Power of Modified Brown-Forsythe and Mixed-Model Approaches in Split-Plot Designs

Methodology ◽  
2017 ◽  
Vol 13 (1) ◽  
pp. 9-22 ◽  
Author(s):  
Pablo Livacic-Rojas ◽  
Guillermo Vallejo ◽  
Paula Fernández ◽  
Ellián Tuero-Herrero
Keyword(s):  
Repeated Measures ◽  
Statistical Power ◽  
Mixed Model ◽  
Covariance Structure ◽  
Simulation Method ◽  
Future Research ◽  
Repeated Measures Design ◽  
Fixed And Random Effects ◽  
Split Plot ◽  
High Level

Abstract. Low precision of the inferences of data analyzed with univariate or multivariate models of the Analysis of Variance (ANOVA) in repeated-measures design is associated to the absence of normality distribution of data, nonspherical covariance structures and free variation of the variance and covariance, the lack of knowledge of the error structure underlying the data, and the wrong choice of covariance structure from different selectors. In this study, levels of statistical power presented the Modified Brown Forsythe (MBF) and two procedures with the Mixed-Model Approaches (the Akaike’s Criterion, the Correctly Identified Model [CIM]) are compared. The data were analyzed using Monte Carlo simulation method with the statistical package SAS 9.2, a split-plot design, and considering six manipulated variables. The results show that the procedures exhibit high statistical power levels for within and interactional effects, and moderate and low levels for the between-groups effects under the different conditions analyzed. For the latter, only the Modified Brown Forsythe shows high level of power mainly for groups with 30 cases and Unstructured (UN) and Autoregressive Heterogeneity (ARH) matrices. For this reason, we recommend using this procedure since it exhibits higher levels of power for all effects and does not require a matrix type that underlies the structure of the data. Future research needs to be done in order to compare the power with corrected selectors using single-level and multilevel designs for fixed and random effects.

The use of MIXED models in the analysis of animal experiments with repeated measures data

2004 ◽  
Vol 84 (1) ◽  
pp. 1-11 ◽  
Author(s):  
Z. Wang and L. A. Goonewardene
Keyword(s):  
Least Squares ◽  
Linear Model ◽  
Repeated Measures ◽  
Mixed Model ◽  
Covariance Structure ◽  
General Linear Model ◽  
Ordinary Least Squares ◽  
General Linear ◽  
Repeated Measures Data ◽  
Split Plot

The analysis of data containing repeated observations measured on animals (experimental unit) allocated to different treatments over time is a common design in animal science. Conventionally, repeated measures data were either analyzed as a univariate (split-plot in time) or a multivariate ANOVA (analysis of contrasts), both being handled by the General Linear Model procedure of SAS. In recent times, the mixed model has become more appealing for analyzing repeated data. The objective of this paper is to provide a background understanding of mixed model methodology in a repeated measures analysis and to use balanced steer data from a growth study to illustrate the use of PROC MIXED in the SAS system using five covariance structures. The split-plot in time approach assumes a constant variance and equal correlations (covariance) between repeated measures or compound symmetry, regardless of their proximity in time, and often these assumptions are not true. Recognizing this limitation, the analysis of contrasts was proposed. If there are missing measurements, or some of the data are measured at different times, such data were excluded resulting in inadequate data for a meaningful analysis. The mixed model uses the generalized least squares method, which is generally better than the ordinary least squares used by GLM, if the appropriate covariance structure is adopted. The presence of unequally spaced and/or missing data does not pose a problem for the mixed model. In the example analyzed, the first order ante dependence [ANTE(1)] covariance model had the lowest value for the Akaike and Schwarz’s Bayesian information criteria fit statistics and is therefore the model that provided the best fit to our data. Hence, F values, least square estimates and standard errors based on the ANTE (1) were considered the most appropriate from among the five models demonstrated. It is recommended that the mixed model be used for the analysis of repeated measures designs in animal studies. Key words: Repeated measures, General Linear Model, Mixed Model, split-plot, covariance structure

scholarly journals Investigating Care Dependency and Its Relation to Outcome (ICARE): Results From a Naturalistic Study of an Intensive Day Treatment Program for Depression

2021 ◽  
Vol 12 ◽  
Author(s):  
Sarah Glanert ◽  
Svenja Sürig ◽  
Ulrike Grave ◽  
Eva Fassbinder ◽  
Sebastian Schwab ◽  
...  
Keyword(s):  
Treatment Outcome ◽  
Depressive Symptoms ◽  
Repeated Measures ◽  
Mixed Model ◽  
Depressive Symptomatology ◽  
Future Research ◽  
Clinical Trial Registration ◽  
Levels Of Care ◽  
Repeated Measures Design ◽  
Care Dependency

Background: This study explores the association of experienced dependency in psychotherapy as measured with the CDQ (Care Dependency Questionnaire) and treatment outcome in depression. Furthermore, the course of care dependency and differences in the CDQ scores depending on the received type of treatment, MCT (metacognitive therapy), or CBASP (cognitive behavioral analysis system of psychotherapy), were investigated.Methods: The study follows a prospective, parallel group observational design. Patients suffering from depression received an 8-week intensive day clinic program, which was either CBASP or MCT. The treatment decision was made by clinicians based on the presented symptomatology and with regard to the patients' preferences. The patients reported depressive symptoms with the QIDS-SR16 (Quick Inventory of Depressive Symptomatology) and levels of experienced care dependency with the German version of the CDQ on a weekly basis. Mixed-model analyses were run to account for the repeated-measures design.Results: One hundred patients were included in the analyses. Results indicate that higher levels of care dependency might predict a less favorable outcome of depressive symptomatology. Levels of care dependency as well as depressive symptoms decreased significantly over the course of treatment. There was no significant between-group difference in care dependency between the two treatment groups.Conclusion: The results suggest that care dependency might be associated with a worse treatment outcome in depressed patients. In general, care dependency seems to be a dynamic construct, as it is changing over time, while the levels of care dependency seem to be independent from the received type of treatment. Future research should continue investigating the mechanisms of care dependency in a randomized controlled design.Clinical Trial Registration:https://www.drks.de/drks_web/, identifier: DRKS00023779.

scholarly journals Carbohydrate-Induced Insulin Signaling Activates Focal Adhesion Kinase: A Nutrient and Mechanotransduction Crossroads

Nutrients ◽  
2020 ◽  
Vol 12 (10) ◽  
pp. 3145 ◽  
Author(s):  
Dylan T. Wilburn ◽  
Steven B. Machek ◽  
Thomas D. Cardaci ◽  
Darryn S. Willoughby
Keyword(s):  
Focal Adhesion Kinase ◽  
Focal Adhesion ◽  
Insulin Signaling ◽  
Repeated Measures ◽  
Vastus Lateralis ◽  
Future Research ◽  
Placebo Condition ◽  
Carbohydrate Ingestion ◽  
Post Exercise ◽  
Repeated Measures Design

Research has suggested that nutrient, exercise, and metabolism-related proteins interact to regulate mammalian target of rapamycin complex one (mTOR) post-exercise and their interactions needs clarification. In a double-blind, cross-over, repeated measures design, ten participants completed four sets to failure at 70% of 1-repitition maximum (1-RM) with 45 s rest on angled leg press with or without pre-exercise maltodextrin (2 g/kg) after a 3 h fast. Vastus lateralis biopsies were collected at baseline before supplementation and 1 h post-exercise to analyze Focal Adhesion Kinase (FAK), ribosomal protein S6 kinase beta-1 (p70S6K), insulin receptor substrate 1 (IRS-1), phosphatidylinositol 3-kinase (PI3K), and 5′ AMP-activated protein kinase (AMPK) activation. FAK and IRS-1 activity were only elevated 1 h post-exercise with carbohydrate ingestion (p < 0.05). PI3K and p70S6K activation were both elevated after exercise in both conditions (p < 0.05). However, AMPK activity did not change from baseline in both conditions (p > 0.05). We conclude that FAK does not induce mTOR activation through PI3K crosstalk in response to exercise alone. In addition, FAK may not be regulated by AMPK catalytic activity, but this needs further research. Interestingly, carbohydrate-induced insulin signaling appears to activate FAK at the level of IRS-1 but did not enhance mTOR activity 1 h post-exercise greater than the placebo condition. Future research should investigate these interactions under different conditions and within different time frames to clearly understand the interactions between these signaling molecules.

scholarly journals Investigating human skeletal muscle physiology with unilateral exercise models: when one limb is more powerful than two

2017 ◽  
Vol 42 (6) ◽  
pp. 563-570 ◽  
Author(s):  
Martin J. MacInnis ◽  
Chris McGlory ◽  
Martin J. Gibala ◽  
Stuart M. Phillips
Keyword(s):  
Skeletal Muscle ◽  
Repeated Measures ◽  
Statistical Power ◽  
Human Skeletal Muscle ◽  
Capillary Density ◽  
Muscle Physiology ◽  
Repeated Measures Design ◽  
Confounding Variables ◽  
Direct Sampling ◽  
Wide Range

Direct sampling of human skeletal muscle using the needle biopsy technique can facilitate insight into the biochemical and histological responses resulting from changes in exercise or feeding. However, the muscle biopsy procedure is invasive, and analyses are often expensive, which places pragmatic restraints on sample sizes. The unilateral exercise model can serve to increase statistical power and reduce the time and cost of a study. With this approach, 2 limbs of a participant are randomized to 1 of 2 treatments that can be applied almost concurrently or sequentially depending on the nature of the intervention. Similar to a typical repeated measures design, comparisons are made within participants, which increases statistical power by reducing the amount of between-person variability. A washout period is often unnecessary, reducing the time needed to complete the experiment and the influence of potential confounding variables such as habitual diet, activity, and sleep. Variations of the unilateral exercise model have been employed to investigate the influence of exercise, diet, and the interaction between the 2, on a wide range of variables including mitochondrial content, capillary density, and skeletal muscle hypertrophy. Like any model, unilateral exercise has some limitations: it cannot be used to study variables that potentially transfer across limbs, and it is generally limited to exercises that can be performed in pairs of treatments. Where appropriate, however, the unilateral exercise model can yield robust, well-controlled investigations of skeletal muscle responses to a wide range of interventions and conditions including exercise, dietary manipulation, and disuse or immobilization.

scholarly journals Interpersonal touch enhances cognitive control: A neurophysiological investigation

2018 ◽  
Author(s):  
Blair Saunders ◽  
Anja Riesel ◽  
Julia Klawohn ◽  
Michael Inzlicht
Keyword(s):  
Cognitive Control ◽  
Inhibitory Control ◽  
Repeated Measures ◽  
Statistical Power ◽  
Performance Monitoring ◽  
Control Task ◽  
Romantic Couples ◽  
Repeated Measures Design ◽  
Interpersonal Touch ◽  
Neural Monitoring

Touch is central to mammalian communication, socialisation, and wellbeing. Despite this prominence, interpersonal touch is relatively understudied. In this preregistered investigation, we assessed the influence of interpersonal touch on the subjective, neural, and behavioural correlates of cognitive control. Forty-five romantic couples were recruited (N=90; dating&gt;6 months), and one partner performed an inhibitory control task while electroencephalography was recorded to assess neural performance monitoring. Interpersonal touch was provided by the second partner, and was manipulated between experimental blocks. A within-subject repeated-measures design was used to maximise statistical power, with our sample size providing 80% power for even small effect sizes (ds &gt; .25). Results indicated that participants were not only happier when receiving touch, but also showed increased neural processing of mistakes. Further exploratory cognitive modelling using indirect effects tests and drift diffusion models of decision making revealed that touch was indirectly associated with both improved inhibitory control and increased rates of evidence accumulation (drift rate) through its influence on neural monitoring. Thus, beyond regulating emotion and stress, interpersonal touch appears to enhance the neurocognitive processes underling flexible goal-directed behaviour.

The cognitive and behavioral impact of a culinary education program on schoolchildren

2020 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Nur Al Ali ◽  
Andres Arriaga ◽  
Margarita Rubio
Keyword(s):  
Diet Quality ◽  
Education Program ◽  
Repeated Measures ◽  
Statistical Power ◽  
Statistical Significance ◽  
Control Group ◽  
Content Type ◽  
Repeated Measures Design ◽  
The Mean ◽  
Culinary Education

Purpose The purpose of this paper is to design a culinary education program and ascertain its impact on knowledge, phobias, culinary skills and diet quality in schoolchildren. Design/methodology/approach Repeated measures design to determine changes after the implementation of the culinary education program in a pre-school and primary school located in a Madrid (Spain) neighborhood with a low socioeconomic level. A total of 58 children agreed to participate in the program and 40 parents authorized the evaluation of the effect of the program. Variables were measured with Student’s t-test or Wilcoxon-signed range test for quantitative variables. Effect size was calculated by Cohen’s d. Findings The mean scores in knowledge and beliefs improved from 5.0 to 7.8 (P < 0.001; Cohen’s d = 1), the mean value obtained in the food phobias test decreased from 24.6 to 20.7 (P = 0.01; Cohen’s d = 0.53), diet quality on Kid Med Test score increased from 7.4 to 8.2 (P = 0.06; Cohen’s d = 0.38), and the children improved their culinary skills from 21.2 to 27.9 (P < 0.001; Cohen’s d = 0.66). Research limitations/implications Self-declared data could lead to information biases. Other limitations were the lack of control group and a scarce statistical power that could explain the absence of statistical significance in the results on diet quality. To observe the effects of this change on eating habits, long-term evaluations would have to be carried out. Originality/value The brand-new culinary education program had a positive effect on children’s knowledge of nutrition and culinary skills as well as reducing food phobias. This proposal is beyond the state-of-the-art and could be implemented elsewhere with a robust effect on children, parents and educators.

scholarly journals Investigating Parietal and Premotor Influence on Motor Cortical Excitability Associated with Visuomotor Associative Plasticity

2021 ◽  
Vol 11 (4) ◽  
pp. 452
Author(s):  
Paul J. Wolfe ◽  
Lynea B. Kaethler ◽  
W. Richard Staines
Keyword(s):  
Repeated Measures ◽  
Cortical Excitability ◽  
Occipital Cortex ◽  
Long Term Potentiation ◽  
Future Research ◽  
Reaching Movement ◽  
Repeated Measures Design ◽  
Term Potentiation ◽  
Motor Cortex Excitability

The brain changes in response to sensory signals it is exposed to. It has been shown that long term potentiation-like neuroplasticity can be experimentally induced via visual paired-associative stimulation (V-PAS). V-PAS combines afferent visual stimuli with a transcranial magnetic stimulation pulse to induce plasticity. Preparation of a reaching movement to generate activity in superior parietal occipital cortex (SPOC) was used in this study as an additional afferent contributor to modulate the resultant plasticity. We hypothesized that V-PAS with a reaching movement would induce greater cortical excitability than V-PAS alone and would exhibit facilitated SPOC to M1 projections. All four experiments enrolled groups of 10 participants to complete variations of V-PAS in a repeated measures design. SPOC to M1 projections facilitated motor cortex excitability following V-PAS regardless of intervention received. We did not observe evidence indicating extra afferent information provided an additive effect to participants. Investigation of PMd to M1 projections confirmed disinhibition and suggested interneuronal populations within M1 may be mechanistically involved. Future research should look to rule out the existence of an upper limit for effective afference during V-PAS and investigate the average influence of V-PAS on cortical excitability in the larger population.

scholarly journals Self-Confidence and Perceptions of Athletic Training Students Following Simulated Experiences: A Mixed-Methods Pilot Study

2021 ◽  
Author(s):  
Hannah Stedge ◽  
Valerie Herzog
Keyword(s):  
Mixed Methods ◽  
Athletic Training ◽  
Clinical Education ◽  
Repeated Measures ◽  
Real Life ◽  
Future Research ◽  
High Fidelity ◽  
Athletic Training Students ◽  
Repeated Measures Design ◽  
Self Confidence

Purpose: Athletic training clinical education provides students with a variety of real-life patient encounters under the supervision of a credentialed athletic trainer. However, clinical education experiences may not allow for all students to practice more invasive or less common skills such as rectal thermometry. The purpose of this study was two-fold: 1) determine the effect of a series of low to high-fidelity simulated rectal thermometry experiences on athletic training students’ self-confidence scores; and 2) through a phenomenological qualitative approach, explore athletic training students’ perceptions following the simulated rectal thermometry experiences. Method: This study used a quasi-experimental, mixed methods, one-group repeated measures design. Participants were nine first year professional Master of Athletic Training students who completed four simulated experiences on the skill of rectal thermometry. The outcome measures were The Athletic Trainer’s Self-Confidence Scale (ATSCS) and qualitative semi-structured open-ended interviews. Results: Friedman’s ANOVA revealed a statistically significant improvement from the participant’s baseline ATSCS scores to the last post-test scores. We identified three major themes following qualitative interviews: 1) Perception changes before and after participation; 2) Benefits of a series of four simulations; 3) Explanations for the improvements in confidence and competence. Conclusions: A series of simulated learning activities using high-fidelity and standardized patient interactions can help improve athletic training students’ self-confidence. Though future research is needed, athletic training education programs should consider implementing simulated experiences to equip athletic training students to transition to practice.

scholarly journals Effects of Temperature and Wetness Duration on the Sporulation Rate of Phomopsis viticola on Infected Grape Canes

Plant Disease ◽  
2013 ◽  
Vol 97 (5) ◽  
pp. 579-589 ◽  
Author(s):  
D. J. Anco ◽  
L. V. Madden ◽  
M. A. Ellis
Keyword(s):  
Repeated Measures ◽  
Goodness Of Fit ◽  
Environmental Data ◽  
Spore Density ◽  
Controlled Environment ◽  
Phomopsis Viticola ◽  
Repeated Measures Design ◽  
Split Plot ◽  
Effects Of Temperature ◽  
Sporulation Rate

Controlled-environment studies were conducted to examine effects of temperature (T) and wetness duration (W) on the sporulation rate of Phomopsis viticola on infected grape canes and to determine effects of interrupted wetness duration (IWD) on sporulation. A split-plot design was used to determine T and W effects, with T (5, 12, 15, 18, 20, 22, 25, 28, and 35°C) as the whole-plot and W (11, 23, 35, 47, and 71 h) as the subplot. Linear and nonlinear mixed models were fitted to the data. Lower and upper limits of sporulation were estimated to be 4 and 36°C, respectively, based on the modeling results, optimum sporulation was near 21°C, and sporulation increased monotonically with increasing wetness duration. Of the examined models, a generalization of the Analytis Beta model fit the data best, based on a collection of goodness-of-fit statistical criteria. To determine effects of IWD, a split-plot was used, with T (12, 15, and 20°C) as the whole-plot and IWD (0, 2, 4, 8, 12, and 24 h) as the subplot. Generally, sporulation declined with increasing IWD. An IWD of 8 h or more resulted in significantly and substantially less sporulation compared to the control (0 h IWD) (P < 0.01). Temporal patterns of spore density in the field were determined using a repeated-measures design, in which spore density and environmental data were measured in the vineyard during and following individual rain events over 3 years. The developed model from the controlled-environment study, coupled with a time-of-season weight function and a dispersal index (based on total rain per rain episode), predicted the trend in spore density over time reasonably well, although the total magnitude of spore density could not be predicted because the density of lesions was not known. Results can be used for improving the accuracy of a disease warning system that currently only considers infection of grapes by P. viticola.

Sign in / Sign up

Export Citation Format

Share Document