Hello Dakshitha,

You could compute a Cohen's d index: (M1 - M2) / SDpooled for each individual study for which you have data. If I follow your query correctly, studies with larger d indices would be suggestive of better diagnosis.

If, on the other hand, your concern was whether the projected classification from the diagnostic procedure under investigation (e.g., "this case is likely diseased") was more accurate in one study than in another, then some other statistic would be better. Classification accuracy (e.g., Cohen's kappa: agreement beyond chance) would be one option.

Good luck with your work.

See Ewout Steyerberg's book @ https://www.bing.com/search?q=ewout+steyerberg+clinical+prediction+models&qs=AS&pq=ewout+steyerberg+&sc=2-17&cvid=76A34AB205D9429B94E280E43C380D36&FORM=QBRE&sp=1&ghc=1

D. Booth

I agree with David Morse and you apply the test by using SPSS, which is simple,