Assessing the Impact of Non-Random Measurement Error on Inference

A Sensitivity Analysis Approach

Max Gallop, Simon Weschle

Research output: Contribution to journalReview article

1 Citation (Scopus)

Abstract

Many commonly used data sources in the social sciences suffer from non-random measurement error, understood as mis-measurement of a variable that is systematically related to another variable. We argue that studies relying on potentially suspect data should take the threat this poses to inference seriously and address it routinely in a principled manner. In this article, we aid researchers in this task by introducing a sensitivity analysis approach to non-random measurement error. The method can be used for any type of data or statistical model, is simple to execute, and straightforward to communicate. This makes it possible for researchers to routinely report the robustness of their inference to the presence of non-random measurement error. We demonstrate the sensitivity analysis approach by applying it to two recent studies.

Original languageEnglish (US)
Pages (from-to)367-384
Number of pages18
JournalPolitical Science Research and Methods
Volume7
Issue number2
DOIs
StatePublished - Apr 1 2019
Externally publishedYes

Fingerprint

social science
threat

ASJC Scopus subject areas

  • Sociology and Political Science
  • Political Science and International Relations

Cite this

Assessing the Impact of Non-Random Measurement Error on Inference : A Sensitivity Analysis Approach. / Gallop, Max; Weschle, Simon.

In: Political Science Research and Methods, Vol. 7, No. 2, 01.04.2019, p. 367-384.

Research output: Contribution to journalReview article

@article{41813fb3e1cd42b4af32711eabca28b3,
title = "Assessing the Impact of Non-Random Measurement Error on Inference: A Sensitivity Analysis Approach",
abstract = "Many commonly used data sources in the social sciences suffer from non-random measurement error, understood as mis-measurement of a variable that is systematically related to another variable. We argue that studies relying on potentially suspect data should take the threat this poses to inference seriously and address it routinely in a principled manner. In this article, we aid researchers in this task by introducing a sensitivity analysis approach to non-random measurement error. The method can be used for any type of data or statistical model, is simple to execute, and straightforward to communicate. This makes it possible for researchers to routinely report the robustness of their inference to the presence of non-random measurement error. We demonstrate the sensitivity analysis approach by applying it to two recent studies.",
author = "Max Gallop and Simon Weschle",
year = "2019",
month = "4",
day = "1",
doi = "10.1017/psrm.2016.53",
language = "English (US)",
volume = "7",
pages = "367--384",
journal = "Political Science Research and Methods",
issn = "2049-8470",
publisher = "Cambridge University Press",
number = "2",

}

TY - JOUR

T1 - Assessing the Impact of Non-Random Measurement Error on Inference

T2 - A Sensitivity Analysis Approach

AU - Gallop, Max

AU - Weschle, Simon

PY - 2019/4/1

Y1 - 2019/4/1

N2 - Many commonly used data sources in the social sciences suffer from non-random measurement error, understood as mis-measurement of a variable that is systematically related to another variable. We argue that studies relying on potentially suspect data should take the threat this poses to inference seriously and address it routinely in a principled manner. In this article, we aid researchers in this task by introducing a sensitivity analysis approach to non-random measurement error. The method can be used for any type of data or statistical model, is simple to execute, and straightforward to communicate. This makes it possible for researchers to routinely report the robustness of their inference to the presence of non-random measurement error. We demonstrate the sensitivity analysis approach by applying it to two recent studies.

AB - Many commonly used data sources in the social sciences suffer from non-random measurement error, understood as mis-measurement of a variable that is systematically related to another variable. We argue that studies relying on potentially suspect data should take the threat this poses to inference seriously and address it routinely in a principled manner. In this article, we aid researchers in this task by introducing a sensitivity analysis approach to non-random measurement error. The method can be used for any type of data or statistical model, is simple to execute, and straightforward to communicate. This makes it possible for researchers to routinely report the robustness of their inference to the presence of non-random measurement error. We demonstrate the sensitivity analysis approach by applying it to two recent studies.

UR - http://www.scopus.com/inward/record.url?scp=85064856470&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064856470&partnerID=8YFLogxK

U2 - 10.1017/psrm.2016.53

DO - 10.1017/psrm.2016.53

M3 - Review article

VL - 7

SP - 367

EP - 384

JO - Political Science Research and Methods

JF - Political Science Research and Methods

SN - 2049-8470

IS - 2

ER -