I would like to measure (to quantify with a statistical test) how much different are paired scores. For example: for gene "x" I performed an experiment with technology A that gives a score, then for the same gene I performed an experiment with technology B that gives a score. I would like to compare the scores and measure how much different they are in order to check how much different the two technologies are.
Data type: continuous numerical data.
Can anyone help me please?
Thank you very much
A paired t-test is appropriate for this data, but only if you are asking "are these different". To ask "how different are they" (usually after you've already asked "are they different"), your question seems to be looking at an effect size. A confidence interval around the mean difference (mentioned by @Harvey Motulsky) could be a good measure. Another common one is Cohen's D, which gives the mean difference in terms of standard deviations. If looking into Cohen's D, I'd recommend checking out this guide for some practical details on interpretation.
As @StevenKolassa pointed out, the paired t test is designed for exactly the kind of data you describe if the goal is to get a P value and a confidence interval for the mean difference between the two methods.
But method comparison is its own world. Look up Bland-Altman plots which are a way to plot your data and assess how the two methods differ. I think that may be more useful to you than a paired t test.
