A comparison of chi-squared statistics for testing homogeneity of survey data

High School and Beyond survey

Jeffrey Wilson, Patricia M. Wilson

Research output: Contribution to journalArticle

Abstract

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.

Original languageEnglish (US)
Pages (from-to)203-213
Number of pages11
JournalJournal of Applied Statistics
Volume18
Issue number2
DOIs
StatePublished - Jan 1 1991

Fingerprint

Chi-squared
Survey Data
Homogeneity
Test Statistic
Statistics
Wald Test
Testing
Design Effect
Two-stage Sampling
Cluster Sampling
Power of Test
Chi-squared test
Sampling Design
Type I error
Method of Moments
Probability Model
Number of Clusters
Normality
Proportion
Interval

ASJC Scopus subject areas

  • Statistics, Probability and Uncertainty
  • Statistics and Probability

Cite this

A comparison of chi-squared statistics for testing homogeneity of survey data : High School and Beyond survey. / Wilson, Jeffrey; Wilson, Patricia M.

In: Journal of Applied Statistics, Vol. 18, No. 2, 01.01.1991, p. 203-213.

Research output: Contribution to journalArticle

@article{62ef4f70a65047c5b878637a613d862f,
title = "A comparison of chi-squared statistics for testing homogeneity of survey data: High School and Beyond survey",
abstract = "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.",
author = "Jeffrey Wilson and Wilson, {Patricia M.}",
year = "1991",
month = "1",
day = "1",
doi = "10.1080/02664769100000016",
language = "English (US)",
volume = "18",
pages = "203--213",
journal = "Journal of Applied Statistics",
issn = "0266-4763",
publisher = "Routledge",
number = "2",

}

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.

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

VL - 18

SP - 203

EP - 213

JO - Journal of Applied Statistics

JF - Journal of Applied Statistics

SN - 0266-4763

IS - 2

ER -