Nonsampling vs. Sampling Errors in Survey Research

This descripto-analytical paper on ensuring quality in survey research discusses the management of errors in administering survey. This paper aims to help the social science researchers to ensure the quality in the process and outcomes of survey research. It begins with the brief conceptual underpinnings of survey research, discusses about reliability and validity tests in survey, elaborates the notion of total survey error approach, and suggests some measures on handling survey errors. Given the wider applications and substantial costs associated with survey research, the issues of sampling and non-sampling errors have always been major concerns in the quality of survey research. Survey research can be instrumental in generating knowledge provided survey errors are handled properly. Though a variety of measures are in practices to ensure quality of survey data, this paper gives importance on total survey approach that gives emphasis on total quality management in the collection, analysis, and interpretation of data. Dealing survey data from the perspective of total survey approach would yield fruitful results from survey research.

Download Full-text

An Empirical Investigation of Sampling Errors In Educational Survey Research

Journal of Educational Statistics ◽

10.3102/10769986004001024 ◽

1979 ◽

Vol 4 (1) ◽

pp. 24-40 ◽

Cited By ~ 6

Author(s):

Kenneth N. Ross

Keyword(s):

Survey Research ◽

Error Estimation ◽

Multivariate Statistics ◽

Empirical Investigation ◽

Sampling Technique ◽

Sampling Errors ◽

Sample Design ◽

Estimation Techniques ◽

Sampling Distributions ◽

Complex Sample

This investigation examines the influence of sample design on the sampling errors of several multivariate statistics which are frequently used in educational survey research. Student’s empirical sampling technique is used to generate sampling distributions for several complex sample designs which are often used to sample schools, classrooms and students. Some results are presented for two error estimation techniques: “Jackknifing” and “Balanced Repeated Replication”.

Download Full-text