Fabricating data: How substituting values for nondetects can ruin results, and what can be done about it
Abstract
The most commonly used method in environmental chemistry to deal with values below detection limits is to substitute a fraction of the detection limit for each nondetect. Two decades of research has shown that this fabrication of values produces poor estimates of statistics, and commonly obscures patterns and trends in the data. Papers using substitution may conclude that significant differences, correlations, and regression relationships do not exist, when in fact they do. The reverse may also be true. Fortunately, good alternative methods for dealing with nondetects already exist, and are summarized here with references to original sources. Substituting values for nondetects should be used rarely, and should generally be considered unacceptable in scientific research. There are better ways.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
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?
Substitution was suitable for handling censored data because specific statistical methods assume a data distribution not applicable in this case and fewer than 60% of samples were below the detection limit.
“Even though there are specific statistical methods for handling censored data, they assume a specific data distribution not applicable in this case and as there are less than 60% of samples below the detection limit, substitution was suitable 49 .”
From Underestimated burden of per- and polyfluoroalkyl substances in global surface waters and groundwaters (2024) · cites this paper as [49]
Cites
Estimation of Average Concentration in the Presence of Nondetectable Values
Hornung, R. W., & Reed, L. D. (1990) · Applied Occupational and Environmental Hygiene · doi:10.1080/1047322x.1990.10389587 · Cited by 3,495 · see citing works
Survival analysis using the SAS system : a practical guide
Allison, P. D. (1995) · Medical Entomology and Zoology · Cited by 2,402 · see citing works
Nondetects and data analysis : statistics for censored environmental data
Helsel, D. R. (2005) · Wiley-Interscience eBooks · Cited by 521 · see citing works
Self-citationEstimation of descriptive statistics for multiply censored water quality data
Helsel, D. R., & Cohn, T. A. (1988) · Water Resources Research · doi:10.1029/wr024i012p01997 · Cited by 329 · see citing works
Self-citationNondetects and Data Analysis: Statistics for Censored Environmental Data
Harter, T. (2006) · Vadose Zone Journal · doi:10.2136/vzj2005.0106br · Cited by 311 · see citing works
Gilliom, R. J., & Helsel, D. R. (1986) · Water Resources Research · doi:10.1029/wr022i002p00135 · Cited by 307 · see citing works
Self-citationBaccarelli, A., Pfeiffer, R., Consonni, D., Pesatori, A. C., Bonzini, M., Patterson, D. G., Bertazzi, P. A., & Landi, M. T. (2005) · Chemosphere · doi:10.1016/j.chemosphere.2005.01.055 · Cited by 185 · see citing works
Statistical Approaches to Estimating Mean Water Quality Concentrations with Detection Limits
Shumway, R. H., Azari, R. S., & Kayhanian, M. (2002) · Environmental Science & Technology · doi:10.1021/es0111129 · Cited by 130 · see citing works
Singh, A. K., & Nocerino, J. M. (2002) · Chemometrics and Intelligent Laboratory Systems · doi:10.1016/s0169-7439(01)00186-1 · Cited by 118 · see citing works
Estimation of distributional parameters for censored trace-level water-quality data
Gilliom, R. J., & Helsel, D. R. (1984) · Antarctica A Keystone in a Changing World · doi:10.3133/ofr84729 · Cited by 97 · see citing works
Self-citationANALYZING CENSORED WATER QUALITY DATA USING A NON‐PARAMETRIC APPROACH1
She, N. (1997) · JAWRA Journal of the American Water Resources Association · doi:10.1111/j.1752-1688.1997.tb03536.x · Cited by 52 · see citing works
Millard, S. P., & Deverel, S. J. (1988) · Water Resources Research · doi:10.1029/wr024i012p02087 · Cited by 52 · see citing works
Biomonitoring of contaminants in birds from two trophic levels in the North Pacific
Rocque, D. A., & Winker, K. (2004) · Environmental Toxicology and Chemistry · doi:10.1897/03-182 · Cited by 46 · see citing works
Mercury concentrations in water from an unconfined aquifer system, New Jersey coastal plain
Barringer, J. L., Szabó, Z., Kauffman, L. J., et al. (2005) · The Science of The Total Environment · doi:10.1016/j.scitotenv.2004.11.013 · Cited by 43 · see citing works
Insider Censoring: Distortion of Data with Nondetects
Helsel, D. R. (2005) · Human and Ecological Risk Assessment An International Journal · doi:10.1080/10807030500278586 · Cited by 39 · see citing works
Self-citationTajimi, M., Uehara, R., Watanabe, M., et al. (2005) · Chemosphere · doi:10.1016/j.chemosphere.2005.03.096 · Cited by 28 · see citing works
McCarthy, L. H., Stephens, G. R., Whittle, D. M., et al. (1997) · The Science of The Total Environment · doi:10.1016/s0048-9697(96)05420-4 · Cited by 18 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Practitioner’s guide to latent class analysis: Methodological considerations and common pitfalls
Sinha, P., Calfee, C. S., & Delucchi, K. L. (2021) · Critical Care Medicine · doi:10.1097/ccm.0000000000004710 · Cited by 1,204 · see citing works
Ackerman Grunfeld, D., Gilbert, D., Hou, J., Jones, A. M., Lee, M. J., Kibbey, T. C. G., & O’Carroll, D. M. (2024) · Nature Geoscience · doi:10.1038/s41561-024-01402-8 · Cited by 403 · see citing works
de Solla, S. R., De Silva, A. O., & Letcher, R. J. (2012) · Environment International · doi:10.1016/j.envint.2011.09.011 · Cited by 198 · see citing works
Munoz, G., Mercier, L., Duy, S. V., Liu, J., Sauvé, S., & Houde, M. (2022) · Environmental Pollution · doi:10.1016/j.envpol.2022.119739 · Cited by 160 · see citing works
Loos, R., Tavazzi, S., Mariani, G., Suurkuusk, G., Paracchini, B., & Umlauf, G. (2017) · Science of The Total Environment · doi:10.1016/j.scitotenv.2017.07.039 · Cited by 150 · see citing works
Munoz, G., Giraudel, J.-L., Botta, F., Lestremau, F., Dévier, M.-H., Budzinski, H., & Labadie, P. (2015) · Science of The Total Environment · doi:10.1016/j.scitotenv.2015.02.043 · Cited by 144 · see citing works
Liu, Y., Robey, N. M., Bowden, J. A., Tolaymat, T. M., da Silva, B. F., Solo-Gabriele, H. M., & Townsend, T. G. (2021) · Environmental Science & Technology Letters · doi:10.1021/acs.estlett.0c00819 · Cited by 89 · see citing works
A general framework for integrative analysis of incomplete multiomics data
Lin, D., Zeng, D., & Couper, D. (2020) · Genetic Epidemiology · doi:10.1002/gepi.22328 · Cited by 16 · see citing works
Liu, M., Saracevic, E., Oudega, T. J., Obeid, A. A. A., Nagy-Kovács, Z., László, B., Kittlaus, S., Zoboli, O., Krampe, J., Derx, J., et al. (2025) · Environmental Sciences Europe · doi:10.1186/s12302-025-01141-6 · Cited by 9 · see citing works
“This semiparametric approach assumes a log-normal distribution for environmental concentration data and imputes censored values accordingly [47, 48].”
- Authors
- Helsel, D. R.
- Affiliations
- United States Geological Survey (US)
- Corresponding author
- Helsel, Dennis R.
- ORCID
- Helsel, Dennis R.
- Venue
- Chemosphere
- Publisher
- Elsevier BV
- Published
- 2006-12
- Volume
- 65
- Issue
- 11
- Pages
- 2434-2439
- ISSN
- 0045-6535 (ISSN-L)
- OpenAlex
- W2125463545
- Cited by
- 505
- References
- 17
- Citations by year
- 2026: 42 · 2025: 28 · 2024: 25 · 2023: 30 · 2022: 39 · 2021: 29 · 2020: 38 · 2019: 26 · 2018: 37 · 2017: 24 · 2016: 23 · 2015: 30 · 2014: 31 · 2013: 24 · 2012: 19
- FWCI
- 8.36
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Chemosphere (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Physical Sciences › Environmental Science › Water Science and Technology › Environmental and Analytical Chemistry Studies
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Scientific Measurement and Uncertainty Evaluation
- Keywords
- detection limit, data fabrication, statistical estimation, regression analysis, data analysis methods
- MeSH terms
- Data Interpretation, Statistical, Sensitivity and Specificity
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
