Confounding and collinearity in regression analysis: a cautionary tale and an alternative procedure, illustrated by studies of British voting behaviour
Open accessAbstract
Many ecological- and individual-level analyses of voting behaviour use multiple regressions with a considerable number of independent variables but few discussions of their results pay any attention to the potential impact of inter-relationships among those independent variables-do they confound the regression parameters and hence their interpretation? Three empirical examples are deployed to address that question, with results which suggest considerable problems. Inter-relationships between variables, even if not approaching high collinearity, can have a substantial impact on regression model results and how they are interpreted in the light of prior expectations. Confounded relationships could be the norm and interpretations open to doubt, unless considerable care is applied in the analyses and an extended principal components method for doing that is introduced and exemplified.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The authors checked for multicollinearity using VIF and confirmed that all variables used had VIF lower than 2.5 following Johnston et al. [28].
“Second, we checked for multicollinearity using the Variation Influence Factor (VIF) and confirmed that all variables we used had VIF lower than 2.5 following Johnston et al. [28].”
From Using an LLM to Help With Code Understanding (2024) · cites this paper as Johnston et al. [28]
GitHub recently launched CopilotX, a tool that offers a comparable set of features to the prototype to enhance developer experience.
“Given that GitHub recently launched CopilotX [27], a tool that offers a comparable set of features to our prototype to enhance developer experience”
From Using an LLM to Help With Code Understanding (2024) · cites this paper as [27]
The Introduction and the sections on British voting behaviour discuss regression, confounding, and collinearity, not developer tools. Searches for CopilotX, GitHub, its launch, and comparable developer features found no passage addressing the claim in our copy. (LLM, conf=0.78)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Baron, R. M., & Kenny, D. A. (1986) · Journal of Personality and Social Psychology · doi:10.1037//0022-3514.51.6.1173 · Cited by 91,542 · see citing works
Baron, R. M., & Kenny, D. A. (1986) · Journal of Personality and Social Psychology · doi:10.1037/0022-3514.51.6.1173 · Cited by 91,532 · see citing works
Sensitivity analysis in observational research: Introducing the E-value
VanderWeele, T. J., & Ding, P. (2017) · Annals of Internal Medicine · doi:10.7326/m16-2607 · Cited by 6,350 · see citing works
Introduction to Statistical Mediation Analysis
MacKinnon, D. P. (2012) · doi:10.4324/9780203809556 · Cited by 5,618 · see citing works
A general approach to causal mediation analysis.
Imai, K., Keele, L., & Tingley, D. (2010) · Psychological Methods · doi:10.1037/a0020761 · Cited by 4,143 · see citing works
Fox, J. (1991) · doi:10.4135/9781412985604 · Cited by 1,008 · see citing works
On the definition of a confounder
VanderWeele, T. J., & Shpitser, I. (2013) · The Annals of Statistics · doi:10.1214/12-aos1058 · Cited by 268 · cited 2× · see citing works
“The result is frequently termed confounding, the situation when the relationship between two variables is distorted because of the strength of the relationships between either one or both of them and a third variable included in the analysis (see, for example, Kish 1959; Morabia 2011; VanderWheele and Shpitser 2013).”
Background
Flagged first, then by citations
Research works printed but not linked: 18
- Allison P Multiple Regression: A Primer 1999 Thousand Oaks Pine Forge Press
- Clarke HD Sanders D Stewart MC Whiteley P Performance Politics and the British Voter 2010 Cambridge Cambridge University Press
- Clarke HD Goodwin M Whiteley P Brexit: Why Britain Voted to leave the European Union 2017 Cambridge Cambridge University Press
- Evans G Chzhen K Re-evaluating the valence model of political choice Polit. Sci Res Methods 2016 4 2 199 220 10.1017/psrm.2013.11 · DOI
- Franzosi R Outside and inside the regression “black box” from exploratory to interior data analysis Qual. Quant. 1994 28 1 21 53 10.1007/BF01098725 · DOI
- Goodwin M Heath O The 2016 referendum, Brexit and the left behind: an aggregate-level analysis of the result Polit. Q. 2016 87 3 323 332 10.1111/1467-923X.12285 · DOI
- Goodwin M Milazzo C UKIP: Inside the Campaign to Redraw the Map of British Politics 2015 Oxford Oxford University Press
- Hayes AF Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach 2013 New York The Guilford Press
- Imai K Keele L Tingley D Yamamoto T Unpacking the black box of causality: learning about causal mechanisms from experimental and observational studies Am. Polit. Sci. Rev. 2011 105 4 765 789 10.1017/S0003055411000414 · DOI · Paper page
- Johnston RJ Jones K Burgess S Propper C Sarker R Bolster A Scale, factor analyses, and neighborhood effects Geogr. Anal. 2004 36 4 350 368 10.1111/j.1538-4632.2004.tb01141.x · DOI
- King G How not to lie with statistics: avoiding common mistakes in quantitative political science Am. J. Polit. Sci. 1986 30 3 666 687 10.2307/2111095 · DOI · Paper page
- Kish L Some statistical problems in research design Am. Sociol. Rev. 1959 24 4 328 338 10.2307/2089381 · DOI
- Massy WF Principal components regression in exploratory statistical research J. Am. Stat. Assoc. 1965 60 234 256 10.1080/01621459.1965.10480787 · DOI
- Moon G Gould M Jones K Brown T Duncan C Twigg L Subramanian SV Litvia A Iggulden P Epidemiology: An Introduction 2000 Buckingham Open University Press
- Morabia A History of the modern epidemiological concept of confounding J. Epidemiol. Commun. Health 2011 65 2 297 300 10.1136/jech.2010.112565 20696848 · DOI
- Riddell JB On structuring a migration model Geogr. Anal. 1970 2 4 403 409 10.1111/j.1538-4632.1970.tb00870.x · DOI
- Shipley B Cause and Correlation in Biology: A User’s Guide to Path Analysis, Structural Equations and Causal Inference 2009 Cambridge Cambridge University Press
- VanderWeele TJ Explanation in Causal Inference: Methods for Mediation and Interaction 2015 Oxford Oxford University Press
Cited by
Sorted by citations
Using an LLM to Help With Code Understanding
Nam, D., Macvean, A., Hellendoorn, V., Vasilescu, B., & Myers, B. (2024) · Proceedings of the IEEE/ACM 46th International Conference on Software Engineering · doi:10.1145/3597503.3639187 · Cited by 345 · see citing works
Du, Y., Xu, T., Yin, Z., Espinoza, S., Xie, Y., Gentry, C., Tian, Q., Zhao, L.-J., Shen, H., Luo, Z., et al. (2022) · BMC Public Health · doi:10.1186/s12889-022-13288-5 · Cited by 39 · see citing works
Khatri, R. B., Durham, J., Karkee, R., & Assefa, Y. (2022) · Reproductive Health · doi:10.1186/s12978-022-01465-z · Cited by 36 · see citing works
Rautio, D., Gumpert, M., Jassi, A., Krebs, G., Flygare, O., Andrén, P., Monzani, B., Peile, L., Jansson-Fröjmark, M., Lundgren, T., et al. (2022) · Behavior Therapy · doi:10.1016/j.beth.2022.04.010 · Cited by 32 · see citing works
McCarty, R. J., Downing, S. T., Daley, M. L., McNamara, J. P. H., & Guastello, A. D. (2023) · Anxiety, Stress, & Coping · doi:10.1080/10615806.2022.2075855 · Cited by 16 · see citing works
Eustress and Distress in the Context of Telework
Van Slyke, C., Lee, J., Duong, B. Q., & Ellis, T. S. (2022) · Information Resources Management Journal · doi:10.4018/irmj.291526 · Cited by 12 · see citing works
- Authors
- Johnston, R., Jones, K., & Manley, D.
- Affiliations
- University of Bristol (GB)
- Corresponding author
- Johnston, Ron
- Venue
- Quality & Quantity
- Published
- 2017-11-13
- Volume
- 52
- Issue
- 4
- Pages
- 1957-1976
- ISSN
- 0033-5177 (ISSN-L)
- 1573-7845
- PMC
- PMC5993839
- OpenAlex
- W2769972173
- Cited by
- 1,181
- References
- 7
- Citations by year
- 2026: 139 · 2025: 206 · 2024: 259 · 2023: 236 · 2022: 203 · 2021: 93 · 2020: 30 · 2019: 6 · 2018: 7 · 2017: 1
- FWCI
- 8.37
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Locations
- Quality & Quantity (publishedVersion, cc-by)
- Europe PMC (PubMed Central) (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Explore Bristol Research (publishedVersion, cc-by)
- RePEc: Research Papers in Economics (submittedVersion)
- Lirias (submittedVersion)
- Bristol Research (University of Bristol) (submittedVersion, cc-by)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Statistical Methods and Models
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Applications
- Social Sciences › Political Science and International Relations › Electoral Systems and Political Participation
- Keywords
- confounding, multicollinearity, multiple linear regression, voting behavior, principal component regression, interpretation of regression coefficients, British elections
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
