scholarly journals Contemporary advanced statistical methods for the science of educational research: Principal components analysis versus L’ analysee factorielle des correspondances

2019 ◽  
Vol 6 (1) ◽  
pp. 246-255
Author(s):  
Paschalina Ntotsi ◽  
Sofia D Anastasiadou
Keyword(s):  
Principal Components Analysis ◽  
Statistical Methods ◽  
Principal Components ◽  
Tertiary Education ◽  
Likert Scale ◽  
Clustering Methods ◽  
Technological Education ◽  
Components Analysis ◽  
Students Satisfaction ◽  
Item Scale

The current paper analyses two different statistical techniques: i.e., principal components analysis (PCA) and correspondence analysis (L’analysee factorielle des correspondances) (AFC). A survey was carried out using a structured questionnaire for a sample of 135 nurses which studied in the School of Pedagogical and Technological Education (ASPETE) in Greece. Tangibility, Reliability, Responsiveness, Assurance, Empathy and Associability subscales are related to Qualitative Services ASPETE offers. These subscales were measured by 24 items, rated on a seven-point Likert scale. The study focuses on the presentation of the two main types of clustering methods, PCA and AFC. Lee and Lin’s model contains a one-item scale developed to measure overall service quality and a one-item scale for customer satisfaction. The assessment of the students’ satisfaction degree is evaluated based on a seven-step on the Likert scale statement, investigating the extent that the respondents are satisfied from the experience they had with the specific tertiary education organisation (CSF). Keywords: Advanced, statistical, methods, AFC, PCA

Comparison of multivariate pattering methods in group/cluster identification regarding the science of educational research: Implicative Statistical Analysis vs. L’ Analysee Factorielle des Correspondances

2019 ◽  
Vol 6 (1) ◽  
pp. 238-245
Author(s):  
Sofia D Anastasiadou
Keyword(s):  
Statistical Analysis ◽  
Principal Components Analysis ◽  
Educational Research ◽  
Principal Components ◽  
Quantitative Research ◽  
Self Control ◽  
Clustering Methods ◽  
Technological Education ◽  
Cluster Identification ◽  
Components Analysis

Nowadays, there is a substantial improvement and utilisation of pattering methods in the science of educational research, a comparison of multivariate methods in group/cluster identification in the scientific domain of quantitative research has not been carried out. This study analyses two different statistical techniques: i.e., the principal components analysis (PCA) and the implicative statistical analysis (ASI). A survey was carried out using a structured questionnaire for a sample of 135 nurses which studied in the School of Pedagogical and Technological Education in order to be qualified in respect The study focuses on the presentation of the two main types of clustering methods, της ASI and L’ Analysee Factorielle des Correspondances (AFC). AFC’s results made it evident that Emotionality, Self-control, Sociability, General items of EI constructs are shaped attitudes and reveal the latent dimension of respondents psychological attributes related to EI conceptual constructs. Keywords: L’ Analysee Factorielle des Correspondances, principal components analysis, implicative statistical analysis, research.

Locomotor morphometry of the Pachydactylus radiation of lizards (Gekkota: Gekkonidae): a phylogenetically and ecologically informed analysis

2005 ◽  
Vol 83 (12) ◽  
pp. 1511-1524 ◽  
Author(s):  
Megan K Johnson ◽  
Anthony P Russell ◽  
Aaron M Bauer
Keyword(s):  
Principal Components Analysis ◽  
Statistical Methods ◽  
Principal Components ◽  
General Pattern ◽  
Diverse Group ◽  
Promising Candidate ◽  
Terrestrial Species ◽  
Terrestrial Locomotion ◽  
Components Analysis ◽  
Limb Proportions

The Pachydactylus radiation comprises a diverse group of African gekkonids that exploit a variety of microhabitats and exhibit both climbing and terrestrial locomotion. The phylogeny of this radiation is well supported, making it a promising candidate for the investigation of relationships between limb proportions, ecology, and behaviour. Skeletal and external measurements were recorded for an array of taxa and analyzed using principal components analysis (PCA). The results of the PCAs were further analyzed using phylogenetic statistical methods to ascertain whether climbing and terrestrial species and (or) clades within the radiation differed significantly from each other in limb proportions. Phylogenetically based comparisons revealed that although there is some differentiation between climbing and terrestrial species, this is not a general pattern but is primarily attributable to certain species and clades within the radiation that differ considerably from other members of the group. The results indicate that Chondrodactylus angulifer Peters, 1870 possesses shortened distal phalanges and that Pachydactylus rangei (Andersson, 1908), P. austeni Hewitt, 1923, and the Rhoptropus clade (particularly R. afer Peters, 1869) possess elongated limbs relative to the rest of the radiation. These differences correlate with aspects of the lifestyles of these species, such as increased terrestriality, a reduction or loss of the subdigital adhesive apparatus, digging behaviour, and a transition to diurnality.

Personality in Domestic Cats

2007 ◽  
Vol 100 (1) ◽  
pp. 27-29 ◽  
Author(s):  
Christina M. Lee ◽  
Joseph J. Ryan ◽  
David S. Kreiner
Keyword(s):  
Principal Components Analysis ◽  
Negative Correlation ◽  
Principal Components ◽  
Likert Scale ◽  
Domestic Cats ◽  
Varimax Rotation ◽  
Personality Ratings ◽  
Weak Negative Correlation ◽  
Components Analysis

Personality ratings of 196 cats were made by their owners using a 5-point Likert scale anchored by 1: not at all and 5: a great deal with 12 items: timid, friendly, curious, sociable, obedient, clever, protective, active, independent, aggressive, bad-tempered, and emotional. A principal components analysis with varimax rotation identified three intepretable components. Component I had high loadings by active, clever, curious, and sociable. Component II had high loadings by emotional, friendly, and protective, Component III by aggressive and bad-tempered, and Component IV by timid. Sex was not associated with any component, but age showed a weak negative correlation with Component I. Older animals were rated less social and curious than younger animals.

scholarly journals Basic Multivariate Statistical Methods for Environmental Monitoring Data Mining: Introductory Course for Master Students

2020 ◽  
Vol 25 (1-2) ◽  
pp. 35-56
Author(s):  
Vasil Simeonov
Keyword(s):  
Environmental Monitoring ◽  
Principal Components Analysis ◽  
Statistical Methods ◽  
Principal Components ◽  
Monitoring Data ◽  
Multivariate Statistical Methods ◽  
Multivariate Statistical ◽  
Introductory Course ◽  
Components Analysis ◽  
Environmental Monitoring Data

Abstract The present introductory course of lectures summarizes the principles and algorithms of several widely used multivariate statistical methods: cluster analysis, principal components analysis, principal components regression, N-way principal components analysis, partial least squares regression and self-organizing maps with respect to their possible application in intelligent analysis, classification, modelling and interpretation to environmental monitoring data. The target group of possible users is master program students (environmental chemistry, analytical chemistry, environmental modelling and risk assessment etc.).

An Examination of the Reliability and Validity of the Breskin Rigidity Scale

1996 ◽  
Vol 82 (1) ◽  
pp. 195-198 ◽  
Author(s):  
John Maltby ◽  
Christopher Alan Lewis
Keyword(s):  
Principal Components Analysis ◽  
Principal Components ◽  
Reliability And Validity ◽  
Attitude Inventory ◽  
Two Samples ◽  
Components Analysis ◽  
Item Scale ◽  
Reliability Coefficients

The present study examined the reliability and validity of the 15-item Breskin Rigidity Scale. Two samples of UK adults ( n = 216; n = 277) completed the rigidity scale alongside the Wilson-Patterson Attitude Inventory and the trait measure of the Sandler-Hazari Obsessionality Inventory. Satisfactory reliability coefficients were obtained for the rigidity scale in both samples. A principal components analysis of the 15-item rigidity scale showed that 13 items loaded on the first component and 2 on a second component. These items were removed and the reliability and validity estimates were recalculated. Comparisons were drawn between the original and amended versions. The amended 13-item scale shows an improved reliability and higher association with other measures of rigidity and rigid character.

scholarly journals The Use of a Five Factor Model in Equine Personality Research

2013 ◽  
Vol 26 (4) ◽  
Author(s):  
Rachel E. Kristiansen ◽  
Stan A. Kuczaj II
Keyword(s):  
Principal Components Analysis ◽  
Principal Components ◽  
Factor Model ◽  
Five Factor Model ◽  
Original Study ◽  
Good Reliability ◽  
Personality Research ◽  
Components Analysis ◽  
Don't Know Responses ◽  
Item Scale

In order to test the validity of a Five Factor Model of personality on horses, a questionnaire was replicated from a previous study, with an added option of don’t know to the traditional 5-point Likert scale. Raters responded to seventeen items of the 60-item scale with don’t know responses greater than 10% of the time and these seventeen items were subsequently removed from the study. A Principal Components Analysis was used with the remaining items, resulting in eight factors: Neuroticism, Active, Conscientiousness, Agreeableness, Openness, Social Extraversion, Temperamental, and Disciplined. These components correspond well to the five components extracted in the original study, indicating good reliability of the scale. However, 17 items from the original questionnaire were deemed irrelevant by raters, indicating a threat to validity. Though the remaining items were able to be used in analyses, further studies should examine if these are in fact the most effective items to use in the investigation of equine personality.//

A Proposed Scale to Measure Beliefs about the Consequences of Employing Homosexuals

1988 ◽  
Vol 63 (2) ◽  
pp. 547-551 ◽  
Author(s):  
Fabius P. O'Brien ◽  
Michael J. Vest
Keyword(s):  
Principal Components Analysis ◽  
Predictive Validity ◽  
Principal Components ◽  
Internal Consistency Reliability ◽  
Service Organizations ◽  
Strong Correlations ◽  
Southeastern Us ◽  
Discrimination In Employment ◽  
Components Analysis ◽  
Item Scale

Preliminary development of a scale to measure beliefs about the consequences of employing homosexuals is described. From responses of 182 managers from Southeastern US manufacturing, government, and service organizations principal components analysis yielded a single 13-item scale which exhibited strong item-total correlations and internal consistency reliability; Cronbach's alpha was .95. Evidence of predictive validity was provided by relatively strong correlations between beliefs about the consequences of hiring homosexuals and two indicators of discrimination in employment.

Sign in / Sign up

Export Citation Format

Share Document