your goal.
We have some good vids for most if not all of these and Cooper covers them well too. If you have a Johnson and Pennypacker text they do an amazing job of presenting experimental design.
Could you explain carryover effects?
The effect of one intervention carries over into the next (b carries to c) It’s a problem if you don’t go back to baseline after each new intervention type. Because then you cannot tell which intervention is the one having the effect (did c change behavior or was it the effect of b carrying over to c?).
@@PsychCore Ahh, so going back to baseline is important so you dont get this problem! I see now! Thank you for putting it in to my terms!! And thanks for the videos!!!
Effect vs necessary components
Pairwise with this design