Table 2.

Estimates and $95\%$ intervals for parameters in fixed-effects and hierarchical models used in meta-analysis of progesterone accuracy studies. The priors used on ${\sigma }_{\alpha }^2$ and ${\sigma }_{\beta }^2$ were $ℐ𝒢\left(\mathrm{0.01,0.01}\right)$.

	Parameter	Pregnancy failure
		Covariate	No covariate
FE	$\alpha$	0.55 (0.45, 0.64)	0.49 (0.40, 0.58)
	boot CI	(0.37, 0.73)	(0.33, 0.65)
	$\beta$	2.47 (2.29, 2.64)	2.42 (2.25, 2.59)
	boot CI	(2.07, 2.87)	(2.02, 2.82)
	$\gamma$	0.10 (0.05, 0.15)	–
	boot CI	(−0.10, 0.30)	–
HM	${\mu }_{{\alpha }_k}$	0.48(0.34,0.62)	0.49(0.33,0.63)
	${\sigma }_{{\alpha }_k}$	0.11 (0.06, 0.27)	0.12 (0.05, 0.25)
	${\mu }_{{\beta }_k}$	2.28 (1.81, 2.76)	2.29 (1.83, 2.76)
	${\sigma }_{{\beta }_k}$	0.89 (0.62, 1.35)	0.88 (0.61, 1.33)
	$\gamma$	−0.02 (−0.16, 0.12)	–