Johannes Ledolter
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Johannes Ledolter has a total of 52 co-authors. Here are the top co-authors by number of shared publications:
- Franklin Dexter
Department of Anesthesia, University of Iowa, Iowa City, Iowa
13 shared publications - Richard H. Epstein
Department of Anesthesiology, Perioperative Medicine & Pain Management, University of Miami, Miller School of Medicine, Miami, FL, USA
5 shared publications Randy H. Kardon
Department of Ophthalmology and Visual Sciences, University of Iowa, Iowa City, IA, United States
4 shared publications- Randy H. Kardon
Department of Ophthalmology and Visual Sciences, University of Iowa, Iowa City, USA
4 shared publications Arthur J. Swersey
Operations Research, Yale School of Management, New Haven, CT 06520, USA
2 shared publications- Bradley J. Hindman
Department of Anesthesia, University of Iowa, United States of America
2 shared publications
Latest Publications
- 19 April 2023
- Published by Springer Nature in Journal of Medical Systems
- Vol. 47 (1) , 1-10
Exceedance Probabilities of Log-normal Distributions for One Group, Two Groups, and Meta-analysis of Multiple Two-group Studies, With Application to Analyses of Prolonged Times to Tracheal Extubation
Many randomized trials measure means and standard deviations of anesthesia recovery time (e.g., times to tracheal extubation). We show how to use generalized pivotal methods to compare the probabilities of exceeding a tolerance limit (e.g., > 15 min, prolonged times to tracheal extubation). The topic is important because the economic benefits of faster anesthesia emergence depend on reducing variability, not means, especially prevention of very long recovery times. Generalized pivotal methods are applied using computer simulation (e.g., using two Excel formulas for one group and three formulas for two group comparisons). The endpoint for each study with two groups is the ratio between groups of the probabilities of times exceeding a threshold or the ratio of the standard deviations. Confidence intervals and variances for the incremental risk ratio of the exceedance probabilities and for ratios of standard deviations are calculated using studies’ sample sizes, sample means in the time scale of recovery times, and sample standard deviations in the time scale. Ratios are combined among studies using the DerSimonian-Laird estimate of the heterogeneity variance estimate, with Knapp-Hartung adjustment for the relatively small (N = 15) numbers of studies in the meta-analysis. We show larger absolute variability among studies’ results when analyzed based on exceedance probabilities rather than standard deviations. Therefore, if an investigator’s primary goal is to quantify reductions in the variability of recovery times (e.g., times until patients are ready for post-anesthesia care unit discharge), we recommend analyzing the standard deviations. When exceedance probabilities themselves are relevant, they can be analyzed from the original studies’ summary measures.
- 19 April 2023
- Published by Springer Nature in Journal of Medical Systems
- Vol. 47 (1) , 1-10
Exceedance Probabilities of Log-normal Distributions for One Group, Two Groups, and Meta-analysis of Multiple Two-group Studies, With Application to Analyses of Prolonged Times to Tracheal Extubation
Many randomized trials measure means and standard deviations of anesthesia recovery time (e.g., times to tracheal extubation). We show how to use generalized pivotal methods to compare the probabilities of exceeding a tolerance limit (e.g., > 15 min, prolonged times to tracheal extubation). The topic is important because the economic benefits of faster anesthesia emergence depend on reducing variability, not means, especially prevention of very long recovery times. Generalized pivotal methods are applied using computer simulation (e.g., using two Excel formulas for one group and three formulas for two group comparisons). The endpoint for each study with two groups is the ratio between groups of the probabilities of times exceeding a threshold or the ratio of the standard deviations. Confidence intervals and variances for the incremental risk ratio of the exceedance probabilities and for ratios of standard deviations are calculated using studies’ sample sizes, sample means in the time scale of recovery times, and sample standard deviations in the time scale. Ratios are combined among studies using the DerSimonian-Laird estimate of the heterogeneity variance estimate, with Knapp-Hartung adjustment for the relatively small (N = 15) numbers of studies in the meta-analysis. We show larger absolute variability among studies’ results when analyzed based on exceedance probabilities rather than standard deviations. Therefore, if an investigator’s primary goal is to quantify reductions in the variability of recovery times (e.g., times until patients are ready for post-anesthesia care unit discharge), we recommend analyzing the standard deviations. When exceedance probabilities themselves are relevant, they can be analyzed from the original studies’ summary measures.
- Open Access
- 7 March 2023
- Published by Austrian Statistical Society in Austrian Journal of Statistics
- Vol. 52 (1) , 87-100
A Statistical Analysis of the Water Levels at Lake Neusiedl
The recent low water levels of Lake Neusiedl have raised concern about the lake's future and have sparked interest on how meteorological variables impact the water level. Data from the last 50 years are used to study the impact of rain and temperature on the cyclical changes of the water levels. Linear and nonlinear regression models are used to describe how rainfall and temperature change the water level, and to assess the future progression of the water level under three scenarios. For average 2022 meteorological conditions the water levels in 2022 may not recover for a dry 2021 autumn, but will recover substantially for a wet ending to the year 2021.
- 7 March 2023
- Published by Austrian Statistical Society in Austrian Journal of Statistics
- Vol. 52 (1) , 87-100
Open AccessA Statistical Analysis of the Water Levels at Lake Neusiedl
The recent low water levels of Lake Neusiedl have raised concern about the lake's future and have sparked interest on how meteorological variables impact the water level. Data from the last 50 years are used to study the impact of rain and temperature on the cyclical changes of the water levels. Linear and nonlinear regression models are used to describe how rainfall and temperature change the water level, and to assess the future progression of the water level under three scenarios. For average 2022 meteorological conditions the water levels in 2022 may not recover for a dry 2021 autumn, but will recover substantially for a wet ending to the year 2021.
- 24 November 2022
- Published by Elsevier in Perioperative Care and Operating Room Management
- Vol. 30
Variability of the times remaining in surgical cases and the importance of knowing when closure has started
- 24 November 2022
- Published by Elsevier in Perioperative Care and Operating Room Management
- Vol. 30
Variability of the times remaining in surgical cases and the importance of knowing when closure has started
- Open Access
- 25 October 2022
- Published by Springer Nature in Cureus
- Vol. 14 (10) , e30683
Patients in Iowa Counties Lacking Hospitals With Labor and Delivery Services Disproportionately Receive Care at Level III Maternal Care Hospitals When Undergoing Cesarean Delivery: A Retrospective Longitudinal Study
Many obstetrical patients from rural areas in the United States lack hospitals that provide labor and delivery care. Our objective was to examine the effects of such patients on caseloads of cesarean deliveries at Iowa hospitals with level III maternal care, as defined by the Iowa Department of Public Health (e.g., with obstetric anesthesiologists). This retrospective longitudinal study included every discharge with cesarean delivery in the state of Iowa from October 2015 through June 2021. There were N=60,534 such deliveries from 76 hospitals, of which three were level III, and the rest were level I or II. Poisson regression models with robust variance estimation and controlling for geography, maternal risk factors, and insurance, were used to evaluate the binary outcome of whether patients received care at the university level III hospital in Eastern Iowa, or not. Similar models were also developed for care at the two private level III hospitals in Central Iowa, or not. Differences in the mean probabilities of receiving care at the level III hospitals were then estimated using logistic regression, with results reported in units of changes in cases per week at the hospitals. Statewide, the university level III hospital performed 7.4% of the cesarean deliveries, and the two private level III hospitals performed 23.4%. Patients from counties in which no cesarean deliveries were performed during the quarter of the year when they underwent a cesarean delivery disproportionately received care at level III hospitals versus levels I and II hospitals. Lower 99% confidence limits for incremental risk ratios were 1.46 and 4.20, respectively. Cesarean deliveries among patients residing in counties where no hospital had a labor and delivery ward were distributed unequally between the counties of the hospitals with level III maternal care. There were approximately 1.09 (standard error 0.10) extra cesarean deliveries per week at the university hospital versus 5.81 (standard error 0.11) at the private hospitals. The 1.09 vs 5.81 difference was caused, in part, by the effects of insurance and other hospitals with similar services. Patients residing in counties without labor and delivery care disproportionately go to level III hospitals. These results can help anesthesiologists, obstetricians, and analysts at hospitals with large tertiary (level III) programs interpret their annual increases in total obstetric anesthesia activity.
- 25 October 2022
- Published by Springer Nature in Cureus
- Vol. 14 (10) , e30683
Open AccessPatients in Iowa Counties Lacking Hospitals With Labor and Delivery Services Disproportionately Receive Care at Level III Maternal Care Hospitals When Undergoing Cesarean Delivery: A Retrospective Longitudinal Study
Many obstetrical patients from rural areas in the United States lack hospitals that provide labor and delivery care. Our objective was to examine the effects of such patients on caseloads of cesarean deliveries at Iowa hospitals with level III maternal care, as defined by the Iowa Department of Public Health (e.g., with obstetric anesthesiologists). This retrospective longitudinal study included every discharge with cesarean delivery in the state of Iowa from October 2015 through June 2021. There were N=60,534 such deliveries from 76 hospitals, of which three were level III, and the rest were level I or II. Poisson regression models with robust variance estimation and controlling for geography, maternal risk factors, and insurance, were used to evaluate the binary outcome of whether patients received care at the university level III hospital in Eastern Iowa, or not. Similar models were also developed for care at the two private level III hospitals in Central Iowa, or not. Differences in the mean probabilities of receiving care at the level III hospitals were then estimated using logistic regression, with results reported in units of changes in cases per week at the hospitals. Statewide, the university level III hospital performed 7.4% of the cesarean deliveries, and the two private level III hospitals performed 23.4%. Patients from counties in which no cesarean deliveries were performed during the quarter of the year when they underwent a cesarean delivery disproportionately received care at level III hospitals versus levels I and II hospitals. Lower 99% confidence limits for incremental risk ratios were 1.46 and 4.20, respectively. Cesarean deliveries among patients residing in counties where no hospital had a labor and delivery ward were distributed unequally between the counties of the hospitals with level III maternal care. There were approximately 1.09 (standard error 0.10) extra cesarean deliveries per week at the university hospital versus 5.81 (standard error 0.11) at the private hospitals. The 1.09 vs 5.81 difference was caused, in part, by the effects of insurance and other hospitals with similar services. Patients residing in counties without labor and delivery care disproportionately go to level III hospitals. These results can help anesthesiologists, obstetricians, and analysts at hospitals with large tertiary (level III) programs interpret their annual increases in total obstetric anesthesia activity.
- Open Access
- 29 June 2022
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 11 (13) , e023912
Sex‐Specific Differences in Endothelial Function Are Driven by Divergent Mitochondrial Ca 2+ Handling
Background: Sex‐specific differences in vasodilation are mediated in part by differences in cytosolic Ca 2+ handling, but how variations in mitochondrial Ca 2+ contributes to this effect remains unknown. Here, we investigated the extent to which mitochondrial Ca 2+ entry via the MCU (mitochondrial Ca 2+ uniporter) drives sex differences in vasoreactivity in resistance arteries. Methods and Results: Enhanced vasodilation of mesenteric resistance arteries to acetylcholine (ACh) was reduced to larger extent in female compared with male mice in 2 genetic models of endothelial MCU ablation. Ex vivo Ca 2+ imaging of mesenteric arteries with Fura‐2AM confirmed higher cytosolic Ca 2+ transients triggered by ACh in arteries from female mice versus male mice. MCU inhibition both strongly reduced cytosolic Ca 2+ transients and blocked mitochondrial Ca 2+ entry. In cultured human aortic endothelial cells, treatment with physiological concentrations of estradiol enhanced cytosolic Ca 2+ transients, Ca 2+ buffering capacity, and mitochondrial Ca 2+ entry in response to ATP or repeat Ca 2+ boluses. Further experiments to establish the mechanisms underlying these effects did not reveal significant differences in the expression of MCU subunits, at either the mRNA or protein level. However, estradiol treatment was associated with an increase in mitochondrial mass, mitochondrial fusion, and the mitochondrial membrane potential and reduced mitochondrial superoxide production. Conclusions: Our data confirm that mitochondrial function in endothelial cells differs by sex, with female mice having enhanced Ca 2+ uptake capacity, and that these differences are attributable to the presence of more mitochondria and a higher mitochondrial membrane potential in female mice rather than differences in composition of the MCU complex.
- 29 June 2022
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 11 (13) , e023912
Open AccessSex‐Specific Differences in Endothelial Function Are Driven by Divergent Mitochondrial Ca 2+ Handling
Background: Sex‐specific differences in vasodilation are mediated in part by differences in cytosolic Ca 2+ handling, but how variations in mitochondrial Ca 2+ contributes to this effect remains unknown. Here, we investigated the extent to which mitochondrial Ca 2+ entry via the MCU (mitochondrial Ca 2+ uniporter) drives sex differences in vasoreactivity in resistance arteries. Methods and Results: Enhanced vasodilation of mesenteric resistance arteries to acetylcholine (ACh) was reduced to larger extent in female compared with male mice in 2 genetic models of endothelial MCU ablation. Ex vivo Ca 2+ imaging of mesenteric arteries with Fura‐2AM confirmed higher cytosolic Ca 2+ transients triggered by ACh in arteries from female mice versus male mice. MCU inhibition both strongly reduced cytosolic Ca 2+ transients and blocked mitochondrial Ca 2+ entry. In cultured human aortic endothelial cells, treatment with physiological concentrations of estradiol enhanced cytosolic Ca 2+ transients, Ca 2+ buffering capacity, and mitochondrial Ca 2+ entry in response to ATP or repeat Ca 2+ boluses. Further experiments to establish the mechanisms underlying these effects did not reveal significant differences in the expression of MCU subunits, at either the mRNA or protein level. However, estradiol treatment was associated with an increase in mitochondrial mass, mitochondrial fusion, and the mitochondrial membrane potential and reduced mitochondrial superoxide production. Conclusions: Our data confirm that mitochondrial function in endothelial cells differs by sex, with female mice having enhanced Ca 2+ uptake capacity, and that these differences are attributable to the presence of more mitochondria and a higher mitochondrial membrane potential in female mice rather than differences in composition of the MCU complex.
- Open Access
- 16 May 2022
- Published by Springer Nature in Cureus
- Vol. 14 (5) , e25054
Benchmarking Surgeons’ Gender and Year of Medical School Graduation Associated With Monthly Operative Workdays for Multispecialty Groups
Female surgeons reportedly receive less surgical block time and fewer procedural referrals than male surgeons. In this study, we compared operative days between female and male surgeons throughout Florida. Our objective was to facilitate benchmarking by multispecialty groups, both the endpoint to use for statistically reliable results and expected differences. The historical cohort study included all 4,060,070 ambulatory procedural encounters and inpatient elective surgical states performed between January 2017 and December 2019 by 8,472 surgeons at 609 facilities. Surgeons’ gender, year of medical school graduation, and surgical specialty were obtained from their National Provider Identifiers. Female surgeons operated an average of 1.0 fewer days per month than matched male surgeons (99% confidence interval 0.8 to 1.2 fewer days, P < 0.0001). The mean differences were 0.8 to 1.4 fewer days per month among each of the five quintiles of years of graduation from medical school (all P ≤ 0.0050). Results were comparable when repeated using the number of monthly cases the surgeons performed. An average difference of ≤1.4 days per month is a conservative estimate for the current status quo of the workload difference in Florida. Suppose that a group’s female surgeons average more than two fewer operative days per month than the group’s male surgeons of the same specialty. Such a large average difference would call for investigation of what might reflect systematic bias. While such a difference may reflect good flexibility of the organization, it may show a lack of responsiveness (e.g., fewer referrals of procedural patients to female surgeons or bias when apportioning allocated operating room time).
- 16 May 2022
- Published by Springer Nature in Cureus
- Vol. 14 (5) , e25054
Open AccessBenchmarking Surgeons’ Gender and Year of Medical School Graduation Associated With Monthly Operative Workdays for Multispecialty Groups
Female surgeons reportedly receive less surgical block time and fewer procedural referrals than male surgeons. In this study, we compared operative days between female and male surgeons throughout Florida. Our objective was to facilitate benchmarking by multispecialty groups, both the endpoint to use for statistically reliable results and expected differences. The historical cohort study included all 4,060,070 ambulatory procedural encounters and inpatient elective surgical states performed between January 2017 and December 2019 by 8,472 surgeons at 609 facilities. Surgeons’ gender, year of medical school graduation, and surgical specialty were obtained from their National Provider Identifiers. Female surgeons operated an average of 1.0 fewer days per month than matched male surgeons (99% confidence interval 0.8 to 1.2 fewer days, P < 0.0001). The mean differences were 0.8 to 1.4 fewer days per month among each of the five quintiles of years of graduation from medical school (all P ≤ 0.0050). Results were comparable when repeated using the number of monthly cases the surgeons performed. An average difference of ≤1.4 days per month is a conservative estimate for the current status quo of the workload difference in Florida. Suppose that a group’s female surgeons average more than two fewer operative days per month than the group’s male surgeons of the same specialty. Such a large average difference would call for investigation of what might reflect systematic bias. While such a difference may reflect good flexibility of the organization, it may show a lack of responsiveness (e.g., fewer referrals of procedural patients to female surgeons or bias when apportioning allocated operating room time).
- 14 February 2021
- Published by Elsevier in Experimental Eye Research
- Vol. 205, 108494
Circumferential trabecular meshwork cell density in the human eye
- 14 February 2021
- Published by Elsevier in Experimental Eye Research
- Vol. 205, 108494
Circumferential trabecular meshwork cell density in the human eye
- 18 September 2020
- Published by Springer Nature in Health Care Management Science
- Vol. 23 (4) , 640-648
Association between leniency of anesthesiologists when evaluating certified registered nurse anesthetists and when evaluating didactic lectures
Daily evaluations of certified registered nurse anesthetists’ (CRNAs’) work habits by anesthesiologists should be adjusted for rater leniency. The current study tested the hypothesis that there is a pairwise association by rater between leniencies of evaluations of CRNAs’ daily work habits and of didactic lectures. The historical cohorts were anesthesiologists’ evaluations over 53 months of CRNAs’ daily work habits and 65 months of didactic lectures by visiting professors and faculty. The binary endpoints were the Likert scale scores for all 6 and 10 items, respectively, equaling the maximums of 5 for all items, or not. Mixed effects logistic regression estimated the odds of each ratee performing above or below average adjusted for rater leniency. Bivariate errors in variables least squares linear regression estimated the association between the leniency of the anesthesiologists’ evaluations of work habits and didactic lectures. There were 29/107 (27%) raters who were more severe in their evaluations of CRNAs’ work habits than other anesthesiologists (two-sided P < 0.01); 34/107 (32%) raters were more lenient. When evaluating lectures, 3/81 (4%) raters were more severe and 8/81 (10%) more lenient. Among the 67 anesthesiologists rating both, leniency (or severity) for work habits was not associated with that for lectures (P = 0.90, unitless slope between logits 0.02, 95% confidence interval −0.34 to 0.30). Rater leniency is of large magnitude when making daily clinical evaluations, even when using a valid and psychometrically reliable instrument. Rater leniency was context dependent, not solely a reflection of raters’ personality or rating style.
- 18 September 2020
- Published by Springer Nature in Health Care Management Science
- Vol. 23 (4) , 640-648
Association between leniency of anesthesiologists when evaluating certified registered nurse anesthetists and when evaluating didactic lectures
Daily evaluations of certified registered nurse anesthetists’ (CRNAs’) work habits by anesthesiologists should be adjusted for rater leniency. The current study tested the hypothesis that there is a pairwise association by rater between leniencies of evaluations of CRNAs’ daily work habits and of didactic lectures. The historical cohorts were anesthesiologists’ evaluations over 53 months of CRNAs’ daily work habits and 65 months of didactic lectures by visiting professors and faculty. The binary endpoints were the Likert scale scores for all 6 and 10 items, respectively, equaling the maximums of 5 for all items, or not. Mixed effects logistic regression estimated the odds of each ratee performing above or below average adjusted for rater leniency. Bivariate errors in variables least squares linear regression estimated the association between the leniency of the anesthesiologists’ evaluations of work habits and didactic lectures. There were 29/107 (27%) raters who were more severe in their evaluations of CRNAs’ work habits than other anesthesiologists (two-sided P < 0.01); 34/107 (32%) raters were more lenient. When evaluating lectures, 3/81 (4%) raters were more severe and 8/81 (10%) more lenient. Among the 67 anesthesiologists rating both, leniency (or severity) for work habits was not associated with that for lectures (P = 0.90, unitless slope between logits 0.02, 95% confidence interval −0.34 to 0.30). Rater leniency is of large magnitude when making daily clinical evaluations, even when using a valid and psychometrically reliable instrument. Rater leniency was context dependent, not solely a reflection of raters’ personality or rating style.
- Open Access
- 3 August 2020
- Published by Association for Research in Vision and Ophthalmology (ARVO) in Investigative Ophthalmology & Visual Science
- Vol. 61 (10) , 32
Focus on Data: Statistical Significance, Effect Size and the Accumulation of Evidence Achieved by Combining Study Results Through Meta-analysis
Purpose: To provide information and perspectives on statistical significance and on meta-analysis, a statistical procedure for combining estimated effects across multiple studies. Methods: Methods are presented for performing a meta-analysis in which results across multiple studies are combined. An example of a meta-analysis of optical coherence tomography thickness of the retina in patients with multiple sclerosis across multiple studies is provided. We show how to combine individual study results and how to weight the results of each study based on its reliability. The method of a meta-analysis is used to derive from all study results a pooled estimate that is closest to the unknown common effect. Results: Differences between the two most common methods for meta-analysis, the fixed-effects approach and the random-effects approach, are reviewed. Meta-analysis is applied to the study of the differences in the thickness of the retinal nerve fiber layers of healthy controls and patients with multiple sclerosis, showing why this is a useful procedure for combining estimated effects across multiple studies to derive the magnitude of retinal thinning caused by multiple sclerosis. Conclusions: This review provides information and perspectives on statistical significance and on meta-analysis, a statistical procedure for combining estimated effects across multiple studies. A discussion is provided to show why statistical significance and low probability values are not all that matter and why investigators should also look at the magnitude of the estimated effects. Combining estimated effects across multiple studies with proper weighting of individual results is the goal of meta-analysis.
- 3 August 2020
- Published by Association for Research in Vision and Ophthalmology (ARVO) in Investigative Ophthalmology & Visual Science
- Vol. 61 (10) , 32
Open AccessFocus on Data: Statistical Significance, Effect Size and the Accumulation of Evidence Achieved by Combining Study Results Through Meta-analysis
Purpose: To provide information and perspectives on statistical significance and on meta-analysis, a statistical procedure for combining estimated effects across multiple studies. Methods: Methods are presented for performing a meta-analysis in which results across multiple studies are combined. An example of a meta-analysis of optical coherence tomography thickness of the retina in patients with multiple sclerosis across multiple studies is provided. We show how to combine individual study results and how to weight the results of each study based on its reliability. The method of a meta-analysis is used to derive from all study results a pooled estimate that is closest to the unknown common effect. Results: Differences between the two most common methods for meta-analysis, the fixed-effects approach and the random-effects approach, are reviewed. Meta-analysis is applied to the study of the differences in the thickness of the retinal nerve fiber layers of healthy controls and patients with multiple sclerosis, showing why this is a useful procedure for combining estimated effects across multiple studies to derive the magnitude of retinal thinning caused by multiple sclerosis. Conclusions: This review provides information and perspectives on statistical significance and on meta-analysis, a statistical procedure for combining estimated effects across multiple studies. A discussion is provided to show why statistical significance and low probability values are not all that matter and why investigators should also look at the magnitude of the estimated effects. Combining estimated effects across multiple studies with proper weighting of individual results is the goal of meta-analysis.
- Open Access
- 1 July 2020
- Published by Association for Research in Vision and Ophthalmology (ARVO) in Investigative Ophthalmology & Visual Science
- Vol. 61 (8) , 25
Parametric Statistical Inference for Comparing Means and Variances
Purpose: The purpose of this tutorial is to provide visual scientists with various approaches for comparing two or more groups of data using parametric statistical tests, which require that the distribution of data within each group is normal (Gaussian). Non-parametric tests are used for inference when the sample data are not normally distributed or the sample is too small to assess its true distribution. Methods: Methods are reviewed using retinal thickness, as measured by optical coherence tomography (OCT), as an example for comparing two or more group means. The following parametric statistical approaches are presented for different situations: two-sample t-test, Analysis of Variance (ANOVA), paired t-test, and the analysis of repeated measures data using a linear mixed-effects model approach. Results: Analyzing differences between means using various approaches is demonstrated, and follow-up procedures to analyze pairwise differences between means when there are more than two comparison groups are discussed. The assumption of equal variance between groups and methods to test for equal variances are examined. Examples of repeated measures analysis for right and left eyes on subjects, across spatial segments within the same eye (e.g. quadrants of each retina), and over time are given. Conclusions: This tutorial outlines parametric inference tests for comparing means of two or more groups and discusses how to interpret the output from statistical software packages. Critical assumptions made by the tests and ways of checking these assumptions are discussed. Efficient study designs increase the likelihood of detecting differences between groups if such differences exist. Situations commonly encountered by vision scientists involve repeated measures from the same subject over time, measurements on both right and left eyes from the same subject, and measurements from different locations within the same eye. Repeated measurements are usually correlated, and the statistical analysis needs to account for the correlation. Doing this the right way helps to ensure rigor so that the results can be repeated and validated.
- 1 July 2020
- Published by Association for Research in Vision and Ophthalmology (ARVO) in Investigative Ophthalmology & Visual Science
- Vol. 61 (8) , 25
Open AccessParametric Statistical Inference for Comparing Means and Variances
Purpose: The purpose of this tutorial is to provide visual scientists with various approaches for comparing two or more groups of data using parametric statistical tests, which require that the distribution of data within each group is normal (Gaussian). Non-parametric tests are used for inference when the sample data are not normally distributed or the sample is too small to assess its true distribution. Methods: Methods are reviewed using retinal thickness, as measured by optical coherence tomography (OCT), as an example for comparing two or more group means. The following parametric statistical approaches are presented for different situations: two-sample t-test, Analysis of Variance (ANOVA), paired t-test, and the analysis of repeated measures data using a linear mixed-effects model approach. Results: Analyzing differences between means using various approaches is demonstrated, and follow-up procedures to analyze pairwise differences between means when there are more than two comparison groups are discussed. The assumption of equal variance between groups and methods to test for equal variances are examined. Examples of repeated measures analysis for right and left eyes on subjects, across spatial segments within the same eye (e.g. quadrants of each retina), and over time are given. Conclusions: This tutorial outlines parametric inference tests for comparing means of two or more groups and discusses how to interpret the output from statistical software packages. Critical assumptions made by the tests and ways of checking these assumptions are discussed. Efficient study designs increase the likelihood of detecting differences between groups if such differences exist. Situations commonly encountered by vision scientists involve repeated measures from the same subject over time, measurements on both right and left eyes from the same subject, and measurements from different locations within the same eye. Repeated measurements are usually correlated, and the statistical analysis needs to account for the correlation. Doing this the right way helps to ensure rigor so that the results can be repeated and validated.