Bias in an Observational Study on Population

DOI HANDLE Web Site 2 Citations Open Access
  • Babazono Akira
    Institute of Health Science, Kyushu University
  • Tsuda Toshihide
    Department of Hygiene and Preventive Medicine, Okayama University, Medical School
  • Mino Yoshio
    Department of Hygiene and Preventive Medicine, Okayama University, Medical School
  • Yamamoto Eiji
    Department of Applied Mathematics, Okayama University of Science

Bibliographic Information

Other Title
  • 集団を対象とした観察研究におけるバイアス
  • シュウダンオタイショウトシタカンサツケンキュウニオケルバイアス
  • シュウダン オ タイショウ ト シタ カンサツ ケンキュウ ニ オケル バイア

Search this article

Abstract

Avoiding bias is essential in an observational study on population since bias leads to a misinterpretation of the magnitude of the effect on a result. Bias is classified into three categories; selection bias, infomation bias and confounding. Selection bias refers to a distortion in the estimate of effect resulting from the manner in which subjects are selected for the study population. Information bias refers to a distortion in the estimate of effect due to measurement error or misclassification of subjects on one or more variables. Confounding is the bias that results when the study factor effect is mixed, in the data, with the effects of extraneous variables. We must maintain the specific characteristics of each of those biases in mind in order to control them. It is not fair to insist that a study result is incorrect because of a small bias since it is impossible to rule out any bias completely when data are collected by observation. Direction of bias is very helpful in judging the study result when some bias is noticed. Neither selection bias nor information bias can be controlled during the data analytic phase. Both selection bias and information bias should be minimized in the reseatch design. Confounding can be controlled in either the research design or the data analytic phase.

Journal

  • 健康科学

    健康科学 17 35-42, 1995-02-25

    Institute of Health Science,Kyushu University

Citations (2)*help

See more

Keywords

Details 詳細情報について

Report a problem

Back to top