Data analysis methods in optometry Part 8: Principal components analysis (PCA) and factor analysis (FA)

PCA/FA is a method of analyzing complex data sets in which there are no clearly defined X or Y variables. It has multiple uses including the study of the pattern of variation between individual entities such as patients with particular disorders and the detailed study of descriptive variables. In most applications, variables are related to a smaller number of ‘factors’ or PCs that account for the maximum variance in the data and hence, may explain important trends among the variables. An increasingly important application of the method is in the ‘validation’ of questionnaires that attempt to relate subjective aspects of a patients experience with more objective measures of vision.

A globe-kutatás kultúraváltozóinak vizsgálata faktoranalízis segítségével (Cultural variables of the GLOBE research computed by factor analysis)

A szervezeti kultúraváltozók nem függetlenek egymástól. A tanulmány a GLOBE társadalmi kultúra, társadalmi értékek és a kettő különbségéből képzett differenciaváltozók faktoranalízisével kísérel meg főfaktorokat meghatározni és azokat értelmezni. / === / The cultural variables are not independent. This paper describes and discusses consolidated cultural variables computed by factor analysis from 9 original GLOBE variables of societal paracices, societal values, and differentiation scales.

A Q factor analysis of college undergraduate students' study behaviors

The purpose of this study was to better understand the study behaviors and habits of university undergraduate students. It was designed to determine whether undergraduate students could be grouped based on their self-reported study behaviors and if any grouping system could be determined, whether group membership was related to students’ academic achievement. A total of 152 undergraduate students voluntarily participated in the current study by completing the Study Behavior Inventory instrument. All participants were enrolled in fall semester of 2010 at Florida International University. The Q factor analysis technique using principal components extraction and a varimax rotation was used in order to examine the participants in relation to each other and to detect a pattern of intercorrelations among participants based on their self-reported study behaviors. The Q factor analysis yielded a two factor structure representing two distinct student types among participants regarding their study behaviors. The first student type (i.e., Factor 1) describes proactive learners who organize both their study materials and study time well. Type 1 students are labeled “Proactive Learners with Well-Organized Study Behaviors”. The second type (i.e., Factor 2) represents students who are poorly organized as well as being very likely to procrastinate. Type 2 students are labeled Disorganized Procrastinators. Hierarchical linear regression was employed to examine the relationship between student type and academic achievement as measured by current grade point averages (GPAs). The results showed significant differences in GPAs between Type 1 and Type 2 students at the .05 significance level. Furthermore, student type was found to be a significant predictor of academic achievement beyond and above students’ attribute variables including sex, age, major, and enrollment status. The study has several implications for educational researchers, practitioners, and policy makers in terms of improving college students' learning behaviors and outcomes.

Analysis of the C-SOSI Instrument in a 9-1-1 Telecommunicator Population

Thesis (Master's)--University of Washington, 2016-08

Multiple Imputation of missing values in exploratory factor analysis of multidimensional scales: estimating latent trait scores

ABSTRACT Researchers frequently have to analyze scales in which some participants have failed to respond to some items. In this paper we focus on the exploratory factor analysis of multidimensional scales (i.e., scales that consist of a number of subscales) where each subscale is made up of a number of Likert-type items, and the aim of the analysis is to estimate participants' scores on the corresponding latent traits. We propose a new approach to deal with missing responses in such a situation that is based on (1) multiple imputation of non-responses and (2) simultaneous rotation of the imputed datasets. We applied the approach in a real dataset where missing responses were artificially introduced following a real pattern of non-responses, and a simulation study based on artificial datasets. The results show that our approach (specifically, Hot-Deck multiple imputation followed of Consensus Promin rotation) was able to successfully compute factor score estimates even for participants that have missing data.

Factor analysis applied to genome prediction for high-dimensional phenotypes in pigs.

The aim of the present study was to propose and evaluate the use of factor analysis (FA) in obtaining latent variables (factors) that represent a set of pig traits simultaneously, for use in genome-wide selection (GWS) studies. We used crosses between outbred F2 populations of Brazilian Piau X commercial pigs. Data were obtained on 345 F2 pigs, genotyped for 237 SNPs, with 41 traits. FA allowed us to obtain four biologically interpretable factors: ?weight?, ?fat?, ?loin?, and ?performance?. These factors were used as dependent variables in multiple regression models of genomic selection (Bayes A, Bayes B, RR-BLUP, and Bayesian LASSO). The use of FA is presented as an interesting alternative to select individuals for multiple variables simultaneously in GWS studies; accuracy measurements of the factors were similar to those obtained when the original traits were considered individually. The similarities between the top 10% of individuals selected by the factor, and those selected by the individual traits, were also satisfactory. Moreover, the estimated markers effects for the traits were similar to those found for the relevant factor.

Factor structure and measurement invariance of the Health Education Impact Questionnaire: does the subjectivity of the response perspective threaten the contextual validity of inferences?

Objective:
On-going evidence is required to support the validity of inferences about change and group differences in the
evaluation of health programs, particularly when self-report scales requiring substantial subjectivity in response generation are used as outcome measures. Following this reasoning, the aim of this study was to replicate the factor structure and investigate the measurement invariance of the latest version of the Health Education Impact Questionnaire, a widely used health program evaluation measure.
Methods:
An archived dataset of responses to the most recent version of the English-language Health Education Impact
Questionnaire that uses four rather than six response options (N=3221) was analysed using exploratory structural equation
modelling and confirmatory factor analysis appropriate for ordered categorical data. Metric and scalar invariance were
studied following recent recommendations in the literature to apply fully invariant unconditional models with minimum
constraints necessary for model identification.
Results:
The original eight-factor structure was replicated and all but one of the scales (Self Monitoring and Insight) was
found to consist of unifactorial items with reliability of ⩾0.8 and satisfactory discriminant validity. Configural, metric and scalar
invariance were established across pre-test to post-test and population sub-groups (sex, age, education, ethnic background).
Conclusion:
The results support the high level of interest in the Health Education Impact Questionnaire, particularly for use as a pre-test/post-test measure in experimental studies, other pre–post evaluation designs and system-level monitoring and evaluation.

Measuring chronic condition self-management in an Australian community: factor structure of the revised Partners in Health (PIH) scale

Purpose To evaluate the factor structure of the revised Partners in Health (PIH) scale for measuring chronic condition self-management in a representative sample from the Australian community.

Methods A series of consultations between clinical groups underpinned the revision of the PIH. The factors in the revised instrument were proposed to be: knowledge of illness and treatment, patient–health professional partnership, recognition and management of symptoms and coping with chronic illness. Participants (N = 904) reporting having a chronic illness completed the revised 12-item scale. Two a priori models, the 4-factor and bi-factor models were then evaluated using Bayesian confirmatory factor analysis (BCFA). Final model selection was established on model complexity, posterior predictive p values and deviance information criterion.

Results Both 4-factor and bi-factor BCFA models with small informative priors for cross-loadings provided an acceptable fit with the data. The 4-factor model was shown to provide a better and more parsimonious fit with the observed data in terms of substantive theory. McDonald’s omega coefficients indicated that the reliability of subscale raw scores was mostly in the acceptable range.

Conclusion The findings showed that the PIH scale is a relevant and structurally valid instrument for measuring chronic condition self-management in an Australian community. The PIH scale may help health professionals to introduce the concept of self-management to their patients and provide assessment of areas of self-management. A limitation is the narrow range of validated PIH measurement properties to date. Further research is needed to evaluate other important properties such as test–retest reliability, responsiveness over time and content validity.

The posttraumatic growth inventory: Factor structure and invariance in a sample ofbreast cancer patients and in a non-clinical sample

The Posttraumatic Growth Inventory (PTGI) is frequently used to assess positive changes following a traumatic event. The aim of the study is to examine the factor structure and the latent mean invariance of PTGI. A sample of 205 (M age = 54.3, SD = 10.1) women diagnosed with breast cancer and 456 (M age = 34.9, SD = 12.5) adults who had experienced a range of adverse life events were recruited to complete the PTGI and a socio-demographic questionnaire. We use Confirmatory Factor Analysis (CFA) to test the factor-structure and multi-sample CFA to examine the invariance of the PTGI between the two groups. The goodness of fit for the five-factor model is satisfactory for breast cancer sample (χ2(175) = 396.265; CFI = .884; NIF = .813; RMSEA [90% CI] = .079 [.068, .089]), and good for non-clinical sample (χ2(172) = 574.329; CFI = .931; NIF = .905; RMSEA [90% CI] = .072 [.065, .078]). The results of multi-sample CFA show that the model fit indices of the unconstrained model are equal but the model that uses constrained factor loadings is not invariant across groups. The findings provide support for the original five-factor structure and for the multidimensional nature of posttraumatic growth (PTG). Regarding invariance between both samples, the factor structure of PTGI and other parameters (i.e., factor loadings, variances, and co-variances) are not invariant across the sample of breast cancer patients and the non-clinical sample.

Neonatal palliative care attitude scale : development of an instrument to measure the barriers to and facilitators of palliative care in neonatal nursing

OBJECTIVE The aim of this research project was to obtain an understanding of the barriers to and facilitators of providing palliative care in neonatal nursing. This article reports the first phase of this research: to develop and administer an instrument to measure the attitudes of neonatal nurses to palliative care. METHODS The instrument developed for this research (the Neonatal Palliative Care Attitude Scale) underwent face and content validity testing with an expert panel and was pilot tested to establish temporal stability. It was then administered to a population sample of 1285 neonatal nurses in Australian NICUs, with a response rate of 50% (N 645). Exploratory factor-analysis techniques were conducted to identify scales and subscales of the instrument. RESULTS Data-reduction techniques using principal components analysis were used. Using the criteria of eigenvalues being 1, the items in the Neonatal Palliative Care Attitude Scale extracted 6 factors, which accounted for 48.1% of the variance among the items. By further examining the questions within each factor and the Cronbach’s of items loading on each factor, factors were accepted or rejected. This resulted in acceptance of 3 factors indicating the barriers to and facilitators of palliative care practice. The constructs represented by these factors indicated barriers to and facilitators of palliative care practice relating to (1) the organization in which the nurse practices, (2) the available resources to support a palliative model of care, and (3) the technological imperatives and parental demands. CONCLUSIONS The subscales identified by this analysis identified items that measured both barriers to and facilitators of palliative care practice in neonatal nursing. While establishing preliminary reliability of the instrument by using exploratory factor-analysis techniques, further testing of this instrument with different samples of neonatal nurses is necessary using a confirmatory factor-analysis approach.

Fan motivation and loyalty : extending the Sport Interest Inventory (SII) to the Australian Football League

This study used the Sport Interest Inventory (SII) to examine the motivation of fans attending a game in the Australian Football League. This is the first study to use the SII for professional men’s team sport outside the United States. Confirmatory factor analysis showed the model provided a good fit for the data collected in Australia, and regression analysis revealed that team interest, vicarious achievement, excitement and player interest were the significant factors in predicting and explaining the level of attitudinal loyalty of fans toward their favourite team.

Measuring school-related stressors in adolescents

Some 1620 high school students responded to 68 items that measure adolescent stressors. Thirty-five of the items were based on J. P. Kohn and G. H. Frazer's Academic Stress Scale [1(1986) An Academic Stress Scale: Identification and Rated Importance of Academic Stressors, Psychological Reports, Vol. 59, pp. 415–426] developed in the United States, while the remaining 33 items were developed from P. Strutynski's [(1985) A Survey of Queensland Year 10, 11 and 12 Student Attitudes to Schools and Schooling, State Planning Committee, International Youth Year, Brisbane] lists of the most frequently named problems of 2336 Australian high school students. Confirmatory Factor Analysis was used to test and develop a measurement model developed from an extensive review of previous scales. The High School Stressors Scale emerged from the analytic process and measures nine school-related stressors. For researchers focusing on school-related problems and stressors among adolescents, the HSSS promises to be a very useful instrument. It has sound construct and predictive validity and adequate reliability, as demonstrated by the goodness-of-fit indices the squared multiple correlations.

The Melbourne Decision-Making Questionnaire : an instrument for measuring patterns for coping with decisional conflict

A study was conducted to examine the factorial validity of the Flinders Decision Making Questionnaire (Mann, 1982), a 31-item self-report inventory designed to measure tendencies to use three major coping patterns identified in the conflict theory of decision making (Janis and Mann, 1977): vigilance, hypervigilance, and defensive avoidance (procrastination, buck-passing, and rationalization). A sample of 2051 university students, comprising samples from Australia (n=262), New Zealand (n=260), the USA (n=475), Japan (n=359), Hong Kong (n=281) and Taiwan (n=414) was administered the DMQ. Factorial validity of the instrument was tested by confirmatory factor analysis with LISREL. Five different substantive models, representing different structural relationships between the decision-coping patterns had unsatisfactory fit to the data and could not be validated. A shortened instrument, containing 22 items, yielded a revised model comprising four identifiable factors-vigilance, hypervigilance, buck-passing, and procrastination. The revised model had adequate fit with data for each country sample and for the total sample, and was confirmed. It is recommended that the 22-item instrument, named the Melbourne DMQ, replace the Flinders DMQ for measurement of decision-coping patterns.

The development of a clinical learning environment scale

Within nursing, there is a strong demand for high-quality, cost-effective clinical education experiences that facilitate student learning in the clinical setting The clinical learning environment (CLE) is the interactive network of forces within the clinical setting that influence the students'clinical learning outcomes The identification of factors that characterize CLE could lead to strategies that foster the factors most predictive of desirable student learning outcomes and ameliorate those which may have a negative impact on student outcomes The CLE scale is a 23-item instrument with five subscales staff–student relationships, nurse manager commitment, patient relationships, interpersonal relationships, and student satisfaction These factors have strong substantive face validity and construct validity, as determined by confirmatory factor analysis Reliability coefficients range from high (0 85) to marginal (0 63) The CLE scale provides the educator with a valid and reliable instrument to evaluate affectively relevant factors in the CLE, direct resources to areas where improvement may be required, and nurture those areas functioning well It will assist in the application of resources in a cost-effective, efficient, productive manner, and will ensure that the clinical learning experience offers the nursing student the best possible learning outcomes