Table 4 Effect estimates and their sampling variances

MD

Raw values

\({\text{MD}} = M_{1} - M_{2}\)

\(V_{{{\text{MD}}}} = \frac{{n_{1} + n_{2} }}{{n_{1} n_{2} }}S_{{{\text{pooled}}}}^{2}\)

where

\(S_{{{\text{pooled}}}} = \sqrt {\frac{{\left( {n_{1} - 1} \right)S_{1}^{2} + \left( {n_{1} - 1} \right)S_{2}^{2} }}{{n_{1} + n_{2} - 2}}}\)

TE

Log-transformed

\({\text{TE}} = \ln s + \frac{1}{{2\left( {n - 1} \right)}}\)

\(V_{TE} = 1/2\left( {n - 1} \right)\)

ICC

Fisher’s z

\(z = 0.5{ } \times {\text{ln}}\left( {\frac{1 + r}{{1 - r}}} \right)\)

\(V_{icc} = 1/\left( {n - 3/2} \right)\)

r

Fisher’s z

\(z = 0.5{ } \times {\text{ln}}\left( {\frac{1 + r}{{1 - r}}} \right)\)

\(V_{r} = 1/\left( {n - 3} \right)\)