1
|
Yu X, Zoh RS, Fluharty DA, Mestre LM, Valdez D, Tekwe CD, Vorland CJ, Jamshidi-Naeini Y, Chiou SH, Lartey ST, Allison DB. Misstatements, misperceptions, and mistakes in controlling for covariates in observational research. eLife 2024; 13:e82268. [PMID: 38752987 PMCID: PMC11098558 DOI: 10.7554/elife.82268] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/29/2022] [Accepted: 04/02/2024] [Indexed: 05/18/2024] Open
Abstract
We discuss 12 misperceptions, misstatements, or mistakes concerning the use of covariates in observational or nonrandomized research. Additionally, we offer advice to help investigators, editors, reviewers, and readers make more informed decisions about conducting and interpreting research where the influence of covariates may be at issue. We primarily address misperceptions in the context of statistical management of the covariates through various forms of modeling, although we also emphasize design and model or variable selection. Other approaches to addressing the effects of covariates, including matching, have logical extensions from what we discuss here but are not dwelled upon heavily. The misperceptions, misstatements, or mistakes we discuss include accurate representation of covariates, effects of measurement error, overreliance on covariate categorization, underestimation of power loss when controlling for covariates, misinterpretation of significance in statistical models, and misconceptions about confounding variables, selecting on a collider, and p value interpretations in covariate-inclusive analyses. This condensed overview serves to correct common errors and improve research quality in general and in nutrition research specifically.
Collapse
Affiliation(s)
- Xiaoxin Yu
- Department of Epidemiology and Biostatistics, Indiana University School of Public Health-BloomingtonBloomingtonUnited States
| | - Roger S Zoh
- Department of Epidemiology and Biostatistics, Indiana University School of Public Health-BloomingtonBloomingtonUnited States
| | - David A Fluharty
- Department of Epidemiology and Biostatistics, Indiana University School of Public Health-BloomingtonBloomingtonUnited States
| | - Luis M Mestre
- Department of Epidemiology and Biostatistics, Indiana University School of Public Health-BloomingtonBloomingtonUnited States
| | - Danny Valdez
- Department of Applied Health Science, Indiana University School of Public Health-BloomingtonBloomingtonUnited States
| | - Carmen D Tekwe
- Department of Epidemiology and Biostatistics, Indiana University School of Public Health-BloomingtonBloomingtonUnited States
| | - Colby J Vorland
- Department of Applied Health Science, Indiana University School of Public Health-BloomingtonBloomingtonUnited States
| | - Yasaman Jamshidi-Naeini
- Department of Epidemiology and Biostatistics, Indiana University School of Public Health-BloomingtonBloomingtonUnited States
| | - Sy Han Chiou
- Department of Statistics and Data Science, Southern Methodist UniversityDallasUnited States
| | - Stella T Lartey
- University of Memphis, School of Public HealthMemphisUnited Kingdom
| | - David B Allison
- Department of Epidemiology and Biostatistics, Indiana University School of Public Health-BloomingtonBloomingtonUnited States
| |
Collapse
|
2
|
Tchoundjeu S, Bigerelle M, Robbe-Valloire F, Da Silva Botelho T, Jarnias F. How to Select 2D and 3D Roughness Parameters at Their Relevant Scales by the Analysis of Covariance. MATERIALS (BASEL, SWITZERLAND) 2020; 13:ma13071526. [PMID: 32225008 PMCID: PMC7177462 DOI: 10.3390/ma13071526] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 02/04/2020] [Revised: 03/10/2020] [Accepted: 03/11/2020] [Indexed: 06/10/2023]
Abstract
In this paper, a multi-scale methodology is proposed to model and characterize the effect of two lubricants on changes in surface morphology during a running-in test. The test concerns two steels samples, mounted on a twin-disc tribometer to test each of lubricants A and B for a period of 42 h. The changes are characterized by the standardized roughness parameters given in ISO 25178. A technique involving replication is used to monitor wear during the test. Using all these replication measurements, a multi-scale methodology is applied. These selected models highlighted the relevant parameters for quantifying wear during lifespan, and also showed that lubricant A was better able to preserve surface integrity during wear than lubricant B.
Collapse
Affiliation(s)
| | - Maxence Bigerelle
- Laboratory of Industrial and Human Automation control, Mechanical engineering and Computer Science, UMR 8201, Polytechnic University of Hauts-de-France, 59313 Valenciennes, France
| | - Francois Robbe-Valloire
- Laboratory Quartz EA 7393, Supméca, 93400 Saint-Ouen-sur-Seine, France; (F.R.-V.); (T.D.S.B.)
| | - Tony Da Silva Botelho
- Laboratory Quartz EA 7393, Supméca, 93400 Saint-Ouen-sur-Seine, France; (F.R.-V.); (T.D.S.B.)
| | - Frederic Jarnias
- TOTAL Marketing & Services, Centre de recherche de Solaize, 69360 Solaize, France;
| |
Collapse
|
3
|
Harrison P, Hardy GE, Barkham M. The relationship between expected engagement and talking therapy outcome. Clin Psychol Psychother 2019; 26:492-501. [PMID: 31018017 PMCID: PMC6772155 DOI: 10.1002/cpp.2369] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/07/2018] [Revised: 03/12/2019] [Accepted: 04/12/2019] [Indexed: 11/10/2022]
Abstract
The aim of the study was to investigate whether client‐reported expected engagement with therapy predicted therapy outcome. It was hypothesized that higher expected engagement with cognitive behavioural therapy (CBT) or person‐centred experiential therapy (PCET) would predict more symptomatic improvement following therapy and higher likelihood of therapy completion. The Sheffield Expected Engagement with Therapy Scale was administered to 96 clients at pre‐therapy assessment with all meeting a diagnosis of moderate or severe depression with 53 receiving CBT and 43 receiving PCET. Higher expected engagement predicted more symptomatic improvement in CBT but not PCET. Expected engagement only predicted improvement in CBT when clients rated the credibility of CBT as low or moderate. Expected engagement did not predict therapy completion in either therapy. Assessment of expected engagement could be a useful tool in prediction of symptomatic improvement in CBT.
Collapse
Affiliation(s)
- Phillippa Harrison
- Department of Psychology, University of Sheffield, Sheffield, UK.,Centre for Affective Disorders, Institute of Psychiatry, Psychology & Neuroscience, King's College London, London, UK
| | - Gillian E Hardy
- Department of Psychology, University of Sheffield, Sheffield, UK
| | - Michael Barkham
- Department of Psychology, University of Sheffield, Sheffield, UK
| |
Collapse
|
4
|
Leppink J. Analysis of Covariance (ANCOVA) vs. Moderated Regression (MODREG): Why the Interaction Matters. HEALTH PROFESSIONS EDUCATION 2018. [DOI: 10.1016/j.hpe.2018.04.001] [Citation(s) in RCA: 11] [Impact Index Per Article: 1.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/17/2022] Open
|
5
|
Valente MJ, MacKinnon DP. Comparing models of change to estimate the mediated effect in the pretest-posttest control group design. STRUCTURAL EQUATION MODELING : A MULTIDISCIPLINARY JOURNAL 2017; 24:428-450. [PMID: 28845097 PMCID: PMC5568008 DOI: 10.1080/10705511.2016.1274657] [Citation(s) in RCA: 68] [Impact Index Per Article: 9.7] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/10/2023]
Abstract
Models to assess mediation in the pretest-posttest control group design are understudied in the behavioral sciences even though it is the design of choice for evaluating experimental manipulations. The paper provides analytical comparisons of the four most commonly used models used to estimate the mediated effect in this design: Analysis of Covariance (ANCOVA), difference score, residualized change score, and cross-sectional model. Each of these models are fitted using a Latent Change Score specification and a simulation study assessed bias, Type I error, power, and confidence interval coverage of the four models. All but the ANCOVA model make stringent assumptions about the stability and cross-lagged relations of the mediator and outcome that may not be plausible in real-world applications. When these assumptions do not hold, Type I error and statistical power results suggest that only the ANCOVA model has good performance. The four models are applied to an empirical example.
Collapse
|
6
|
Pfister R, Obhi SS, Rieger M, Wenke D. Action and perception in social contexts: intentional binding for social action effects. Front Hum Neurosci 2014; 8:667. [PMID: 25228869 PMCID: PMC4151154 DOI: 10.3389/fnhum.2014.00667] [Citation(s) in RCA: 32] [Impact Index Per Article: 3.2] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/25/2014] [Accepted: 08/11/2014] [Indexed: 11/29/2022] Open
Abstract
The subjective experience of controlling events in the environment alters the perception of these events. For instance, the interval between one's own actions and their consequences is subjectively compressed—a phenomenon known as intentional binding. In two experiments, we studied intentional binding in a social setting in which actions of one agent prompted a second agent to perform another action. Participants worked in pairs and were assigned to a “leader” and a “follower” role, respectively. The leader's key presses triggered (after a variable interval) a tone and this tone served as go signal for the follower to perform a keypress as well. Leaders and followers estimated the interval between the leader's keypress and the following tone, or the interval between the tone and the follower's keypress. The leader showed reliable intentional binding for both intervals relative to the follower's estimates. These results indicate that human agents experience a pre-reflective sense of agency for genuinely social consequences of their actions.
Collapse
Affiliation(s)
- Roland Pfister
- Department of Psychology III, Julius Maximilians University of Würzburg Würzburg, Germany
| | - Sukhvinder S Obhi
- Department of Psychology, Wilfrid Laurier University Waterloo, ON, Canada
| | - Martina Rieger
- Institute for Psychology, UMIT, University for Health Sciences, Medical Informatics and Technology Hall in Tirol, Austria ; Department of Psychology, Max-Planck-Institute for Human Cognitive and Brain Sciences Leipzig, Germany
| | - Dorit Wenke
- Department of Psychology, Max-Planck-Institute for Human Cognitive and Brain Sciences Leipzig, Germany ; Department of Psychology, Humboldt University at Berlin Berlin, Germany
| |
Collapse
|
7
|
Kisbu-Sakarya Y, MacKinnon DP, Aiken LS. A Monte Carlo Comparison Study of the Power of the Analysis of Covariance, Simple Difference, and Residual Change Scores in Testing Two-Wave Data. EDUCATIONAL AND PSYCHOLOGICAL MEASUREMENT 2013; 73:47-62. [PMID: 26412869 PMCID: PMC4583144 DOI: 10.1177/0013164412450574] [Citation(s) in RCA: 31] [Impact Index Per Article: 2.8] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 05/13/2023]
Abstract
This study compares the analysis of covariance (ANCOVA), difference score, and residual change score methods in testing the group effect for pretest–posttest data in terms of statistical power and Type I error rates using a Monte Carlo simulation. Previous research has mathematically shown the effect of stability of individual scores from pretest to posttest, reliability, and nonrandomization (i.e., pretest imbalance) on the performance of the ANCOVA, difference score, and residual change score methods. However, related power issues have not been adequately addressed. The authors examined the impact of stability of measurement over time, reliability of covariate and criterion, nonrandomization, sample size, and treatment effect size on statistical power of the three methods. Across conditions, ANCOVA and residual change score methods had similar power rates. When reliability was less than perfect, ANCOVA had more power than the difference score method when there was an increase from pretest to posttest and a positive baseline imbalance (i.e., treatment group had higher pretest scores than the control group), or when there was a decrease from pretest to posttest and a negative baseline imbalance, and vice versa. In case of perfect reliability, the statistical power of ANCOVA did not differ from the difference score method. For the difference score method, when reliability was low, there was no effect of stability on power, whereas when reliability was high or perfect, power increased as stability increased for medium and large effect sizes. Difference scores may be preferred over ANCOVA under certain circumstances.
Collapse
|
8
|
Hwang DM, Kim DJ. Nonparametric Method using Placement in an Analysis of a Covariance Model. COMMUNICATIONS FOR STATISTICAL APPLICATIONS AND METHODS 2012. [DOI: 10.5351/ckss.2012.19.5.721] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/11/2022]
|
9
|
Pfister R. Gender effects in gaming research: a case for regression residuals? CYBERPSYCHOLOGY BEHAVIOR AND SOCIAL NETWORKING 2011; 14:603-6. [PMID: 21486141 DOI: 10.1089/cyber.2010.0547] [Citation(s) in RCA: 7] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/13/2022]
Abstract
Numerous recent studies have examined the impact of video gaming on various dependent variables, including the players' affective reactions, positive as well as detrimental cognitive effects, and real-world aggression. These target variables are typically analyzed as a function of game characteristics and player attributes-especially gender. However, findings on the uneven distribution of gaming experience between males and females, on the one hand, and the effect of gaming experience on several target variables, on the other hand, point at a possible confound when gaming experiments are analyzed with a standard analysis of variance. This study uses simulated data to exemplify analysis of regression residuals as a potentially beneficial data analysis strategy for such datasets. As the actual impact of gaming experience on each of the various dependent variables differs, the ultimate benefits of analysis of regression residuals entirely depend on the research question, but it offers a powerful statistical approach to video game research whenever gaming experience is a confounding factor.
Collapse
Affiliation(s)
- Roland Pfister
- Department of Psychology, University of Würzburg, Würzburg, Germany.
| |
Collapse
|
10
|
Pandit JJ. The analysis of variance in anaesthetic research: statistics, biography and history. Anaesthesia 2010; 65:1212-20. [DOI: 10.1111/j.1365-2044.2010.06542.x] [Citation(s) in RCA: 43] [Impact Index Per Article: 3.1] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/28/2022]
|
11
|
Twisk JWR, de Vente W. The analysis of randomised controlled trial data with more than one follow-up measurement. A comparison between different approaches. Eur J Epidemiol 2008; 23:655-60. [DOI: 10.1007/s10654-008-9279-6] [Citation(s) in RCA: 25] [Impact Index Per Article: 1.6] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/20/2008] [Accepted: 07/18/2008] [Indexed: 11/27/2022]
|
12
|
García-Berthou E. On the misuse of residuals in ecology: testing regression residuals vs. the analysis of covariance. J Anim Ecol 2001. [DOI: 10.1046/j.1365-2656.2001.00524.x] [Citation(s) in RCA: 413] [Impact Index Per Article: 18.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/20/2022]
|