Thank you very much for your reply. Yes, I conducted a two-tailed paired t-test to get the statistical significance between both hand powers.

My question is, if I now want to use the parameter power to do multiple regression analysis, if I can use the power values from both hands as one sample or if I do need to split it (as they are not independent)?

Marion

How did you determine that there was no significant difference between the left and right hands? Did you use a paired t test (or Wilcoxon signed-ranks test)? You don’t need to consider correlations with these tests.

Charles ]]>

First of all, thank you very much for the description above, this is really helpful. I have a question about one of my analysis and can’t find the answer, I hope you don’t mind asking you.

I am analysing biomechanics of hands. For this, I measured the pressure a hand can apply to a handle after different neurological issues. I measured both hands. Between the left and the right hand was no significant difference in none of the parameters I am using. Now, I am not really sure how to deal with it. Can I use the measured values from both hands or do I need to analyse left and right always separately? Is there a possibility to correct for the inter-hand correlation to use all values?

Thank you very much for your help in advance!

]]>1. Which transformation to use depends on a number of factors. This is more art than science. See the following webpages for some suggestions:

http://www.real-statistics.com/descriptive-statistics/data-transformations/

http://www.real-statistics.com/correlation/box-cox-transformation/

Also depending on what you are going to do with your data, you may or may not need to satisfy some of the stated assumptions: linearity, normality, homogeneity of variances.

2. You can do this before or after the transformation, but what you shouldn’t do is mix the results. You should one or the other. Based on your comments, you probably want to eliminate variables after making the transformations.

Charles ]]>

With values this high, you don’t need to use a table of critical values. Instead you can use a normal approximation. This is explained on the following webpage:

http://www.real-statistics.com/non-parametric-tests/mann-whitney-test/

Charles ]]>

The hypothesis depends on what you want to test. Only you can determine this. I am not able to look at your data and tell you what hypothesis you should use.

Charles ]]>

Charles ]]>

There is no agreed upon criteria for acceptable values of Cronbach’s alpha. Some would say that a value higher than .7 is acceptable; some would accept slightly lower values.

I don’t know what you mean by “Should I go by the scale?”

Charles ]]>

The regression analysis actually includes this result (i.e. the Multiple Correlation value on the report). If only one IV has a significant p-value, then only that variable has a significant influence on predicting the DV.

Charles ]]>

Lag = 0. See the following webpage for a more complete explanation:

http://www.real-statistics.com/time-series-analysis/autoregressive-processes/augmented-dickey-fuller-test/

Charles ]]>