PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Factor Analysis”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2Linked to original sources
PubMed2004

Convergent and discriminative validity of interview and questionnaire measures of personality disorder in mentally disordered offenders: a multitrait-multimethod analysis using confirmatory factor analysis.

Measures of personality disorder from the International Personality Disorder Examination, Personality Diagnostic Questionnaire, and Millon Clinical Multiaxial Inventory (MCMI-II) were obtained from detained male mentally disordered offenders (N = 156), and convergent and discriminant validity were examined by confirmatory factor analysis of the multitrait-multimethod matrix. Hierarchical comparisons of models varying in their specification of trait and method variance established the appropriateness of a model supporting both convergence and discrimination across methods, but these were variable across constructs and measures. Convergence was good for avoidant, schizoid, and antisocial disorders, but poor for histrionic, narcissistic, and obsessive-compulsive disorders. Avoidant, schizoid, and schizotypal disorders were not clearly distinguishable from each other. Measurement error attributable to method variance was substantial for all instruments and for most disorders. The commonly alleged superiority of interview over questionnaire methods was not supported, and the MCMI-II demonstrated proportionately more "true" variance. However, assessment methods may be differentially sensitive to different kinds of personality disorder problems.

Adult↗
PubMedDate not supplied

A guide for applying principal-components analysis and confirmatory factor analysis to quantitative electroencephalogram data.

Principal-components analysis (PCA) has been used in quantitative electroencephalogram (qEEG) research to statistically reduce the dimensionality of the original qEEG measures to a smaller set of theoretically meaningful component variables. However, PCAs involving qEEG have frequently been performed with small sample sizes, producing solutions that are highly unstable. Moreover, solutions have not been independently confirmed using an independent sample and the more rigorous confirmatory factor analysis (CFA) procedure. This paper was intended to illustrate, by way of example, the process of applying PCA and CFA to qEEG data. Explicit decision rules pertaining to the application of PCA and CFA to qEEG are discussed. In the first of two experiments, PCAs were performed on qEEG measures collected from 102 healthy individuals as they performed an auditory continuous performance task. Component solutions were then validated in an independent sample of 106 healthy individuals using the CFA procedure. The results of this experiment confirmed the validity of an oblique, seven component solution. Measures of internal consistency and test-retest reliability for the seven component solution were high. These results support the use of qEEG data as a stable and valid measure of neurophysiological functioning. As measures of these neurophysiological processes are easily derived, they may prove useful in discriminating between and among clinical (neurological) and control populations. Future research directions are highlighted.

Adolescent↗
PubMed1983

The COMSTAT algorithm for multimodal factor analysis: an improvement of Tucker's three-mode factor analysis method.

Three-mode factor analysis, as developed by Tucker in 1966, is a method for nonredundant representation of data arrays with three subscripts (e.g. observations classified according to persons by variables by conditions). Tucker , however, did not succeed in obtaining a least-squares solution for his model. We derive in this paper a necessary condition for a least-squares solution and construct an algorithm which improves any given initial solution in the least-squares sense. We show that this algorithm converges to a representation satisfying the necessary conditions for a least-squares solution. The method is in no way restricted to three dimensions, but can be applied to any multimodal data array.

Alpha Rhythm↗
PubMed2005

An illustration of multilevel factor analysis.

The factor analysis of repeated measures psychiatric data presents interesting challenges for researchers in terms of identifying the latent structure of an assessment instrument. Specifically, repeated measures contain both within and between individual sources of variance. Although a number of techniques exist for separating out these 2 sources of variance, all are problematic. Recently, researchers have proposed that exploratory multilevel factor analysis (MFA) be used to appropriately analyze the latent structure of repeated measures data. The chief objective of this report is to provide a didactic step-by-step guide on how MFA may be applied to psychiatric data. In the discussion, we describe difficulties associated with MFA and consider challenges in factor analyzing life event appraisals in psychiatric samples.

Adolescent↗
PubMed2003

An approach to construct simplified measures of dietary patterns from exploratory factor analysis.

Exploratory factor analysis might work well in elucidating the major dietary patterns prevailing in specific study populations. However, patterns extracted in one study population and their associations with disease risk cannot be reproduced with this data-specific method in other study populations. To construct less population-dependent pattern variables of similar content as original exploratory patterns, we proposed to derive so-called simplified pattern variables. They represent the sum of the unweighted standardised food variables which loaded high at the pattern of interest. Data from the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam study suggest that these simplified pattern variables might adequately approximate factor analysis-based dietary patterns. A simplified pattern variable based on the six highest loading food variables showed a correlation >0.95 with the originally derived factor score, which consisted of forty-seven food variables. Moreover, simplified pattern variables might adequately approximate patterns across different study populations. A simplified pattern variable showed similar factor loadings, ranging from 0.34 to 0.52, as well as similar associations with nutrient intake as a 'western' pattern originally reported from an US study population. These simplified pattern variables can subsequently be used to study pattern associations with disease risk, especially in multi-centre studies. It is therefore an approach that might overcome one of the most frequently claimed limitations of factor analyses applied in epidemiology: their non-comparable risk estimates.

Diet↗
PubMed2002

The Age Universal I-E Scale-12 and orientation toward religion: confirmatory factor analysis.

Confirmatory factor analysis was used to explore the suggested 2-factor and 3-factor models of the Age Universal I-E Scale-12 among 4,160 respondents. The present findings suggested that the 3-factor model provided a better fit of the data than the 2-factor model did. The findings with this version of the Age Universal I-E Scale were consistent with recent theories that the main distinctions in intrinsic and extrinsic religious orientation comprise intrinsic, extrinsic-personal, and extrinsic-social religious dimensions.

Adolescent↗
PubMed1993

Factor structure of the Chinese version of the General Health Questionnaire (GHO-30): a confirmatory factor analysis.

The factor structure of responses to the Chinese version of the General Health Questionnaire (CGHQ-30) in a sample of 2,150 Chinese secondary school students was examined using the LISREL approach to confirmatory factor analysis. The results showed that while a five-factor model (Anxiety, Depression, Inadequate Coping, Social Dysfunctioning and Sleep Disturbances) was able to fit the data, a higher-order factor model with five primary factors (Anxiety, Depression, Inadequate Coping, Social Dysfunctioning and Sleep Disturbances) and a second-order factor (General Psychopathology) was found to be a more parsimonious model. The present findings are discussed with reference to the controversies that surround the dimensionality of the General Health Questionnaire.

Adaptation, Psychological↗
PubMed1994

Applications of standard error estimates in unrestricted factor analysis: significance tests for factor loadings and correlations.

Estimates of standard errors of factor loadings and factor correlations in the unrestricted factor analysis model can be computed for oblique or orthogonal solutions under maximum likelihood. This information can be used to test individual coefficients for significance, to evaluate whether an orthogonal or oblique structure is most consistent with sample data, or to compute confidence intervals for single parameters or confidence regions for arbitrary groups of coefficients. Because the number of parameters estimated in factor analysis is approximately the product of number of variables multiplied by number of factors, a Bonferroni correction for the critical point of the individual test statistics is recommended to control the probability of a Type I error. Several examples are presented.

Aptitude Tests↗
PubMed2004

Factor analysis of household factors: are they associated with respiratory conditions in Chinese children?

BACKGROUND: We explored methods to develop uncorrelated variables for epidemiological analysis models. They were used to examine associations between respiratory health outcomes and multiple household risk factors. METHODS: We analysed data collected in the Four Chinese Cities Study (FCCS) to examine health effects on prevalence rates of respiratory symptoms and illnesses in 7058 school children living in the four Chinese cities: Lanzhou, Chongqing, Wuhan, and Guangzhou. We used factor analysis approaches to reduce the number of the children's lifestyle/household variables and to develop new uncorrelated 'factor' variables. We used unconditional logistic regression models to examine associations between the factor variables and the respiratory health outcomes, while controlling for other covariates. RESULTS: Five factor variables were derived from 21 original variables: heating coal smoke, cooking coal smoke, socioeconomic status, ventilation, and environmental tobacco smoke (ETS) and parental asthma. We found that higher exposure to heating coal smoke was associated with higher reporting of cough with phlegm, wheeze, and asthma. Cooking coal smoke was not associated with any of the outcomes. Lower socioeconomic status was associated with lower reporting of persistent cough and bronchitis. Higher household ventilation was associated with lower reporting of persistent cough, persistent phlegm, cough with phlegm, bronchitis, and wheeze. Higher exposure to ETS and the presence of parental asthma were associated with higher reporting of persistent cough, persistent phlegm, cough with phlegm, bronchitis, wheeze, and asthma. CONCLUSIONS: Our study suggests that independent respiratory effects of exposure to indoor air pollution, heating coal smoke, and ETS may exist for the studied children.

Air Pollution, Indoor↗
PubMed2001

A confirmatory factor analysis of the cognitive capacity screening examination in a clinical sample.

Structured mental status examinations offer several advantages over unstructured mental status examinations; however, few have been subjected to advanced psychometric analysis. Confirmatory factor analysis empirically tests the predictive validity of individual test indices within an a priori methodological framework. With such analysis, one can test hypotheses about the structure of latent variability within a given data set. The purpose of this study was to perform a confirmatory factor analysis of the Cognitive Capacity Screening Examination, the most comprehensive of the brief structured mental status examinations. A confirmatory factor analysis of the Cognitive Capacity Screening Examination (CCSE) was performed by applying LISREL 7 to a sample of 924 male veterans, 409 patients from a chemical dependence treatment program, and 515 individuals from a psychology consultation service. Constructs were derived from previous exploratory analysis of the scale. The results of the confirmatory factor analysis and three indices of model fit support an 11 factor model more complex than that originally formulated for the CCSE. However, three of these factors (digit span with interference, complex mental mathematics, and verbal memory) were more sensitive to impairment than other factors, accounting for over 90% of the CCSE total score variance. Although the CCSE is a more complex test than originally envisioned by its designers, it may not be necessary to give all items on the test. Either a subset of the CCSE items (the CCSE-A) or a relatively brief, informal mental status exam may be adequate for many patients.

Cognition Disorders↗
PubMed2006

Factor analysis of neuroanatomical and clinical characteristics of holoprosencephaly.

The objective of this study is to better understand the relationship between neuroradiologic and clinical characteristics in holoprosencephaly (HPE) using the multivariate analysis called factor analysis. HPE is a brain malformation characterized by incomplete cleavage of the cerebral hemispheres and deep gray structures. We performed evaluations on 89 children with HPE that included their history, developmental assessment, and physical examination. Ten clinical variables included in factor analysis were the grade of spasticity, dystonia, choreoathetosis, hypotonia, mobility, upper extremity/hand function, expressive language, feeding/swallowing difficulty, endocrinopathies, and temperature dysregulation. Five neuroimaging variables graded by pediatric neuroradiologists were the grade of HPE (from least to most severe: lobar, semilobar, and alobar) and the degree of non-separation of caudate, lentiform, thalamic, and hypothalamic nuclei. Factor analysis using principle component extraction and varimax rotation was utilized. Four significant factors were identified: (1) neuroimaging/developmental factor, (2) motor factor, (3) hypothalamic/oromotor factor, and (4) hypotonia factor. These four factors accounted for 65.2% of the variance. In this factor analysis of HPE patients, we were able to reduce the large number of clinical and radiological variables into four factors. These factors and the constructs underlying them provide structure to the data and provide key parameters for future studies involving neurodevelopmental outcomes in HPE.

Brain↗
PubMedDate not supplied

A confirmatory factor analysis of the Caregiving Appraisal Scale for caregivers of home-based ventilator-assisted individuals.

OBJECTIVE: To confirm the underlying dimensions of the Caregiving Appraisal Scale (CAS) with use of data collected from caregivers of home-based ventilator-assisted individuals (VAIs). DESIGN: Cross-sectional survey. SETTING: Residences of home-based VAIs. SAMPLE: Two hundred seventy-seven primary family caregivers of VAIs. MEASURES: Twenty-eight-item CAS developed by Lawton et al. (1989), and an investigator-developed instrument to assess physical health and sociodemographic characteristics of both VAIs and their caregivers. INTERVENTION: None. ANALYSIS: Confirmatory factor analysis with principal components extraction. An oblique (oblimin) solution was used for rotation of the factor matrices. The number of common factors needed to obtain the best fit of the factor model was determined with use of maximum-likelihood estimation. Confirmatory factor analysis with linear structural equation modeling was also performed. RESULTS: Confirmatory factor analysis did not fully replicate the factor structure proposed by Lawton et al. CONCLUSIONS: The model proposed by Lawton et al. provides a useful foundation for examining the appraisal of family caregivers of home-based VAIs. Additional development work is needed for the CAS.

Adult↗
PubMed1989

[Serodiagnostic tests by factor analysis and stepwise discriminating analysis with tumor markers for the detection of ovarian cancer].

We measured five tumor markers simultaneously for serodiagnostic testing as a method for the early detection of ovarian cancer. To decrease both false negativity and false positivity in the results of combination assay, statistical analysis including factor analysis and stepwise discriminating function was applied in this study. At least one of these tumor markers was detected as positive in 75.2% (76 of 101 patients) of sera from patients with ovarian cancer before treatment. Six hundred and ninety-three of 7,097 normal sera (9.8%) gave spuriously positive combination assay results. Falsely positive combination assay results were observed in 1,107 of 3,139 patients with benign disease, which could largely be attributed to the high CA125 values in patients with endometriosis. On the basis of factor analysis in order to decrease false positivity, CA125, TPA, and CA125/TPA were selected as the best parameters for distinguishing between ovarian cancer and benign conditions including those pelvic endometriosis showing positive results in combination assays. Subsequently, the function derived by factor analysis made possible the correct classification of 58 of 71 patients with ovarian cancer detected by combination assay, and 84.8% of pelvic endometriosis and benign ovarian tumor subjects were correctly classified into the non-cancer group. Next, in order to decrease false negativity, statistical analysis was also applied. Malignancy could be correctly diagnosed by this procedure in 14 of 25 patients whose tumor was undetectable by combination assay, whereas of subjects without cancer 14.5% were errorenously classified into the ovarian cancer group. In the search for the best method for accurately detecting ovarian cancer, we used image diagnosis (ultrasonography) in combination with serological diagnosis.(ABSTRACT TRUNCATED AT 250 WORDS)

Adult↗
PubMed1995

Confirmatory factor analysis of a 4-factor model of chronic pain evaluation.

Extending the earlier work of Mikail et al. (1993), a confirmatory factor analysis (CFA; LISREL VII) of a 4-factor model of pain assessment was tested. This model, comprised of the Beck Depression Inventory (BDI) and 13 subscales of the McGill Pain Questionnaire (MPQ) and the West Haven-Yale Multidimensional Pain Inventory (WHYMPI), adequately accounted for the pain experience with minimal overlap. Subjects were 306 outpatient chronic pain patients seen at a multidisciplinary chronic pain clinic. Subscale scores were subjected to CFA procedures that yielded a well-fitting final model that explained 91% of the covariance in the observed data. This final model was derived through an exploratory post-hoc procedure that allowed for correlated errors among subscales of the same instrument. The 4 factors were identified as Affective Distress, Support, Pain Description, and Functional Capacity. Results supported the hypothesis that the MPQ, WHYMPI and BDI are representative of the multidimensionality of the pain experience with minimal overlap among measures. Theoretical and clinical implications of reducing the overlap among existing measures in the assessment of pain patients are discussed.

Adult↗
PubMed1995

Examples for the improvements in AES depth profiling of multilayer thin film systems by application of factor analysis data evaluation.

Factor analysis has proved to be a powerful tool for the full exploitation of the chemical information included in the peak shapes and peak positions of spectra measured by AES depth profiling. Due to its ability to extract the number of independent chemical components, their spectra and their depth distributions, its information content exceeds the one of the usual peak-to-peak height evaluation of AES depth profile data. Using modern software with a graphically interactive user interface the analyst is put into a position, where he can work with Factor Analysis on a physically intuitive level despite of all the matrix algebra mathematics which it is based upon. The progress brought about by Factor Analysis to AES depth profiles of thin films is demonstrated by the analysis of two thin film systems. The first one is a Pt/Ti metallisation used as bottom electrode for ferroelectric thin films, the second one is a multilayer system where a Ti silicide formation of buried Ti/Si bilayers has been induced. Both examples show that Factor Analysis evaluation of AES depth profile data is capable to give access to stoichiometry information and to reveal interfacial layer phases, information which is hardly obtained from the conventional peak-to-peak height data evaluation.

Journal Article↗
PubMed2002

Exploratory factor analysis in behavior genetics research: factor recovery with small sample sizes.

Results of a Monte Carlo study of exploratory factor analysis demonstrate that in studies characterized by low sample sizes the population factor structure can be adequately recovered if communalities are high, model error is low, and few factors are retained. These are conditions likely to be encountered in behavior genetics research involving mean scores obtained from sets of inbred strains. Such studies are often characterized by a large number of measured variables relative to the number of strains used, highly reliable data, and high levels of communality. This combination of characteristics has special consequences for conducting factor analysis and interpreting results. Given that limitations on sample size are often unavoidable, it is recommended that researchers limit the number of expected factors as much as possible.

Animals↗