Home

adaptivetesting

Features

Ability Estimators
Maximum Likelihood	$\arg \max, L(Y=1| \theta)$
Bayes Modal	$\arg \max \frac{L(\theta)P_\text{prior}(\theta)}{\int_\infty^\infty L(\theta)P_\text{prior}(\theta)}$
Expected a Posteriori	$\frac{\int_{-\infty}^{\infty} \theta L(\theta) P_\text{prior}(\theta), d\theta}{\int_{-\infty}^{\infty} L(\theta) P_\text{prior}(\theta), d\theta}$
Item Selection
MFI	$\arg \max I_i(\theta)$
Exposure Control
Randomesque
Maximum Priority Index	$PI_i = I_i \prod_{k=1}^K (w_k f_k)^{c_{ik}}$
Content Balancing
Maximum Priority Index	$PI_i = I_i \prod_{k=1}^K (w_k f_k)^{c_{ik}}$
Weighted Penalty Model	$F_i = w'F_i' + w'' F_i''$
Stopping Criterion
Standard Error	$\leq \frac{1}{\sqrt{I(\theta)}}$
Test Length	# of items