The analysis of repeated measures designs: A review

Department of Psychology, University of Manitoba, 190 Dysart Road, Winnipeg, Manitoba, Canada R3T 2N2.
British Journal of Mathematical and Statistical Psychology (Impact Factor: 2.17). 06/2001; 54(Pt 1):1-20. DOI: 10.1348/000711001159357
Source: PubMed

ABSTRACT Repeated measures ANOVA can refer to many different types of analysis. Specifically, this vague term can refer to conventional tests of significance, one of three univariate solutions with adjusted degrees of freedom, two different types of multivariate statistic, or approaches that combine univariate and multivariate tests. Accordingly, it is argued that, by only reporting probability values and referring to statistical analyses as repeated measures ANOVA, authors convey neither the type of analysis that was used nor the validity of the reported probability value, since each of these approaches has its own strengths and weaknesses. The various approaches are presented with a discussion of their strengths and weaknesses, and recommendations are made regarding the 'best' choice of analysis. Additional topics discussed include analyses for missing data and tests of linear contrasts.

Download full-text


Available from: Harvey Jay Keselman, Sep 26, 2015
365 Reads
  • Source
    • "This procedure is especially adequate to analyze changes in heterogeneous groups such as patients recovering from brain injury [22]. Mixed models use all available data, can properly account for correlation between repeated measurements on the same subject, and have greater flexibility to model time effects [36]. LoC was included as fixed factor and the individual subjects were included as random factors. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Objectifs Des stimuli auditifs sont souvent appliqués aux patients non répondants qui souffrent de traumatisme crânien sévère. Nous avons étudié les processus visuels des patients en état végétatif au cours de leur évolution vers un état de conscience au moyen de potentiels évoqués visuels (PEV) et de réponses comportementales à une stimulation visuelle. Méthodes Les réponses à la stimulation visuelle – localisation visuelle, compréhension des instructions écrites, et manipulation d’objets – et les PEV ont été mesurés toutes les deux semaines durant une période moyenne de 2,6 mois chez onze patients en état végétatif au cours de leur évolution vers un état de conscience. Les PEV des patients étaient comparés à ceux mesurés dans un groupe de sujets témoins. Une mesure de la récupération à long terme a été réalisée 2 à 3 ans plus tard. Résultats Les réponses visuelles comportementales étaient plus grandes au cours de l’évolution de l’état végétatif vers l’état de conscience pour toutes les sous-échelles. Les amplitudes des PEV étaient moindres et les temps de latence étaient plus longs chez les patients par comparaison au groupe témoin. Les caractéristiques des PEV lors de la première mesure étaient corrélées au niveau de récupération à long terme. Conclusion Nos résultats montrent une amélioration des réponses visuelles parallèlement à la récupération de la conscience chez les patients initialement en état végétatif. Le traitement visuel élémentaire est présent, mais, au vu des PEV, moindre chez les patients en état végétatif et en état de conscience minimale par comparaison aux sujets témoins et il demeure moindre chez les patients ayant récupéré un état de conscience plus élevé. Cependant, les PEV obtenus initialement sont corrélés avec la récupération à long terme.
    Neurophysiologie Clinique/Clinical Neurophysiology 09/2014; 44(5). DOI:10.1016/j.neucli.2014.08.008 · 1.24 Impact Factor
  • Source
    • "In cases of the violation of the assumption of sphericity, Greenhouse Geisser corrected p-values for the F-statistics are reported. Unequal sample sizes and heterogeneous variances between groups may lead to progressive decisions in favour of the H1 [82]. Therefore, we subsequently inspected significant repeated measures ANOVA effects by means of the Welch test (Table 1), when heterogeneous variances are assumed. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The Mismatch Negativity (MMN) is an event-related potential (ERP) sensitive to early auditory deviance detection and has been shown to be reduced in schizophrenia patients. Moreover, MMN amplitude reduction to duration deviant tones was found to be related to functional outcomes particularly, to neuropsychological (working memory and verbal domains) and psychosocial measures. While MMN amplitude is thought to be correlated with deficits of early sensory processing, the functional significance of MMN latency remains unclear so far. The present study focused on the investigation of MMN in relation to neuropsychological function in schizophrenia. Forty schizophrenia patients and 16 healthy controls underwent a passive oddball paradigm (2400 binaural tones; 88% standards [1 kHz, 80 db, 80 ms], 11% frequency deviants [1.2 kHz], 11% duration deviants [40 ms]) and a neuropsychological test-battery. Patients were assessed with regard to clinical symptoms. Compared to healthy controls schizophrenia patients showed diminished MMN amplitude and shorter MMN latency to both deviants as well as an impaired neuropsychological test performance. Severity of positive symptoms was related to decreased MMN amplitude to duration deviants. Furthermore, enhanced verbal memory performance was associated with prolonged MMN latency to frequency deviants in patients. The present study corroborates previous results of a diminished MMN amplitude and its association with positive symptoms in schizophrenia patients. Both, the findings of a shorter latency to duration and frequency deviants and the relationship of the latter with verbal memory in patients, emphasize the relevance of the temporal aspect of early auditory discrimination processing in schizophrenia.
    PLoS ONE 04/2014; 9(4):e84536. DOI:10.1371/journal.pone.0084536 · 3.23 Impact Factor
  • Source
    • "The common assumption that data is missing at random was made and hence unbalanced data were allowed (this procedure is implemented within SAS). Mixed-effect models address missing data by providing numerical solutions based on all available data [44]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Outcomes of health and rehabilitation services for children and youth with disabilities increasingly include assessments of health-related quality of life (HRQoL). The purpose of this research was to 1) describe overall patterns of HRQoL, 2) examine changes in parent's perceptions of child's HRQoL across 18 months and 3) explore factors that predict these changes. Participants in this study included 427 parents of children (229 boys and 198 girls) with a physically-based disability between the ages of 6 to 14 years. The Child Health Questionnaire (CHQ) was administered three times, at nine month intervals. Comparisons to the CHQ normative data were analyzed at Time 1 using t-tests, and change over time was examined using linear mixed-effects models. Possible predictors were modeled: 1) child's factors measured by the Activities Scale for Kids, Strengths and Difficulties Questionnaire, and general health measured by SF-36, 2) family characteristics measured by the Impact on Family Scale and 3) environmental barriers measured by the Craig Hospital Inventory of Environmental Factors. CHQ scores of the study's participants demonstrated significantly lower summary scores from the normative sample for both CHQ Physical and Psychosocial summary scores. On average, children did not change significantly over time for physical summary scores. There was an average increase in psychosocial health that was statistically significant, but small. However, there was evidence of heterogeneity among children. Environmental barriers, behavioral difficulties, family functioning/impact, general health and child physical functioning had negative and significant associations with physical QoL at baseline. Change in physical QoL scores over time was dependent on children's behavioral difficulties, family functioning and environmental barriers. Environmental barriers, behavioral difficulties, family functioning/impact and general health had significant associations with psychosocial scores at baseline, but none served as predictors of change over time. Children with physical disabilities differ from the normative group on parent ratings of their physical and psychosocial health. While there was little average change in CHQ scores over 18 months, there is evidence of heterogeneity among children. Factors such as environmental barriers, family functioning/impact, child physical functioning and behavioral difficulties and general health significantly influence QoL scores as measured by the CHQ.
    BMC Pediatrics 01/2014; 14(1):26. DOI:10.1186/1471-2431-14-26 · 1.93 Impact Factor
Show more