Dear Professor,
I am currently working with the dNBS toolbox. I am having difficulty running post-hoc on the toolbox. Specifically, I have three groups and successfully ran ANOVA (one-way anova) on my data, obtaining the correct result. However, I don't know how to use the text file named 'Test stats' for conducting multiple comparisons. I attempted to run a t-test between two groups, but it did not yield any significant results at any threshold. Furthermore, neither Matlab nor SPSS can use the stats generated from the dNBS toolbox. Could you please assist me with what steps I should take?
Thank you.
Hi,
dNBS is an independent package. You might want to reach out to the developers of dNBS. Perhaps they might follow this forum and respond...
Originally posted by mahtabb_99:
Dear Professor,
I am currently working with the dNBS toolbox. I am having difficulty running post-hoc on the toolbox. Specifically, I have three groups and successfully ran ANOVA (one-way anova) on my data, obtaining the correct result. However, I don't know how to use the text file named 'Test stats' for conducting multiple comparisons. I attempted to run a t-test between two groups, but it did not yield any significant results at any threshold. Furthermore, neither Matlab nor SPSS can use the stats generated from the dNBS toolbox. Could you please assist me with what steps I should take?
Thank you.
Thank you very much for your response.
How can I contact the developers of the dNBS toolbox?
Originally posted by mahtabb_99:
Thank you very much for your response.
How can I contact the developers of the dNBS toolbox?
Hi mahtabb_99,
as I read, you are interested in doing post-hoc testing.
In your case you mentioned that you want to use the "Test stats"
file which contains the observed test statistics of the
connection.
I have a newer version of dNBS in which I have included a file called "dNBS_allAsiNetsWithObs". This updated file contains all the necessary information, including the test statistics, the mean of the observations (connectivity values of the subjects), and the observations for all subjects for all edges. I think this file contains the values you need for your post hoc test.
Example of the File:
AsiNetwork 1 1 BA5_LH to 6 BA45_LH
Test stat 3.45
Sample mean: 0.00 Sample subgroup means (according to design
matrix columns): 0.01 -0.00 Sample values :
0.113 0.019 0.008 0.019 0.000 -0.004
-0.001 -0.011 0.007 -0.004 -0.044 -0.017 0.036
0.006 0.008 -0.009 0.005 -0.008 -0.003
0.015 0.019 0.035 0.001 0.069
0.043 -0.007 0.028 -0.001 -0.014 0.014
0.004 0.005 -0.005 -0.010 0.040 0.136
-0.135 0.004 -0.239 0.013 -0.032 -0.009 -0.006
0.008 -0.002 0.027 0.014 0.186 -0.015
-0.016 0.002 0.011 -0.006 0.009 -0.022 -0.013
0.005 0.026 0.002 -0.047 -0.014 -0.000 -0.017
0.001 -0.001 -0.004
AsiNetwork 1 1 BA5_LH to 11 BA28_RH
Test stat 3.42
Sample mean: 0.00 Sample subgroup means (according to design
matrix columns): 0.01 -0.01 Sample values :
0.097 -0.020 0.003 -0.006 -0.074 0.017
0.039 0.010 0.053 0.001 -0.006 0.004
0.010 -0.012 0.056 -0.019 0.010 -0.045 -0.010
-0.009 -0.007 0.008 -0.003 0.019 0.149
0.000 0.060 0.079 0.012 -0.012 0.050
-0.014 -0.022 -0.026 0.032 -0.006 -0.008 0.042
0.008 0.001 -0.008 -0.007 0.011 -0.024
0.064 -0.003 0.057 -0.124 -0.053 0.035 -0.146
-0.017 0.003 0.001 0.025 0.049 0.003
0.009 -0.005 -0.098 0.003 0.002 -0.021 -0.035
0.032 -0.104
Thank you very much for your help.
However, I also need the Square root of the mean squared error for ANOVA1 in MATLAB.
Otherwise, I will have to include additional parameters for N-way ANOVA.
You should be able to calculate this from the file:
This Result File is form the Example Data. We have 2 Groups with 15 Subjects.
Sample Mean = Overall mean
The first 15 Sample Values are from group 1 the last 15 Sample Values are from Group 2 (Coresponding the design Matrix you use)
AsiNetwork 1 2 Frontal_Mid_R to 36 Temporal_Mid_R Test stat 4.31 Sample mean: 0.03 Sample subgroup means (according to design matrix columns): 0.05 0.01 Sample values : 0.002 0.392 0.001 0.005 0.001 0.103 0.003 0.146 0.003 0.003 0.003 0.005 0.004 0.003 0.131 0.003 0.002 0.005 0.005 0.002 0.002 0.002 0.006 0.002 0.002 0.002 0.002 0.006 0.001 0.105
To calculate the square root of the mean squared error (RMSE) for ANOVA1, you'll need to perform the following steps:
The first 3 Results are allready given (but rounded)
Calculate the overall mean: Calculate the mean of all the sample values from both groups.
Calculate the group means: Calculate the mean of each group's sample values separately.
Calculate the squared deviations: Calculate the squared difference between each individual sample value and its corresponding group mean. Sum up these squared deviations.
Calculate the mean squared error (MSE): Divide the sum of squared deviations by the total number of samples minus the number of groups.
Calculate the square root of the mean squared error (RMSE): Take the square root of the MSE calculated in step 4.
Here's the calculation based on the provided data:
Sample values group 1: [0.002, 0.392, 0.001, 0.005, 0.001,
0.103, 0.003, 0.146, 0.003, 0.003, 0.003, 0.005, 0.004, 0.003,
0.131]
Sample values group 2: [0.003, 0.002, 0.005, 0.005, 0.002, 0.002,
0.002, 0.006, 0.002, 0.002, 0.002, 0.002, 0.006, 0.001, 0.105]
Overall mean: (Sum of all values) / (Total number of values) =
(Sum of all values) / 30
= (0.002 + 0.392 + ... +
0.002 + 0.105) / 30
Group 1 mean: Sum of group 1 values / Number of group 1
values
= (0.002 + 0.392 + ... +
0.004 + 0.131) / 15
Group 2 mean: Sum of group 2 values / Number of group 2
values
= (0.003 + 0.002 + ... +
0.006 + 0.105) / 15
Sum of squared deviations: Sum[(individual value - corresponding group mean)^2] for all values
MSE: Sum of squared deviations / (Total number of values -
Number of groups)
= Sum of squared deviations / (30 - 2)
RMSE: Square root of MSE
= sqrt(MSE)