Extrapolation of Correlation Between 2 Variables in 4 General Medical Journals

Kuo, Yen-Hong

doi:10.1001/jama.287.21.2815

Cited by 13 publications

(15 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…While they found that the majority of subjects with heavy prenatal alcohol exposure performed worse than expected based on nonlinear regression estimates of WCST and IQ scores derived from a large comparison group, data from some heavily exposed subjects appear outside the range of IQ scores obtained in the comparison group. In regression, extrapolation outside of the range of scores used to derive the line can result in erroneous or unreliable conclusions (Cohen et al., 2003; Kuo, 2002).…”

Section: Discussionmentioning

confidence: 99%

“…However, these results should be interpreted cautiously since the heavily exposed subjects with the lowest IQ scores do not appear to fall within the range of scores used to calculate the regression lines. Regression lines tend to be unreliable when based on few observations (i.e., at the extremes of a distribution) and extrapolation can lead to extraneous results (Cohen et al., 2003; Kuo, 2002). No studies using the CST exist with which to compare results to general intellectual functioning.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Children With Heavy Prenatal Alcohol Exposure Demonstrate Deficits on Multiple Measures of Concept Formation

McGee

Schonfeld

Roebuck-Spencer

et al. 2008

Alcoholism Clin & Exp Res

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 99%

Children With Heavy Prenatal Alcohol Exposure Demonstrate Deficits on Multiple Measures of Concept Formation

McGee

Schonfeld

Roebuck-Spencer

et al. 2008

Alcoholism Clin & Exp Res

View full text Add to dashboard Cite

show abstract

“…Much of the problematic usage of regression analysis is not derived from flaws in the methodology itself, but from the lack of consideration given to the requisite assumptions for drawing valid inferences based on the method, for example, failure of the strict exogeneity assumption (Porter 1999), or the failure to consider the consequences of diagnostic problems such as influence points or extrapolation off support of the data (Kuo 2002). Failures on any of these dimensions can lead to inferences that are seriously flawed.…”

Section: Introductionmentioning

confidence: 99%

Evolutionary regression? Assessing the problem of hidden biases in criminal justice applications using propensity scores

et al. 2015

View full text Add to dashboard Cite

Objectives Propensity score methods rely on an untestable assumption of unconfoundedness for making causal inference. Yet, empirical applications using propensity scores in criminology routinely invoke this assumption without careful scrutiny. Methods We use a dataset with a wide range of observable, potential confounders, which allows us to evaluate recidivism outcomes for adolescent offenders who are sentenced to either placement or probation. We then systematically withhold important known confounders from the matching process to demonstrate the effectiveness of sensitivity checks in sizing up the robustness of these treatment effect estimates in the case where hidden biases clearly exist. Results We find important variability in the estimated treatment effect, and a large degree of imbalance in 'unobserved' covariates, which we did not explicitly control for. The hidden biases observed in our controlled analysis would have at least been suggested in an actual application by the low gamma statistics that attended our analysis, a statistic that is not reported in most criminological applications of propensity score analysis. Conclusions Researchers who use propensity score methods should openly discuss potential limitations of their analysis due to hidden bias and report bias sensitivity checks based on the gamma statistic when statistically significant treatment effects are reported.

show abstract

“…Clearly, this dataset would have provided no support for the 'mixing vessel' hypothesis if all the data used to construct the phylogeny had also been used in the regression analysis. There is always a risk in drawing conclusions from extrapolation of a regression analysis (Kuo 2002). In the case of emerging infectious disease, this technique is especially suspect because extrapolation relies on the assumption of a constant rate of evolution over time.…”

Section: Retrospective Analyses Based On Phylogeneticsmentioning

confidence: 99%

Influenza as a model system for studying the cross–species transfer and evolution of the SARS coronavirus

Bush

2004

Phil. Trans. R. Soc. Lond. B

View full text Add to dashboard Cite

Severe acute respiratory syndrome coronavirus (SARS-CoV) moved into humans from a reservoir species and subsequently caused an epidemic in its new host. We know little about the processes that allowed the cross-species transfer of this previously unknown virus. I discuss what we have learned about the movement of viruses into humans from studies of influenza A, both how it crossed from birds to humans and how it subsequently evolved within the human population. Starting with a brief review of severe acute respiratory syndrome to highlight the kinds of problems we face in learning about this viral disease, I then turn to influenza A, focusing on three topics. First, I present a reanalysis of data used to test the hypothesis that swine served as a "mixing vessel" or intermediate host in the transmission of avian influenza to humans during the 1918 "Spanish flu" pandemic. Second, I review studies of archived viruses from the three recent influenza pandemics. Third, I discuss current limitations in using molecular data to study the evolution of infectious disease. Although influenza A and SARS-CoV differ in many ways, our knowledge of influenza A may provide important clues about what limits or favours cross-species transfers and subsequent epidemics of newly emerging pathogens.

show abstract

Extrapolation of Correlation Between 2 Variables in 4 General Medical Journals

Abstract: A high proportion of the articles analyzed from all 4 weekly general medical journals involved extrapolation without indication. Researchers, reviewers, and editors should be aware of this problem and work to eliminate it.

Cited by 13 publications

References 45 publications

Children With Heavy Prenatal Alcohol Exposure Demonstrate Deficits on Multiple Measures of Concept Formation

Children With Heavy Prenatal Alcohol Exposure Demonstrate Deficits on Multiple Measures of Concept Formation

Evolutionary regression? Assessing the problem of hidden biases in criminal justice applications using propensity scores

Influenza as a model system for studying the cross–species transfer and evolution of the SARS coronavirus

Contact Info

Product

Resources

About