TY - JOUR
T1 - A comparison of chi-squared statistics for testing homogeneity of survey data
T2 - High School and Beyond survey
AU - Wilson, Jeffrey
AU - Wilson, Patricia M.
N1 - Funding Information:
This research was supported in part by the Faculty Grant-in-Aid Program and a Research Incentive grant from the Bureau of Research at Arizona State University. The authors wish to thank Professor Dave L. Turner of Utah State University who gave great assistance in writing the FORTRAN programs used to simulate the data.
PY - 1991/1/1
Y1 - 1991/1/1
N2 - The performance of several test statistics for comparing vectors of proportions from certain survey data was compared. The statistics were used to analyze a subsample of data from the ‘High School and Beyond’ survey. These tests include the Wald test statistic and the modified Wald test statistic Fw, the chi-squared test statistic XRSB and its modification FRSB, a test X2DMB based on a probability model, and a method of moments approach, X2H. Data were also simulated based on two-stage cluster sampling design and the type I error level, and the power of these tests was obtained for selected combinations of parameter values. The statistics X2DMB, X2RSB, FRSB and X2H performed well both for a small number of clusters or a small number of units within clusters. The power performance of these tests is quite stable. Approximate intervals were constructed for design effect constants. Methods of estimating these constants based on a normality assumption worked best.
AB - The performance of several test statistics for comparing vectors of proportions from certain survey data was compared. The statistics were used to analyze a subsample of data from the ‘High School and Beyond’ survey. These tests include the Wald test statistic and the modified Wald test statistic Fw, the chi-squared test statistic XRSB and its modification FRSB, a test X2DMB based on a probability model, and a method of moments approach, X2H. Data were also simulated based on two-stage cluster sampling design and the type I error level, and the power of these tests was obtained for selected combinations of parameter values. The statistics X2DMB, X2RSB, FRSB and X2H performed well both for a small number of clusters or a small number of units within clusters. The power performance of these tests is quite stable. Approximate intervals were constructed for design effect constants. Methods of estimating these constants based on a normality assumption worked best.
UR - http://www.scopus.com/inward/record.url?scp=84958363263&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84958363263&partnerID=8YFLogxK
U2 - 10.1080/02664769100000016
DO - 10.1080/02664769100000016
M3 - Article
AN - SCOPUS:84958363263
SN - 0266-4763
VL - 18
SP - 203
EP - 213
JO - Journal of Applied Statistics
JF - Journal of Applied Statistics
IS - 2
ER -