Table 2

Power and type I error rate of publication bias detection methods

Publication biasNτ2I2 (%)EggerRankHarbordPetersSchwarzerArc-EggerArc-rankTrim
None100.02170.060.030.060.040.030.060.030.01
None100.12560.060.020.060.040.020.060.020.00
None100.90900.060.010.070.030.010.070.010.00
None200.02170.060.030.060.050.020.060.030.02
None200.12560.080.020.080.040.020.080.020.01
None200.90900.090.010.110.030.010.110.010.00
None500.02170.070.020.070.040.020.070.020.03
None500.12560.110.020.110.040.020.110.020.02
None500.90900.130.010.150.030.010.150.010.01
None1000.02170.080.030.080.050.020.080.020.04
None1000.12560.140.020.140.050.020.140.020.03
None1000.90900.150.020.180.040.010.180.020.02
RR=4100.02170.050.020.050.040.020.050.030.00
RR=4100.12560.060.020.060.030.020.060.020.00
RR=4100.90900.050.010.060.030.020.060.020.00
RR=4200.02170.060.020.050.030.020.050.020.01
RR=4200.12560.070.010.070.030.010.080.010.00
RR=4200.90900.080.010.090.030.020.090.020.00
RR=4500.02170.080.020.080.040.020.080.020.04
RR=4500.12560.100.020.110.040.020.110.020.02
RR=4500.90900.110.020.120.030.050.130.050.00
RR=41000.02170.080.020.080.040.020.080.020.05
RR=41000.12560.130.030.150.050.040.140.040.03
RR=41000.90900.130.030.150.040.130.150.110.00
RR=10100.02170.050.020.050.040.020.050.020.00
RR=10100.12560.050.020.050.030.020.050.020.00
RR=10100.90900.080.030.070.040.020.070.030.00
RR=10200.02170.050.020.050.030.020.050.020.02
RR=10200.12560.060.020.060.030.020.060.020.00
RR=10200.90900.110.040.090.040.020.090.030.00
RR=10500.02170.060.020.060.030.020.060.020.05
RR=10500.12560.070.020.070.030.020.080.020.01
RR=10500.90900.190.060.140.040.030.140.030.00
RR=101000.02170.070.020.060.040.010.060.020.09
RR=101000.12560.090.030.090.030.050.090.040.03
RR=101000.90900.300.080.190.060.030.180.030.00
  • All proportions were based on 10 000 samples.