Exact-and approximate bayes credibility intervals for the one-way analysis of variance model

1990 ◽  
Vol 19 (7) ◽  
pp. 2541-2565
Author(s):  
A.J. Van Der Merwe ◽  
C.A. Van Der Merwe
Keyword(s):  
Analysis Of Variance ◽  
Variance Model ◽  
The One

scholarly journals Generalization of the order-restricted information criterion for multivariate normal linear models (pre-print )

2021 ◽  
Author(s):  
Rebecca M. Kuiper ◽  
Herbert Hoijtink
Keyword(s):  
Model Selection ◽  
Analysis Of Variance ◽  
Linear Models ◽  
Inequality Constraints ◽  
Information Criterion ◽  
Variance Model ◽  
Order Restrictions ◽  
Population Means ◽  
Order Restricted ◽  
The One

The Akaike information criterion for model selection presupposes that the parameter space is not subject to order restrictions or inequality constraints.Anraku (1999) proposed a modified version of this criterion, called the order-restricted information criterion, for model selection in the one-way analysis of variance model when the population means are monotonic.We propose a generalization of this to the case when the population means may be restricted by a mixture of linear equality and inequality constraints.If the model has no inequality constraints, then the generalized order-restricted information criterion coincides with the Akaike information criterion.Thus, the former extends the applicability of the latter to model selection in multi-way analysis of variance models when some models may have inequality constraints while others may not. Simulation shows that the information criterion proposed in this paper performs well in selecting the correct model.

Model Comparison in the One-Way Between-Subjects Design: An Sas Implementation

1992 ◽  
Vol 75 (3_suppl) ◽  
pp. 1124-1126
Author(s):  
John F. Walsh
Keyword(s):  
Analysis Of Variance ◽  
Model Comparison ◽  
Statistical Test ◽  
Statistical System ◽  
Test Statistic ◽  
Cell Means ◽  
Treatment Conditions ◽  
Competing Models ◽  
Proportional Increase ◽  
The One

A statistical test is developed based on the comparison of sums of squared errors associated with two competing models. A model based on cell means is compared to a representation that specifies the means for the treatment conditions. Comparing models is more general than the traditional H0 in analysis of variance wherein all the cell means are assumed equal. The test statistic, Proportional Increase in Error, is computed using the SAS statistical system.

scholarly journals A further consideration in the application of an analysis-of-variance model for the intrasubject replication design

1974 ◽  
Vol 7 (4) ◽  
pp. 629-633 ◽  
Author(s):  
T. Kratochwill ◽  
K. Alden ◽  
D. Demuth ◽  
D. Dawson ◽  
C. Panicucci ◽  
...  
Keyword(s):  
Analysis Of Variance ◽  
Variance Model

scholarly journals Belief in importance of information literacy abilities among undergraduates. Underlying factors and analysis of variance

2020 ◽  
Vol 48 (4) ◽  
pp. 559-577
Author(s):  
Maria Pinto ◽  
David Caballero ◽  
Dora Sales ◽  
Alicia Segura
Keyword(s):  
Analysis Of Variance ◽  
Cognitive Processing ◽  
Information Literacy ◽  
Self Assessment ◽  
Academic Degree ◽  
Content Type ◽  
Gender And Age ◽  
Technological Processing ◽  
And Gender ◽  
The One

Purpose The purpose of this paper is to analyse the levels of belief in importance of information literacy abilities (BILAs) among an undergraduates’ sample. The aim is, on the one hand, to discover if there is a representative latent structure and, on the other hand, to know the existing differences according to external variables such as academic degree, course, gender and age. Design/methodology/approach A self-assessment questionnaire (IL-HUMASS) was applied to a sample of 749 students in English Studies, Translation and Interpreting and Education in Spain. Three types of statistical methods have been used to study the results: descriptive, factorial and analysis of variance. Findings Students’ levels of BILAs are acceptable but improvable. A framework of six underlying factors has been uncovered: evaluation-ethics, searching-using, technological processing, communication, dissemination and cognitive processing of the information. Significant differences on degree, course and gender have been found. Practical implications This paper is intended for a broad academic sector, including faculty, librarians and students in higher education. The BILAs construct helps to improve the diagnosis of the perception of the BILAs. Its representation through a reduced number of latent factors simplifies results and possible applications. The results show that variations in degree, course and gender are significant and should be taken into account. Originality/value Although much has been written about information literacy abilities, we still know little about the importance students place on them. The BILAs construct is intended to improve that knowledge.

scholarly journals Effect of Immediate Dentin Sealing and Surface Conditioning on the Microtensile Bond Strength of Resin-based Composite to Dentin

2019 ◽  
Vol 44 (6) ◽  
pp. E289-E298 ◽  
Author(s):  
CRG van den Breemer ◽  
M Özcan ◽  
MS Cune ◽  
AP Almeida Ayres ◽  
B Van Meerbeek ◽  
...  
Keyword(s):  
Bond Strength ◽  
Analysis Of Variance ◽  
Water Storage ◽  
Silica Coating ◽  
Microtensile Bond Strength ◽  
Surface Conditioning ◽  
Clinical Procedures ◽  
Significant Difference ◽  
The One ◽  
Storage Times

SUMMARY This study evaluated the microtensile bond strength (μTBS) of resin-based composite (RBC) to dentin after different immediate dentin sealing (IDS) strategies and surface-conditioning (SC) methods and on two water storage times. Human molars (n=48) were randomly divided into eight experimental groups involving four different IDS strategies—IDS-1L with one layer of adhesive, IDS-2L with two layers of adhesive, IDS-F with one layer of adhesive and one layer of flowable RBC, and DDS (delayed dentin sealing) with no layer of adhesive (control)—and two different SC methods—SC-P with pumice rubbing and SC-PC with pumice rubbing followed by tribochemical silica coating. The μTBS test was performed after one week and after six months of water storage, being recorded as the “immediate” and “aged” μTBS, respectively. Composite-adhesive-dentin microspecimens (0.9×0.9×8–9 mm) were stressed in tension until failure to determine the μTBS. Failure mode and location of failure were categorized. Two-way analysis of variance was applied to analyze the data for statistically significant differences between the experimental groups (p<0.05). Two-way analysis of variance revealed no significant differences between the one-week μTBS specimens for IDS strategy (p=0.087) and SC methods (p=0.806). However, the interaction of IDS strategy and SC methods appeared statistically significant (p=0.016). The six-month specimen evaluation showed no significant difference in μTBS for SC (p=0.297) and SC/IDS interaction (p=0.055), but the μTBS of the IDS strategies differed significantly among them (p=0.003). For tribochemical silica-coated IDS, no significant effect of aging on μTBS was recorded (p=0.465), but there was a highly significant difference in μTBS depending on the IDS strategy (p<0.001). In addition, the interaction of IDS and aging was borderline statistically significant (p=0.045). The specimens failed mainly at the adhesive-dentin interface for all experimental groups. Dentin exposure during clinical procedures for indirect restorations benefits from the application of IDS, which was shown to result in higher bond strength. No significant differences were found between cleaning with solely pumice or pumice followed by tribochemical silica coating.

An Analysis of Variance Model for Assessing Reliability of Naturalistic Observations

1978 ◽  
Vol 47 (1) ◽  
pp. 271-278 ◽  
Author(s):  
Ronald A. Berk
Keyword(s):  
Analysis Of Variance ◽  
Interrater Reliability ◽  
Factorial Analysis ◽  
Future Research ◽  
Variance Model ◽  
Unbiased Estimates ◽  
Variance Approach

This paper presents a subjects × raters partially nested factorial analysis of variance model for estimating coefficients of interrater reliability. Procedures and formulae are described for computing unbiased estimates using the between-subjects and error mean squares from the model. Inclusion or exclusion of rater variance in the estimates is also discussed. Nine specific advantages for using the analysis of variance approach over existing methods are listed. A direction for future research on reliability is suggested.

Sign in / Sign up

Export Citation Format

Share Document