help
help > RE: Design matrix and contrast for 2x2 design
Aug 7, 2017 11:08 AM | Andrew Zalesky
RE: Design matrix and contrast for 2x2 design
Hi David,
it is possible for the ROIs to be the same for the 1 and -1 contrasts, but the actual edges (connections) should generally be different. You might want to try different thresholds.
Unfortunately, the BCT does not include extensive functionality for statistical testing. You might need to import the network measures computed by BCT into a statistical package that you feel comfortable with and perform hypothesis testing there.
Andrew
Originally posted by David de Wide:
it is possible for the ROIs to be the same for the 1 and -1 contrasts, but the actual edges (connections) should generally be different. You might want to try different thresholds.
Unfortunately, the BCT does not include extensive functionality for statistical testing. You might need to import the network measures computed by BCT into a statistical package that you feel comfortable with and perform hypothesis testing there.
Andrew
Originally posted by David de Wide:
Dear Andrew,
Great, I suspected that was the difference based on the areas, but the -1 contrast of Music was very similar to the 1 contrast of the interaction effect (same ROIs, similar t-scores), so I wanted to make sure.
Regarding the BCT, I've already been working with it and made several loop based scripts (for different thresholds) to generate binary matrices and run several of weighted unidirect/binary undirected metrics.
However, with regards to statistical testing, I've only found a way to compare two conditions (i.e. Drug v Placebo or Music v No music), but have been unable to determine how/if to test for an interaction effect using all four conditions. Unlike regular mean scores, I can't subtract the means of the music conditions and then compare those results between drug and placebo.
Sincerely,
David
Great, I suspected that was the difference based on the areas, but the -1 contrast of Music was very similar to the 1 contrast of the interaction effect (same ROIs, similar t-scores), so I wanted to make sure.
Regarding the BCT, I've already been working with it and made several loop based scripts (for different thresholds) to generate binary matrices and run several of weighted unidirect/binary undirected metrics.
However, with regards to statistical testing, I've only found a way to compare two conditions (i.e. Drug v Placebo or Music v No music), but have been unable to determine how/if to test for an interaction effect using all four conditions. Unlike regular mean scores, I can't subtract the means of the music conditions and then compare those results between drug and placebo.
Sincerely,
David
Threaded View
| Title | Author | Date |
|---|---|---|
| David de Wide | Aug 3, 2017 | |
| Max Kathofer | 8 hours ago | |
| yue zhang | Jan 11, 2020 | |
| Andrew Zalesky | Aug 5, 2017 | |
| Liam Nestor | Sep 27, 2019 | |
| Andrew Zalesky | Sep 28, 2019 | |
| Liam Nestor | Sep 28, 2019 | |
| Andrew Zalesky | Sep 28, 2019 | |
| Liam Nestor | Sep 30, 2019 | |
| Andrew Zalesky | Oct 1, 2019 | |
| Liam Nestor | Oct 1, 2019 | |
| David de Wide | Aug 6, 2017 | |
| Andrew Zalesky | Aug 7, 2017 | |
| David de Wide | Aug 7, 2017 | |
| Andrew Zalesky | Aug 7, 2017 | |
| David de Wide | Aug 11, 2017 | |
| Andrew Zalesky | Aug 12, 2017 | |
| David de Wide | Aug 18, 2017 | |
| Andrew Zalesky | Aug 19, 2017 | |