Random definitions

Question 1

Two statistical approaches used in meta-analyses to evaluate extent of variability (heterogeneity) between individual studies

Answer 1

Cochran’s Q
I2 statistic

Question 2

Standardised effect size for measuring difference between two group means (standardised mean difference) - can be used in meta analysis

Answer 2

Cohen’s d

Question 3

Quantitative measure of reliability for two raters that are rating the same thing, correcting for how often the raters may agree by chance

Answer 3

Cohen’s kappa

Question 4

A non-parametric correlation co-efficient that measures the strength of association between two variables

Answer 4

Spearman’s rho

Question 5

A parametric correlation co-efficient that measures the strength of association between two variables

Answer 5

Pearson’s r

Question 6

A coefficient that measures the internal consistency, or reliability, of a set of survey items

Answer 6

Cronbach’s alpha

Question 7

Number of trials needed to negate meta analysis results (quantifying publication bias)

Answer 7

Failsafe N (Rosenthal)

Question 8

A consensus method for creating treatment guidelines

Answer 8

Delphi method

Question 9

Co-efficient of determination - provides info on goodness of fit of predictive model

Answer 9

R squared

Question 10

Method for adjusted odds ratio when controlling for confounders in causal studies (e.g. cohort) or in meta analysis when big difference between treatment and control group sizes

Answer 10

Mantel Haenszel

Question 11

Odds ratio in meta analysis when minimal difference between treatment and control group sizes

Answer 11

Peto method