It is done by Odds Ratio. It is presented in two ways:

(a) Forest plot : The usual way of displaying data from a meta-analysis is by pictorial representation in two columns. The left-hand column lists the names of the studies commonly in chronological order from top downwards, while the right-hand column is a plot of the measure of effect in odds ratio for each of these studies (often represented by a square) incorporating confidence intervals represented by horizontal lines. Overall measure of the effect is plotted as a diamond, the lateral points of which indicate confidence interval.

(b) Heterogeneity : It is the extent to which the selected studies are different from each other. It can be tested by–

– Cochrane’s Q : a statistic based on the chi-squared test.

– I2 statistic (It is preferred) : I2 value 25%, 50% and 75% correspond to low, moderate and high heterogeneity respectively.

The presence or absence of heterogeneity influences the subsequent method of analysis. If heterogeneity is absent, then the analysis is done by fixed-effects modelling. If heterogeneity is large, analysis is done by meta-regression.

Hi

Yes, if your data is continuous, the statistical method for weighting is inverse variance and finally your effect measures can be mean difference or standardized MD. 

Thank you Mr Ahmad, for answering this question 

Thank you Dr Bobby for your time in answering this question 

We can use random effect model also to counter heterogeneity.