Skip to main content

Table 6 2013 BRFSS Survey Data: Estimation for the probabilities of satisfaction with health care received for the Hispanic participants who were unable to work with annual household income less than 15000 dollars, N=1430 (overall missing rate=25.4%)

From: A nonparametric multiple imputation approach for missing categorical data

 

Pr(Y=Very Satisfied)

Pr(Y=Somewhat Satisfied)

Method

Est (SE)

95% CI

Est (SE)

95% CI

CC

0.585 (0.015)

(0.555, 0.614)

0.335 (0.014)

(0.306, 0.363)

CE

0.553 (0.016)

(0.521, 0.584)

0.349 (0.016)

(0.319, 0.380)

PMI

0.552 (0.014)

(0.524, 0.581)

0.345 (0.014)

(0.318, 0.372)

NNMI MLR (5,0.4,0.4;0.2)

0.560 (0.019)

(0.522, 0.598)

0.353 (0.020)

(0.314, 0.392)

NNMI MLR (5,0.1,0.7;0.2)

0.556 (0.019)

(0.519, 0.592)

0.351 (0.021)

(0.310, 0.391)

NNMI MLR (5,0.7,0.1;0.2)

0.550 (0.022)

(0.507, 0.594)

0.359 (0.019)

(0.322, 0.396)

NNMI CLR (5,0.4,0.4;0.2)

0.547 (0.021)

(0.506, 0.588)

0.358 (0.017)

(0.324, 0.392)

NNMI CLR (5,0.1,0.7;0.2)

0.559 (0.016)

(0.528, 0.590)

0.352 (0.016)

(0.320, 0.383)

NNMI CLR (5,0.7,0.1;0.2)

0.555 (0.018)

(0.520, 0.590)

0.350 (0.019)

(0.314, 0.387)

  1. Est: Estimates of probabilities; SE: Estimate of standard error; 95%CI: 95% confidence interval
  2. X: covariates as gender, general health, education level, having health care coverage, and having delayed getting medical care, that are used in working models
  3. CC: Complete Cases; CE: Calibration estimator; PMI: Parametric Multiple Imputation; NNMI MLR (NN,ω 1,ω 2;ω 3): denotes the NNMI method using Multinomial Logistic Regressions, NN is the number of nearest neighbors and weights are ω 1,ω 2, and ω 3; NNMI CLR : the NNMI method using Cumulative Logistic Regressions; K = 10 imputed datasets are used for PMI and NNMI methods
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us