China 
Africa 
Explore chapters and articles related to this topic
Exploratory Factor Analysis
Published in Jhareswar Maiti, Multivariate Statistical Modeling in Engineering and Management, 2023
Exploratory factor analysis (EFA) is a data reduction technique similar to principal component analysis (PCA) with subtle differences in conceptualization, estimation to model adequacy testing including the purpose of the modeling. The development of factor analysis dates back to the classical works of Charles Spearman on association of two things (1904a) and general theory of intelligence (1904b). Although primarily developed in the domain of psychology, its applications spread across almost all disciplines where the issues involve the measurement of hidden things. The major purpose is to find out how many latent dimensions (or common factors) are involved in a group of measurements and what these dimensions are (Vincent, 1953). It exploits the covariance relationships of many observed variables (X) in search of the factors (F) and falls under covariance structure analysis. In this chapter, we discuss on the conceptual factor model with assumptions and certain useful results followed by three extraction methods. Then we discuss the model adequacy tests and different criteria to select the number of factors to be retained. Subsequently, factor rotation is described followed by the estimation of factor scores. Finally, a case study is presented which demonstrates the usefulness of EFA in solving real-world problems.
Dimension Reduction Breaking the Curse of Dimensionality
Published in Chong Ho Alex Yu, Data Mining and Exploration, 2022
Principal component analysis (PCA) is a popular dimension reduction technique, but it is commonly mixed up with exploratory factor analysis (EFA). This confusion can be partly attributed to the fact that in SPSS, the default extraction method for factor analysis is principal components. As a matter of fact, these are two different procedures, even though they yield almost identical results on many occasions. Principal components analysis aims to find the optimal way of collapsing many correlated variables into a small number of subsets to make the study more manageable. The subsets do not need to make any theoretical sense; they are for convenience only. In contrast, the purpose of exploratory factor analysis is to identify the underlying theoretical structure of diverse variables. If certain items are loaded into a subscale, then the items must be related to this construct both mathematically and conceptually. Additionally, the former takes all variances into account, while the latter extracts shared variances (Bandalos and Boehm-Kaufman 2009).
Personality and driver behaviour questionnaire: Correlational exploratory study
Published in Gianluca Dell’Acqua, Fred Wegman, Transport Infrastructure and Systems, 2017
J.F. Dourado, A.T. Pereira, V. Nogueira, A.M.C. Bastos Silva, A.J.M. Seco
The Exploratory Factor Analysis was performed using the principal components method, with varimax rotation (Pereira et al., 2013). The suitability of data for factor analysis was assessed with the Kaiser-Meyer-Oklin (KMO) and the Barlett’s Tests of Sphericity (Kline, 1994). The data was considered suitable for factorial analysis when KMO > 0.50 (Sharma, 1996) and the null hypothesis (H0) was rejected (p-value ≤ 0.05). To help establishing the correct number of factors to extract from the factorial analysis, the criteria used were: i) the Kaiser criteria, to retain factors with eigenvalue >1 (Kaiser, 1958); ii) Cattel Scree Plot criteria, which implies the retention of all components in the sharp descent part of the plot before the eigenvalues start to level off, where line changes slope (Cattel, 1966; Kline, 2000). The selection of the items for each factor consisted of retaining items that showed strong factor loadings. As other authors have done (Elal et al., 2000), items with factor loadings >0.30 were chosen.
Network orientation of logistics service providers: the construct, dimensionality and measurement scale
Published in International Journal of Logistics Research and Applications, 2020
Wojciech Czakon, Arkadiusz Kawa, Stephanie Scott
The study then assessed the construct validity (Bagozzi, Yi, and Phillips 1991) on a sample of 305 Polish logistics service managers. Through exploratory factor analysis, indicators with the highest factor loadings were identified. This process helped to reduce the complexity of the measurement scale and to obtain a better statistical adjustment of the factors. The reliability analysis by the Cronbach’s alpha was completed with the AVE indicator. The AMOS software was run for convergent and discriminant validity tests, through confirmatory factor analysis, and to assess the quality of the measurement model. A nomological validity assessment, using structural equation modeling on the proposed conceptual framework (Figure 1), tested a theoretical prediction that (1) increasing network orientation improves organisational capabilities, and (2) through improved organisational capabilities, a logistics firm’s performance increases.
Understanding key behavioral factors affecting road traffic citation and crash involvement of professional bus and passenger van drivers using a modified driver behavior questionnaire: an Indian perspective
Published in International Journal of Occupational Safety and Ergonomics, 2022
Vijaya Bandyopadhyaya, Ranja Bandyopadhyaya, Santanu Barman
Bus and passenger van drivers’ responses were analyzed separately. Initially, the internal consistency of the scale used was tested with reliability analysis using Cronbach’s α. The factor structure was first obtained by exploratory factor analysis (EFA) using the principle component analysis (PCA) method with varimax rotation with Kaiser normalization, the most commonly used method, with SPSS version 20.0. The factor structure obtained using EFA was confirmed using confirmatory factor analysis (CFA) with AMOS version 20.0. EFA was conducted with 70% of observations and CFA with the remaining 30% of the observations.
Analysis of risk management performance with innovative approach: a case study of China’s shipping companies
Published in Maritime Policy & Management, 2020
Hengbin Yin, Zhuo Chen, Yi Xiao, Su Wang
The analysis pertinent to the purpose of this study was conducted using the SPSS v. 21.0 statistical package. First, a frequency analysis was performed to investigate the subjects’ general characteristics. Second, an exploratory factor analysis was conducted to verify the validity of the measurement tools and the Cronbach’s α was calculated for reliability verification. Third, a descriptive statistical analysis was performed, providing the mean and standard deviation of major variables, and a correlation analysis was conducted. Fourth, a three-step regression analysis was conducted for hypothesis testing.