Table 2.

Expected Number of Limitations Across Complete Case (N=5959) and Full Sample (N=9914) FIML Analyses

	(1)	(2)
	Complete Case Estimation	Full Sample Estimation—
		FIML As Observed
Round 2
Mean Predicted Limitations	1.59 (1.89)	1.86 (2.15)
Skewness	1.96	1.78
Round 6
Mean Predicted Limitations	2.04 (1.59)	2.45 (2.13)
Skewness	1.66	1.45
Round 10
Mean Predicted Limitations	2.92 (1.38)	3.36 (1.68)
Skewness	1.49	1.20
Average Across Rounds
Mean Predicted Limitations	2.13 (1.61)	2.52 (1.99)
Skewness	1.69	1.40

Estimates are based on predicted values from LCGA estimates under two common sets of assumptions. When estimated with the full sample, the complete case sample is estimated to have, on average, 1.6 limitations at survey round 2 and 3.13 limitations at Round 10.