Skip to main content

Table 3 How often are predictions of blood use better for random than for purposive sampling strategies?

From: Aiming for a representative sample: Simulating random versus purposive strategies for hospital selection

RAND versus: MAXVAR LARG
  RBC FFP PLT RBC FFP PLT
Lower hospital level prediction error for RAND 0 % 3 % 14 % 100 % 35 % 84 %
Lower national level prediction error for RAND 25 % 3 % 79 % 100 % 81 % 87 %
REGVAR versus: MAXVAR LARG
Lower hospital level prediction error for REGVAR 0 % 6 % 25 % 100 % 44 % 90 %
Lower national level prediction error for REGVAR 20 % 4 % 79 % 100 % 79 % 88 %
2REG versus: MAXVAR LARG
Lower hospital level prediction error for 2REG 0 % 0 % 0 % 100 % 9 % 52 %
Lower national level prediction error for 2REG 24 % 3 % 83 % 100 % 70 % 88 %
  1. Percentage of all simulations that the random strategies (RAND, REGVAR and 2REG) outperform the purposive strategies (MAXVAR and LARG) in terms of hospital and national level prediction error for n = 12 (n = 4 per hospital type)