Estimation accuracy in the psychological sciences.
Ontology highlight
ABSTRACT:
SUBMITTER: Davis-Stober CP
PROVIDER: S-EPMC6261010 | biostudies-literature | 2018
REPOSITORIES: biostudies-literature
Publications
Estimation accuracy in the psychological sciences.
Davis-Stober Clintin P CP Dana Jason J Rouder Jeffrey N JN
PloS one 20181126 11
Sample means comparisons are a fundamental and ubiquitous approach to interpreting experimental psychological data. Yet, we argue that the sample and effect sizes in published psychological research are frequently so small that sample means are insufficiently accurate to determine whether treatment effects have occurred. Generally, an estimator should be more accurate than any benchmark that systematically ignores information about the relations among experimental conditions. We consider two suc ...[more]
Similar Datasets
Project description:[This corrects the article DOI: 10.1371/journal.pone.0207239.].
| S-EPMC6394978 | biostudies-literature
Project description:An important goal for psychological science is developing methods to characterize relationships between variables. Customary approaches use structural equation models to connect latent factors to a number of observed measurements, or test causal hypotheses between observed variables. More recently, regularized partial correlation networks have been proposed as an alternative approach for characterizing relationships among variables through off-diagonal elements in the precision matrix. While the graphical Lasso (glasso) has emerged as the default network estimation method, it was optimized in fields outside of psychology with very different needs, such as high dimensional data where the number of variables (p) exceeds the number of observations (n). In this article, we describe the glasso method in the context of the fields where it was developed, and then we demonstrate that the advantages of regularization diminish in settings where psychological networks are often fitted ( p≪n ). We first show that improved properties of the precision matrix, such as eigenvalue estimation, and predictive accuracy with cross-validation are not always appreciable. We then introduce nonregularized methods based on multiple regression and a nonparametric bootstrap strategy, after which we characterize performance with extensive simulations. Our results demonstrate that the nonregularized methods can be used to reduce the false-positive rate, compared to glasso, and they appear to provide consistent performance across sparsity levels, sample composition (p/n), and partial correlation size. We end by reviewing recent findings in the statistics literature that suggest alternative methods often have superior performance than glasso, as well as suggesting areas for future research in psychology. The nonregularized methods have been implemented in the R package GGMnonreg.
| S-EPMC6736701 | biostudies-literature
Project description:High-risk mechanisms in trauma usually dictate certain treatment and evaluation in protocolized care. A 10-15 feet (ft) fall is traditionally cited as an example of a high-risk mechanism, triggering trauma team activations and costly work-ups. The height and other details of mechanism are usually reported by lay bystanders or prehospital personnel. This small observational study was designed to evaluate how accurate or inaccurate height estimation may be among typical bystanders.This was a blinded, prospective study conducted on the grounds of a community hospital. Four panels with lines corresponding to varying heights from 1-25 ft were hung within a building structure that did not have stories or other possibly confounding factors by which to judge height. The participants were asked to estimate the height of each line using a multiple-choice survey-style ballot. Participants were adult volunteers composed of various hospital and non-hospital affiliated persons, of varying ages and genders. In total, there were 96 respondents.For heights equal to or greater than 15 ft, less than 50% of participants of each job description were able to correctly identify the height. When arranged into a scatter plot, as height increased, the likelihood to underestimate the correct height was evident, having a strong correlation coefficient (R=+0.926) with a statistically significant p value = <0.001.The use of vertical height as a predictor of injury severity is part of current practice in trauma triage. This data is often an estimation provided by prehospital personnel or bystanders. Our small study showed bystanders may not estimate heights accurately in the field. The greater the reported height, the less likely it is to be accurate. Additionally, there is a higher likelihood that falls from greater than 15 ft may be underestimated.
| S-EPMC6123090 | biostudies-other
Project description:Estimates of prehospital transport times are an important part of emergency care system research and planning; however, the accuracy of these estimates is unknown. The authors examined the accuracy of three estimation methods against observed transport times in a large cohort of prehospital patient transports.This was a validation study using prehospital records in King County, Washington, and southwestern Pennsylvania from 2002 to 2006 and 2005 to 2011, respectively. Transport time estimates were generated using three methods: linear arc distance, Google Maps, and ArcGIS Network Analyst. Estimation error, defined as the absolute difference between observed and estimated transport time, was assessed, as well as the proportion of estimated times that were within specified error thresholds. Based on the primary results, a regression estimate was used that incorporated population density, time of day, and season to assess improved accuracy. Finally, hospital catchment areas were compared using each method with a fixed drive time.The authors analyzed 29,935 prehospital transports to 44 hospitals. The mean (± standard deviation [±SD]) absolute error was 4.8 (±7.3) minutes using linear arc, 3.5 (±5.4) minutes using Google Maps, and 4.4 (±5.7) minutes using ArcGIS. All pairwise comparisons were statistically significant (p < 0.01). Estimation accuracy was lower for each method among transports more than 20 minutes (mean [±SD] absolute error was 12.7 [±11.7] minutes for linear arc, 9.8 [±10.5] minutes for Google Maps, and 11.6 [±10.9] minutes for ArcGIS). Estimates were within 5 minutes of observed transport time for 79% of linear arc estimates, 86.6% of Google Maps estimates, and 81.3% of ArcGIS estimates. The regression-based approach did not substantially improve estimation. There were large differences in hospital catchment areas estimated by each method.Route-based transport time estimates demonstrate moderate accuracy. These methods can be valuable for informing a host of decisions related to the system organization and patient access to emergency medical care; however, they should be employed with sensitivity to their limitations.
| S-EPMC4251659 | biostudies-literature
Project description:Background and objectivesAdequate estimation of renal function in obese patients is essential for the classification of patients in CKD category as well as the dose adjustment of drugs. However, the body size descriptor for GFR indexation is still debatable, and formulas are not validated in patients with extreme variations of weight.Design, setting, participants, & measurementsThis study included 209 stages 1-5 CKD obese patients referred to the Department of Renal Function Study at the University Hospital in Lyon between 2010 and 2013 because of suspected renal dysfunction. GFR was estimated with the Chronic Kidney Disease and Epidemiology equation (CKD-EPI) and measured with a gold standard method (inulin or iohexol) not indexed (mGFR) or indexed to body surface area determined by the Dubois and Dubois formula with either real (mGFRr) or ideal (mGFRi) body weight. Mean bias (eGFR-mGFR), precision, and accuracy of mGFR were compared with the results obtained for nonobese participants (body mass index between 18.5 and 24.9) who had a GFR measurement during the same period of time.ResultsMean mGFRr (51.6 ± 24.2 ml/min per 1.73 m(2)) was significantly lower than mGFR, mGFRi, and eGFRCKD-EPI. eGFRCKD-EPI had less bias with mGFR (0.29; -1.7 to 2.3) and mGFRi (-1.62; -3.1 to 0.45) compared with mGFRr (8.7; 7 to 10). This result was confirmed with better accuracy for the whole cohort (78% for mGFR, 84% for mGFRi, and 72% for mGFRr) and participants with CKD stages 3-5. Moreover, the Bland Altman plot showed better agreement between mGFR and eGFRCKD-EPI. The bias between eGFRCKD-EPI and mGFRr was greater in obese than nonobese participants (8.7 versus 0.58, P<0.001).ConclusionsThis study shows that, in obese CKD patients, the performance of eGFRCKD-EPI is good for GFR ? 60 ml/min per 1.73 m(2). Indexation of mGFR with body surface area using ideal body weight gives less bias than mGFR scaled with body surface area using real body weight.
| S-EPMC3974350 | biostudies-literature
Project description:BackgroundFoamy viruses (FVs) are the most genetically stable viruses of the retrovirus family. This is in contrast to the in vitro error rate found for recombinant FV reverse transcriptase (RT). To investigate the accuracy of FV genome copying in vivo we analyzed the occurrence of mutations in HEK 293T cell culture after a single round of reverse transcription using a replication-deficient vector system. Furthermore, the frequency of FV recombination by template switching (TS) and the cross-packaging ability of different FV strains were analyzed.ResultsWe initially sequenced 90,000 nucleotides and detected 39 mutations, corresponding to an in vivo error rate of approximately 4 x 10-4 per site per replication cycle. Surprisingly, all mutations were transitions from G to A, suggesting that APOBEC3 activity is the driving force for the majority of mutations detected in our experimental system. In line with this, we detected a late but significant APOBEC3G and 3F mRNA by quantitative PCR in the cells. We then analyzed 170,000 additional nucleotides from experiments in which we co-transfected the APOBEC3-interfering foamy viral bet gene and observed a significant 50% drop in G to A mutations, indicating that APOBEC activity indeed contributes substantially to the foamy viral replication error rate in vivo. However, even in the presence of Bet, 35 out of 37 substitutions were G to A, suggesting that residual APOBEC activity accounted for most of the observed mutations. If we subtract these APOBEC-like mutations from the total number of mutations, we calculate a maximal intrinsic in vivo error rate of 1.1 x 10-5 per site per replication. In addition to the point mutations, we detected one 49 bp deletion within the analyzed 260000 nucleotides.Analysis of the recombination frequency of FV vector genomes revealed a 27% probability for a template switching (TS) event within a 1 kilobase (kb) region. This corresponds to a 98% probability that FVs undergo at least one additional TS event per replication cycle. We also show that a given FV particle is able to cross-transfer a heterologous FV genome, although at reduced efficiency than the homologous vector.ConclusionOur results indicate that the copying of the FV genome is more accurate than previously thought. On the other hand recombination among FV genomes appears to be a frequent event.
| S-EPMC2678077 | biostudies-literature
Project description:MotivationModel quality assessment programs are used to predict the quality of modeled protein structures. They can be divided into two groups depending on the information they are using: ensemble methods using consensus of many alternative models and methods only using a single model to do its prediction. The consensus methods excel in achieving high correlations between prediction and true quality measures. However, they frequently fail to pick out the best possible model, nor can they be used to generate and score new structures. Single-model methods on the other hand do not have these inherent shortcomings and can be used both to sample new structures and to improve existing consensus methods.ResultsHere, we present an implementation of the ProQ2 program to estimate both local and global model accuracy as part of the Rosetta modeling suite. The current implementation does not only make it possible to run large batch runs locally, but it also opens up a whole new arena for conformational sampling using machine learned scoring functions and to incorporate model accuracy estimation in to various existing modeling schemes. ProQ2 participated in CASP11 and results from CASP11 are used to benchmark the current implementation. Based on results from CASP11 and CAMEO-QE, a continuous benchmark of quality estimation methods, it is clear that ProQ2 is the single-model method that performs best in both local and global model accuracy.Availability and implementationhttps://github.com/bjornwallner/ProQ_scriptsContactbjornw@ifm.liu.seSupplementary informationSupplementary data are available at Bioinformatics online.
| S-EPMC4848402 | biostudies-literature
Project description:Age determination of unknown human bodies is important in the setting of a crime investigation or a mass disaster because the age at death, birth date, and year of death as well as gender can guide investigators to the correct identity among a large number of possible matches. Traditional morphological methods used by anthropologists to determine age are often imprecise, whereas chemical analysis of tooth dentin, such as aspartic acid racemization, has shown reproducible and more precise results. In this study, we analyzed teeth from Swedish individuals using both aspartic acid racemization and radiocarbon methodologies. The rationale behind using radiocarbon analysis is that aboveground testing of nuclear weapons during the cold war (1955-1963) caused an extreme increase in global levels of carbon-14 ((14)C), which has been carefully recorded over time. Forty-four teeth from 41 individuals were analyzed using aspartic acid racemization analysis of tooth crown dentin or radiocarbon analysis of enamel, and 10 of these were split and subjected to both radiocarbon and racemization analysis. Combined analysis showed that the two methods correlated well (R(2) = 0.66, p < 0.05). Radiocarbon analysis showed an excellent precision with an overall absolute error of 1.0 +/- 0.6 years. Aspartic acid racemization also showed a good precision with an overall absolute error of 5.4 +/- 4.2 years. Whereas radiocarbon analysis gives an estimated year of birth, racemization analysis indicates the chronological age of the individual at the time of death. We show how these methods in combination can also assist in the estimation of date of death of an unidentified victim. This strategy can be of significant assistance in forensic casework involving dead victim identification.
| S-EPMC2871409 | biostudies-literature
Project description:The usage of psychological networks that conceptualize behavior as a complex interplay of psychological and other components has gained increasing popularity in various research fields. While prior publications have tackled the topics of estimating and interpreting such networks, little work has been conducted to check how accurate (i.e., prone to sampling variation) networks are estimated, and how stable (i.e., interpretation remains similar with less observations) inferences from the network structure (such as centrality indices) are. In this tutorial paper, we aim to introduce the reader to this field and tackle the problem of accuracy under sampling variation. We first introduce the current state-of-the-art of network estimation. Second, we provide a rationale why researchers should investigate the accuracy of psychological networks. Third, we describe how bootstrap routines can be used to (A) assess the accuracy of estimated network connections, (B) investigate the stability of centrality indices, and (C) test whether network connections and centrality estimates for different variables differ from each other. We introduce two novel statistical methods: for (B) the correlation stability coefficient, and for (C) the bootstrapped difference test for edge-weights and centrality indices. We conducted and present simulation studies to assess the performance of both methods. Finally, we developed the free R-package bootnet that allows for estimating psychological networks in a generalized framework in addition to the proposed bootstrap methods. We showcase bootnet in a tutorial, accompanied by R syntax, in which we analyze a dataset of 359 women with posttraumatic stress disorder available online.
| S-EPMC5809547 | biostudies-literature
Project description:Study on the MS1 interference problem for reporter ion quantification using orthogonal proteomes: human and pyrococcus furiosus.
2012-08-13 | PXD000013 | Pride