Analysis Of Variance Design Research Articles

Experimental design and statistical decisions tutorial: Comments on longitudinal ideomotor apraxia recovery

Basso, Burgio, Paulin, and Prandoni (2000) claimed evidence for recovery from ideomotor apraxia within the first year post-cerebrovascular accident. However, concerns about the excessive use of paired t tests and incongruous experimental designs challenge the reported findings. The ideomotor apraxia recovery data should be analysed with a completely within-subjects design one-way analysis of variance (ANOVA). Further, the locus of lesion by clinical examination time data should be analysed in a two-way mixed design ANOVA with repeated measures on the three levels of examination time. In addition, circularity assumptions (i.e., composition of the covariance matrices) should be addressed for the repeated measures in the recommended analyses. In a tutorial approach, using the proper experimental designs and statistics, the ideomotor apraxia data were reanalysed. The one-way results support the recovery improvements between the first and second examinations. However, conflict with the original findings was found in the two-way mixed design results. Thus, the overall conclusions should be amended, especially given that the Locus of Lesion - Clinical Examination Time interaction failed to reach significance.

Effect of elastic lumbosacral supports on lumbar spine mechanical stresses

Subjects lifted a 6.82 kg weight from the floor to 1.23 meters high using an elastic lumbosacral support worn the conventional way, without a support, and with the support wide side forward. The lifts were analyzed using the "Lift-track" 2 dimensional analysis program, and the results subjected to repeated measures design analysis of variance (ANOVA) and post hoc Bonferroni analysis. This study indicates that use an elastic lumbosacral support is able to influence lifting posture enough to have a significant effect on mechanical stress being placed on the L5-S1 joint in lifting from the floor, if worn with the wide side facing the abdomen. Wearing the support in the conventional manner had no significant effect in reducing the mechanical stress on the lumbar spine.

USE of outdoor freshwater pond microcosms: I. Microcosm design and fate of pyridaben

AbstractAn outdoor freshwater microcosm study was conducted in which pyridaben, an insecticide‐miticide, was directly applied to water to determine its fate and biological effects on an aquatic community. The following paper describes the design, specific techniques, and fate of pyridaben in microcosms including model prediction of the estimated environmental concentration, whereas the subsequent paper describes the responses of biota. An analysis of variance design was used with four treatments in which pyridaben was applied as an aqueous suspension (75% wettable powder) at three concentrations (0.34, 3.4, 34.0 μg/L) plus an untreated control to 24 microcosm tanks (23 m3). Each treatment was replicated six times. Pyridaben was applied to each microcosm once in April and once in May 1993 to simulate actual crop application. The Exposure Analysis Modeling System predicted that as a result of a drift (5.0%) exposure scenario, the half‐life of pyridaben in water was 30 to 34 h. The model also predicted negligible concentrations of pyridaben in sediment after drift. Surface runoff was not considered an important source of pyridaben to aquatic systems, because the Pesticide Root Zone Model predicted a maximum runoff concentration of 0.1 μg/L. The half‐life in microcosm water for the low, middle, and high treatment concentrations for both applications ranged from 11.8 to 28.5 h. After both applications at 3.4 μg/L, pyridaben was not detected in sediment within 24 h, whereas at 34.0 μg/L the half‐life of pyridaben in sediment was 9.8 d.

USE of outdoor freshwater pond microcosms: II. Responses of biota to pyridaben

AbstractThe effects of pyridaben, an insecticide‐miticide on zooplankton, macroinvertebrates, and fish were studied in outdoor freshwater microcosms using an analysis of variance design with three chemical concentrations (0.34, 3.4, 34.0 μg/L) and one untreated control randomized among 24 tanks. Each treatment was replicated six times. Monitoring was conducted during an 11‐month baseline phase followed by a total of three months for treatment and posttreatment phases. Two applications of a wettable powder formulation were sprayed directly below the water surface with a 30‐d interval between treatments. Copepoda adult abundance was significantly reduced at 34.0 μg/L but recovery occurred within 6 weeks after application. Abundance of copepoda nauplii was significantly reduced at 3.4 and 34.0 μg/L, after applications one and two; effects were more severe at 34.0 μg/L and recovery was more rapid at 3.4 μg/L. Abundance of Rotifera was reduced at 34.0 μg/L, after applications one and two, and recovery occurred within 8 weeks for all groups except Polyarthra and Keratella. Of the most abundant Cladocera, abundance of Alona was not significantly affected and abundance of Latonopsis was significantly reduced at 34.0 μg/L, after applications one and two, but recovery occurred within 6 weeks. Abundance of Latonopsis also was significantly reduced at 3.4 μg/L, after applications one and two, but recovery occurred within 2 weeks. A significant decrease occurred in the abundances of Cnidaria, Insecta, and Hydracarina at 34.0 μg/L, only after application one. Pyridaben was toxic to bluegill at 34.0 μg/L, but was not acutely toxic at the laboratory 96‐h LC50 concentration (∼3.4 μg/L).

Language-Based Cognitive Functioning in Parents of Offspring With ADHD Comorbid for Tourette Syndrome or Learning Disabilities

The parents of children diagnosed with attention deficit hyperactivity disorder (ADHD) were examined for characteristics symptomatic of 2 comorbidities (Co) within their offspring with ADHD: Tourette syndrome (TS) and language-based learning disabilities (LD). A 2 ×2 multivariate analysis of variance design was used; the parents were divided according to whether the offspring with ADHD had TS (Co-TS) or not (No Co-TS) and whether offspring had LD (Co-LD) or not (No Co-LD). Parents (86 mothers and 70 fathers) were administered the short form of the Wechsler Adult Intelligence Scale-Revised (WAIS-R; Wechsler, 1981), subtests of the Woodcock-Johnson Cognitive and Achievement Batteries (Woodcock & Johnson, 1989), and word fluency (semantic and letter; Benton & Hamsher, 1989; Wiig & Semel, 1987). The fathers of offspring with Co-TS were more likely to be diagnosed with LD and had lower WAIS-R Vocabulary, Arithmetic, and Picture Arrangement scores. In addition, when the WAIS-R full scale IQ was used as a covariate, these fathers had lower Woodcock-Johnson Oral Vocabulary, Spelling, and Reading scores (as measured by Letter-Word Identification). The mothers of children with Co-TS had lower WAIS-R Vocabulary scores. Parents of the children with No Co-TS showed a higher proportion of error patterns on the word fluency tasks involving repetitions and rule breaks. The findings show that it was the parents of the children with Co-TS, not the parents of the children with Co-LD, who showed language-based learning problems. In separate discriminant analyses for the fathers and mothers, when the aforementioned significant parental measures were used as predictors, 81% of offspring with Co-TS were correctly classified as having a diagnosis of TS, solely on the basis of characteristics in their parents.

Fleet Assessments of a Virtual Reality Training System

In recent years, virtual reality (VR), often called virtual environments (VE), has received an enormous amount of attention among training developers. This is due in part to media hyperbole over applications of VR in the entertainment industry, but even more so, because training developers recognize the potential of VR as a flexible and effective training medium. A prime candidate for examining the effectiveness of VR systems is the training of Officer of the Deck (OOD) ship handling for the surfaced submarine. The Virtual Environment for Submarine OOD Ship-Handling Training (VESUB) project was developed to demonstrate and evaluate the training potential of a VR-based system for OOD training. This article documents the effectiveness of the training provided by the VESUB trainer and the subjective assessments of the VESUB system obtained from 42 trainees and 12 observers during the Training Effectiveness Evaluation of the VESUB technology demonstration.Data were collected on 15 ship-handling variables grouped into 7 skill categories. A repeated measures analysis of variance design, with scenario session (training and testing) as the within-subjects variable, found significant learning (skill improvements on ship-handling tasks) on 11 of the 15 variables.Trainees and observers filled out an opinion questionnaire consisting of forced-choice and open-ended questions. Eighty-seven percent of the trainees and 100% of the observers indicated that the VESUB system provided a positive learning experience. Ninety-two percent of the trainees and 99% of the observers believed that VESUB would increase their confidence in their ship-handling skills. Fleet recommendations on how these technologies should be used and assessments of system components are provided.

Difficulty, Consequence, and Effort in Academic Task Performance

This study extends previous research on the expectancy construct of the expectancy-value theory of motivation developed by Pintrich. In addition to value and expectancy for success, it is proposed that people take into account how much effort is required to complete a task. A 2 × 2 × 2 analysis of variance design was used to investigate the relationship of Difficulty, Consequence, and Required Effort on the willingness of participants to complete a task. The dependent variable was the score obtained from completing statistical problems that varied in difficulty and effort required to complete. 128 students at a public university were participants. Analysis showed significant main effects and an interaction for difficulty and required effort, but no main effect for consequence. Results are discussed in terms of creating optimal motivation for classroom tasks and for both classroom and standardized examinations. Implications for students in subject pools are also discussed.

Cambridge Dictionary of Statistics

B.S. Everitt (1998) Pp. viii +359. Cambridge University Press, Cambridge. £19.95 (hardback), ISBN 0-521-59346-8. This book by B.S. Everitt contains 360 pages. It was first published by CUP in 1998. It is a 50% expansion of Everitt’s earlier successful dictionary, has an attractive layout and is reasonably priced. The aim of the book is to ‘provide students of statistics, working statisticians and researchers in many disciplines who are users of statistics with relatively concise definitions of statistical terms’. Most pages contain between eight and ten definitions. Some are accompanied by one or more formulae. This is a valuable reference book for all statisticians, and medical researchers who use and need to understand statistics. It is not essential for someone learning statistics for the first time as it refers to many advanced methods. Most definitions are followed by a reference which the reader can source for more details. There is a list of the most important reference sources at the beginning of the book. As well as definitions, the dictionary includes concise biographies of over 100 statisticians who have made major contributions to the subject. It also includes entries on the major statistical packages. Some entries are very short. For example, the entry for ‘non-orthogonal designs’ is ‘Analysis of variance designs with two or more factors in which the number of observations in each cell are not equal’. Others are longer and include equations. For example, the entry on ‘Least squares estimation’ includes 94 words, four equations and a reference. Many other entries include graphs. Up-to-date statistical methods are defined, mainly using medical examples, so that the non-medical reader will find much irrelevant material. Examples given range from the very elementary to advanced and they reflect the author’s background to such an extent that a better title may be The Cambridge Dictionary of Medical Statistics. Students of applied biology following a basic course in statistics will find most of the book difficult to understand, mainly because many of the definitions pertain to topics not covered in introductory courses. For non-mathematical students, even elementary definitions may cause problems. For example, for someone who does not know the meaning of the term standard deviation the explanation ‘The most commonly used measure of the spread of a set of observations. Equal to the square root of the variance’ would lead to ‘the second moment about the mean’ as the entry for variance. The definition given for moment may then result in further problems of understanding. However, to be fair to the author, he does give references to other texts. In particular, for more information on variance, standard deviation and many other elementary definitions the student is referred to Practical Statistics for Medical Research by D.G. Altman (1991), published by Chapman and Hall. Professional statisticians will find the dictionary a useful reference to statistical methods they may not yet be familiar with. It is particularly useful as a reference book for medical researchers when reading scientific papers. This book should be purchased by statisticians, libraries that support medical and veterinary departments, and research workers in medicine.

"Quick and Easy" Formulae for Approximating Statistical Power in Repeated Measures ANOVA

The determination of statistical power for a repeated measures experimental study is a necessary, but difficult and seldom done, calculation. The purpose of this study was to provide practical methods for the a priori determination of power for the repeated measures (RM) analysis of variance (ANOVA). Stepwise regression analysis procedures were used to derive 10 power approximation equations for the main and interaction effects of the 1-way RM, 2-way mixed and 2-way RM ANOVA designs. The theoretically correct power values were calculated using the program DATASIM (Bradley, 1988) for various levels of effect size, number of levels of each factor, sample size, and mean correlations among RM factor levels. Potvin and Schutz's (in press) equations for estimating the error variance were utilized for computing power for the 2-way RM ANOVA. The derived equations showed a high level of precision in approximating power (R2 = .97 to .99, SEE = .0 12 to .035) with 9 to 16 predictor variables for the 1-way RM and the 2-way mixed designs. For the 2-way RM design, however, the levels of precision were relatively low for the main effect (R2 = .66 and .81 for α levels .01 and .05, respectively), and the residuals of the regression equation for the interaction effect showed a nonlinear relation with the criterion variable. Recommendations are given for the practical usage of the derived equations, and examples for calculating power for each design are provided.

Changes in intrauterine pressure after oxytocin administration in reproductively normal mares and in those with a delay in uterine clearance

Intrauterine pressure was measured in 4 reproductively normal mares and 4 mares with delay in uterine clearance after administration of oxytocin to determine if intrauterine pressure varied between dosage and group. Changes in intrauterine pressure were measured during estrus, when a follicle was ≥ 35 mm, using a Millar “Mikro-tip”® catheter that had 3 discrete pressure sensors/channels . Mares received 4 different treatments of 10, 5, 2.5 or 0 IU (vehicle) of oxytocin. The protocol for each treatment consisted of a 10-min baseline recording, administration of treatment and measurement of changes in intrauterine pressure for 65 min. After administration of the first two treatments, mares were rested for 2 h and the protocol repeated for the remaining 2 treatments. Changes in intrauterine pressure were measured on a physiograph and stored in a computer. The results were analyzed by 4×4 Latin Square Design analysis of variance (ANOVA) using the GLM procedure of the Statistical Analysis System. The ANOVA detected a main effect of treatment (P <0.01) and mare (nested within group; P <0.01) but no effect of channels, group or treatment-by-group interaction. There was a dose-dependent increase in uterine activity in both normal mares and those with delayed uterine clearance. A dose of 10 IU of oxytocin induced a larger number of uterine contractions (5.67 ± 0.06) for a longer time (24.09 ± 1.18 min) than the 5 IU (4.16 ± 0.06 contractions and 16.31 ± 1.18; P <0.01 min) or 2.5 IU dose (4.08 ± 0.06 contractions and 17.61 ± 1.18 min). The first intrauterine wave occurred most often near the tip of the horn in 10 of 12 recordings in normal mares and in 8 of 12 recordings in mares with delayed uterine clearance. It was then propagated from the middle of the horn to the uterine body just cranial to the cervix. There was no pattern of propagation for subsequent intrauterine pressure waves. We conclude that the difference in spontaneous clearance of the uterus between the 2 groups is not reflected in their response to exogenous oxytocin as determined by changes in intrauterine pressure.

Evidence in support of a model that predicts how biological and environmental factors interact to influence spatial skills.

It was proposed, based on M. Annett (1985), that individuals biologically predisposed to poorer spatial skills are less likely to capitalize on opportunities to develop these skills. Using an analysis of variance design assessing mental rotation skills in 2 cohorts of 8th graders (365 students), the authors found a significant 3-way interaction (Brothers x Family Handedness x Gender). For the girls with brothers, those from all right-handed families had lower mental rotation scores than did the other girls. For the 2nd cohort, among those children who participated in mental-rotation-type activities with their brothers, both boys and girls from all right-handed families performed more poorly on the mental rotation test than did the other children with brothers. Thus, compared with other children, the children from all right-handed families do not appear to be able to use their spatial experiences with male siblings to increase their spatial skills.

Amine-based aqueous polymers for the simultaneous titration and extraction of lactic acid in aqueous two-phase systems

The partitioning of 10% (w/w) lactic acid in ethylene oxide propylene oxide (EOPO) random copolymers and dextran T500 aqueous two-phase systems was studied. An analysis of variance design was applied to investigate the effect of pH, polymer concentration, and addition of polyethyleneimine to the aqueous two-phase systems. The lowest lactate partition coefficient of 0.09 was obtained at pH 6 in the systems containing 7.2% (w/w) polyethyleneimine. The use of polyethyleneimine as titrating base during the fermentative production of lactic acid was evaluated in batch fermentations with 100 g/l glucose. Yield and productivity of polyethyleneimine titrated fermentations compared with those obtained in fermentations titrated with NaOH and KOH.

COMMUNITY COLLEGE FACULTY RECRUITMENT PRACTICES: THE EFFECTS OF APPLICANT GENDER, INSTRUCTIONAL PROGRAMS, AND JOB ATTRIBUTES

Achieving desired results in education partly depends on the quality and motivation of the faculty hired to deliver instruction. Although recruitment is the first step in hiring capable faculty, no empirical research has been done about community college recruitment practices occurring prior to the employment interview. This study applied marketing and advertising theory to the task of recruiting community college business faculty. The reactions of two groups (male, female) of target applicants (N = 180) were assessed with respect to recruitment advertisements varied by instructional program description and by attributes used to describe the job. The investigation was conducted using a 2 × 3 × 3 factorial analysis of variance design. Male applicants reacted more favorably to advertisements when the organization was depicted as being oriented toward academic transfer instructional programs, and the job was described using extrinsic job attributes. Female applicants reacted more favorably when the organization was depicted as being oriented toward academic transfer instructional programs, and the job was described using intrinsic job attributes. Implications for practice, theory, and research are discussed.

Nonparametric alternatives to the f statistic in analysis of variance

Populations are frequently nonnormally distributed in applied behavioral and social science research, affecting the properties of the F test in Analysis of Variance designs. The purpose of this study is to compare some of the most promising nonparametric alternatives to the F test for interaction:McSweeney (Marascuilo and McSweeney, 1977), Harwell and Serlin (1989; Harwell, 1991), and Blair and Sawilowsky (1990). The properties of the competing statistics were examined on data sampled from four theoretical distributions and two real data sets in the context of the balanced 2 × 2 × 2 design. Cell sample sizes of n= 7, 21, and 35 were obtained, and statistics were calculated at the nominal α = 0.05 and 0.01 levels. The number of repetitions was set at 20,000 per experiment. Results indicated that Harwell and Serlin's L test was conservatively and liberally nonrobust, and not a powerful competitor. Although robust, McSweeney's test was conservative when the effect size was large, and as the number of nonnull effects increased, the test's power decreased. If the distribution is heavy-tailed or skewed, the Blair and Sawilowsky test statistic demonstrated superior power properties when compared with ANOVA. The F test retained a slight power advantage over the Blair and Sawilowsky statistic when testing for interactions on data sampled from populations that are symmetric with light tails, such as the normal or uniform.

Overcoming feelings of powerlessness in "aging" researchers: a primer on statistical power in analysis of variance designs.

A general rationale and specific procedures for examining the statistical power characteristics of psychology-of-aging empirical studies are provided. First, 4 basic ingredients of statistical hypothesis testing are reviewed. Then, 2 measures of effect size are introduced (standardized mean differences and the proportion of variation accounted for by the effect of interest), and methods are given for estimating these measures from already-completed studies. Power and sample size formulas, examples, and discussion are provided for common comparison-of-means designs, including independent samples I-factor and factorial analysis of variance (ANOVA) design, analysis of covariance designs, repeated measures (correlated samples) ANOVA designs, and split-plot (combined between- and within-subjects) ANOVA designs. Because of past conceptual differences, special attention is given to the power associated with statistical interactions, and cautions about applying the various procedures are indicated. Illustrative power estimations also are applied to a published study from the literature. It is argued that psychology-of-aging researchers will be both better informed consumers of what they read and more "empowered" with respect to what they research by understanding the important roles played by power and sample size in statistical hypothesis testing.

Overcoming feelings of powerlessness in "aging" researchers: A primer on statistical power in analysis of variance designs.

Overcoming feelings of powerlessness in "aging" researchers: A primer on statistical power in analysis of variance designs.

Cerebral blood flow correlates of apathy in Alzheimer disease.

Apathy is a pervasive noncognitive neuropsychiatric disturbance in Alzheimer disease, which causes significant caregiver distress. The neuroanatomical substrate of apathy is not well understood. To study the relationship between regional cerebral blood flow and the presence and severity of the personality disturbance, apathy, in individuals with Alzheimer disease. Analysis of the relationship between regional cerebral blood flow as measured by single photon emission computed tomography and severity of apathy as measured by the Neuropsychiatric Inventory using an analysis of variance design. We examined regional cerebral perfusion alterations as measured by xenon 133Xecalibrated technetium Tc 99m hexamethyl-propyleneamine-oxime single photon emission computed tomography in relation to the presence and severity of apathy. The neurology clinics of the University of California, Los Angeles, UCLA School of Medicine, and Harbor-UCLA Medical Center. Thirty-one community-dwelling patients fulfilling National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer's Disease and Related Disorders Association diagnostic criteria for probable Alzheimer disease who had a single photon computed tomographic scan performed within 3 months of administration of the Neuropsychiatric Inventory. The presence of apathy was associated with more severe prefrontal and anterior temporal dysfunction. These regional cerebral perfusion relationships with apathy were independent of cognitive decline except in the dorsolateral prefrontal cortex. These results demonstrate the association of apathetic syndromes with prefrontal and anterior temporal regional brain dysfunction and are consistent with similar findings previously reported in other disorders.

Agonist and antagonist moment and EMG-angle relationship during isokinetic eccentric and concentric exercise

The examination of moment and electromyographic (EMG) activity-angle curves of the agonists and antagonists is important for the determination of the mechanisms responsible for moment production during maximal activation of knee extensors and flexors. The purpose of this study was the examination of agonist and antagonistic activity of knee extensors and flexors at different knee joint angles during isokinetic exercise. Twelve females performed maximum isometric, eccentric and concentric isokinetic efforts of knee extensors and flexors on a Biodex dynamometer. The isokinetic tests were performed at angular velocities of 30, 60, 90, 120 and 150° Is. The EMG activity of vastus lateralis, rectus femoris, vastus medialis and hamstrings was also recorded. The knee extensor and flexor maximal activity was found at angles ranging from 50° to 70° and from 20° to 40°, respectively. The antagonist activity was greater at the initial and final stages of the movement. Three way analysis of variance designs indicated significant differences in the agonist and antagonistic activity of knee extensors and flexors between eccentric and concentric exercise. The results of this study demonstrate that, despite changes in magnitude, the moment-angle and IEMG activity-angle patterns of the agonists were not affected by angular velocity and muscle action conditions. On the contrary, the IEMG activity-angle curves of the antagonists during maximal isokinetic voluntary efforts are affected by angular velocity conditions but were independent of the type of muscle action examined. Copyright © 1996 Elsevier Science Ireland Ltd.

Item Response Theory Models and Spurious Interaction Effects in Factorial ANOVA Designs

In many psychological experiments, interaction effects in factorial analysis of variance (ANOVA) designs are often estimated using total scores derived from classical test theory. However, interaction effects can be reduced or eliminated by nonlinear monotonic transformations of a dependent variable. Although cross-over interactions cannot be eliminated by trans formations, the meaningfulness of other interactions hinges on achieving a measurement scale level for which nonlinear transformations are inappropriate (i.e., at least interval scale level). Classical total test scores do not provide interval level measurement according to contemporary item response theory (IRT). Nevertheless, rarely are IRT models applied to achieve more optimal measurement properties and hence more meaningful interaction effects. This paper provides several condi tions under which interaction effects that are estimated from classical total scores, rather than IRT trait scores, can be misleading. Using derived asymptotic expecta tions from an IRT model, interaction effects of zero on the IRT trait scale were often not estimated as zero from the total score scale. Further, when nonzero inter actions were specified on the IRT trait scale, the esti mated interaction effects were biased inward when estimated from the total score scale. Test difficulty level determined both the direction and the magnitude of the biased interaction effects. Index terms: facto rial designs, interaction effects, interval measurement, item response theory, level of measurement, measure ment scales, statistical inference.

Detecting Unit of Analysis Problems in Nested Designs: Statistical Power and Type I Error Rates of the F Test for Groups-within-Treatments Effects

The nested analysis of variance (ANOVA) model is often recommended for analysis of educational data in which students receive treatments within classrooms. The Type I error rate and statistical power of the F test for groups-within-treatments effects associated with such nested ANOVA designs were evaluated in a Monte Carlo study. Data were generated for ANOVA models comprising two and three levels of the treatment variable; two, three, and five groups nested within each treatment; and 3, 10, and 30 observations within each group. The intraclass correlation among scores within groups was controlled at levels ranging from zero (independent observations) to .50, and the F test for groups-within-treatments effects was conducted at nominal alpha levels ranging from .05 to .30. The results indicated that the test for groups within treatments did not indicate sufficient power, even at the most liberal alpha level examined, to detect intraclass correlation in the sample data. The procedure was not effective in supporting decisions to use individual observations rather than group means as the unit of analysis.