Search

Dimension reduction, by means of Principal Component Analysis (PCA), is often employed to obtain a reduced set of components preserving the largest possible part of the total variance of the observed variables. Several methodologies have been proposed either to improve the interpretation of PCA results (e.g., by means of orthogonal, oblique rotations, shrinkage methods), or to model oblique components or factors with a hierarchical structure, such as in Bi-factor and High-Order Factor analyses. In this paper, we propose a new methodology, called Hierarchical Disjoint Principal Component Analysis (HierDPCA), that aims at building a hierarchy of disjoint principal components of maximum variance associated with disjoint groups of observed variables, from Q up to a unique, general one. HierDPCA also allows choosing the type of the relationship among disjoint principal components of two sequential levels, from the lowest upwards, by testing the component correlation per level and changing from a reflective to a formative approach when this correlation turns out to be not statistically significant. The methodology is formulated in a semi-parametric least-squares framework and a coordinate descent algorithm is proposed to estimate the model parameters. A simulation study and two real applications are illustrated to highlight the empirical properties of the proposed methodology. [ABSTRACT FROM AUTHOR]

Abstract The choice of statistical data analysis should be guided by a critical analysis that supports the theoretical relationship between the construct and its indicators. This theoretical article reviews the three main existing psychometric paradigms and their proposals for explaining the relationship between indicators and their constructs. The discussion begins with the standard paradigm that guides the construction and analysis of data in psychology, reflective model. Then, a description of the formative models is performed and finally the Network Analysis as an alternative. The definitions, consequences, and limitations of the use of each measurement model are presented such as a reflection on making decisions about which data generation mechanisms are more appropriate.

The choice of statistical data analysis should be guided by a critical analysis that supports the theoretical relationship between the construct and its indicators. This theoretical article reviews the three main existing psychometric paradigms and their proposals for explaining the relationship between indicators and their constructs. The discussion begins with the standard paradigm that guides the construction and analysis of data in psychology, reflective model. Then, a description of the formative models is performed and finally the Network Analysis as an alternative. The definitions, consequences, and limitations of the use of each measurement model are presented such as a reflection on making decisions about which data generation mechanisms are more appropriate. [ABSTRACT FROM AUTHOR]

Research on the effects of adversity has led to mounting interest in examining the differential impact of adversity as a function of its timing and type. The current study examines whether the effects of different types (i.e., physical, sexual, and emotional abuse) and timing (i.e., early, middle childhood, adolescence, or adulthood) of adversity on maternal mental and physical health outcomes in pregnancy, are best accounted for by a cumulative model or independent effects model. Women from a prospective pregnancy cohort (N =3,362) reported retrospectively on their experiences of adversity (i.e., physical, sexual, and emotional abuse) in early childhood (0–5 years], middle childhood (6–12 years], adolescence (13–18 years], and adulthood (19+ years]. Measures of overall health, stress, anxiety, and depression were gathered in pregnancy. Results showed that a cumulative formative latent model was selected as more parsimonious than a direct effects model. Results also supported a model where the strength of the effect of adversity did not vary across abuse timing or type. Thus, cumulative adversity resulted in greater physical and mental health difficulties. In conclusion, cumulative adversity is a more parsimonious predictor of maternal physical and mental health outcomes than adversity at any one specific adversity timing or subtype. [ABSTRACT FROM AUTHOR]

Despite substantial evidence for the link between an individual's intelligence and successful life outcomes, questions about what defines intelligence have remained the focus of heated dispute. The most common approach to understanding intelligence has been to investigate what performance on tests of intellect is and is not associated with. This psychometric approach, based on correlations and factor analysis is deficient. In this review, we aim to substantiate why classic psychometrics which focus on between-person accounts will necessarily provide a limited account of intelligence until theoretical considerations of within-person accounts are incorporated. First, we consider the impact of entrenched psychometric presumptions that support the status quo and impede alternative views. Second, we review the importance of process-theories, which are critical for any serious attempt to build a within-person account of intelligence. Third, features of dynamic tasks are reviewed, and we outline how static tasks can be modified to target within-person processes. Finally, we explain how multilevel models are conceptually and psychometrically well-suited to building and testing within-individual notions of intelligence, which at its core, we argue is cognitive flexibility. We conclude by describing an application of these ideas in the context of microworlds as a case study. [ABSTRACT FROM AUTHOR]

BACKGROUND: From funerary rituals to amphitheater spectacles, gladiators played social roles that changed along the VII century B.C. and the V century A.D. in Ancient Rome. OBJECTIVE: Historically and educationally focused, this text aims to analyze training schools (Ludi gladiatorial) as a formative model for the ways of fighting, arming, physical preparation, the structure used in spectacles, fighting, and living rules, and ultimately, the professionals inside roman combats. METHODS: Therefore, I undertook bibliographical research through the works of Veyne (2009, 2015), Poehler (2013), Garraffoni (2008), and the Site of the Pompeii Archeological Park, among other works that could help me understand gladiators' training and trajectory to meet their contemporary roman society needs. Though considered part of infamous life, gladiators were requested to participate in commemorative religious and military rituals and in entertainment held at amphitheaters, thus, institutionalizing the places destined for training and the troupe (familiae gladiatoriae) to satisfy the roman and non-roman social, cultural, and administrative contexts. RESULTS: Training schools were the institutions that served for physical preparation, technique and coexistence between different subjects (lanista, trainers, fighters and fans) and institutionalized the fighting profession through presentations in local cities and in troupes in the surrounding cities and spread images of fighters, athletes and gladiators evidencing the fame of these representatives of the infamy class; however, they also reveal the formative model to youngsters by referring to the courage to "go down fighting", bordering on heroism, whether for military and political purposes or for disciplinary and punitive ones. CONCLUSION: In that institution, the gladiator's training model was rooted by funerary rituals (munus), religious celebrations, war conquests and punishment for those considered "outside the Roman law", let us not forget to mention that this model was an example of virility, strength and fame, luring young ones and volunteers from Ancient Rome and institutionalizing gladiatorial combat throughout the whole Roman imperial domain. [ABSTRACT FROM AUTHOR]

Despite substantial evidence for the link between an individual’s intelligence and successful life outcomes, questions about what defines intelligence have remained the focus of heated dispute. The most common approach to understanding intelligence has been to investigate what performance on tests of intellect is and is not associated with. This psychometric approach, based on correlations and factor analysis is deficient. In this review, we aim to substantiate why classic psychometrics which focus on between-person accounts will necessarily provide a limited account of intelligence until theoretical considerations of within-person accounts are incorporated. First, we consider the impact of entrenched psychometric presumptions that support the status quo and impede alternative views. Second, we review the importance of process-theories, which are critical for any serious attempt to build a within-person account of intelligence. Third, features of dynamic tasks are reviewed, and we outline how static tasks can be modified to target within-person processes. Finally, we explain how multilevel models are conceptually and psychometrically well-suited to building and testing within-individual notions of intelligence, which at its core, we argue is cognitive flexibility. We conclude by describing an application of these ideas in the context of microworlds as a case study.

Within evolutionary biology, life-history theory is used to explain cross-species differences in allocation strategies regarding reproduction, maturation, and survival. Behavioral scientists have recently begun to conceptualize such strategies as a within-species individual characteristic that is predictive of behavior. Although life history theory provides an important framework for behavioral scientists, the psychometric approach to life-history strategy measurement—as operationalized by K-factors—involves conceptual entanglements. We argue that current psychometric approaches attempting to identify K-factors are based on an unwarranted conflation of functional descriptions and proximate mechanisms—a conceptual mix-up that may generate unviable hypotheses and invites misinterpretation of empirical findings. The assumptions underlying generic psychometric methodology do not allow measurement of functionally defined variables; rather these methods are confined to Mayr’s proximate causal realm. We therefore conclude that K-factor scales lack validity, and that life history strategy cannot be identified with psychometrics as usual. To align theory with methodology, suggestions for alternative methods and new avenues are proposed. [ABSTRACT FROM AUTHOR]

The values associated with a tourism product, such as a hospitality facility, an attraction, or a destination, are intangible for independent evaluation. Meanwhile, tangible features of a tourism product can be used as predictor variables in causal research to assess the values as dependent variables. In the meanwhile, the multidimensional values may function independently from each other. Under this condition, consideration should be made in designing the values as a formative model and using partial least squares structural equation modeling to illustrate its relationships with the predictor variables. This paper uses the Taipei Zoo as a case study by conducting a survey of 472 visitors. The results demonstrate how different conceptual model designs can affect the statistical results and ultimate decision-making concerned with the value design of a tourism product. [ABSTRACT FROM AUTHOR]

Introduction: The examination of cognitive reserve (CR) literature reveals a lack of consensus regarding conceptualization and pervasive problems with its measurement. This study aimed at examining the conceptual nature of CR through the analysis of reflective and formative models using eight proxies commonly employed in the CR literature. We hypothesized that all CR proxies would significantly contribute to a one-factor reflective model and that educational and occupational attainment would produce the strongest loadings on a single CR factor.Method: The sample consisted of 149 participants (82 male/67 female), with 18.1 average years of education and ages of 45–99 years. Participants were assessed for eight proxies of CR (parent socioeconomic status, intellectual functioning, level of education, health literacy, occupational prestige, life leisure activities, physical activities, and spiritual and religious activities). Primary statistical analyses consisted of confirmatory factor analysis (CFA) to test reflective models and structural equation modeling (SEM) to evaluate multiple indicators multiple causes (MIMIC) models.Results: CFA did not produce compelling support for a unitary CR construct when using all eight of our CR proxy variables in a reflective model but fairly cogent evidence for a one-factor model with four variable proxies. A second three-factor reflective model based upon an exploratory principal components analysis of the eight proxies was tested using CFA. Though all eight indicators significantly loaded on their assigned factors, evidence in support of overall model fit was mixed. Based upon the results involving the three-factor reflective model, two alternative formative models were developed and evaluated. While some support was obtained for both, the model in which the formative influences were specified as latent variables appeared to best account for the contributions of all eight proxies to the CR construct.Conclusion: While the findings provide partial support for our hypothesis regarding CR as a one-dimensional reflective construct, the results strongly suggest that the construct is more complex than what can be captured in a reflective model alone. There is a need for theory to better identify and differentiate formative from reflective indicators and to articulate the mechanisms by which CR develops and operates. [ABSTRACT FROM PUBLISHER]

A number of personality frameworks assume traits describe central tendencies of action--for instance, calling someone assertive indicates they have a tendency to perform assertive actions. But what makes it appropriate to characterize an action by terms like assertive, kind, or honest? We propose that actions are characterized by such terms in large part by having expected effects on the environment which match particular conceptual templates. In the present studies, we attempt to better identify the expected effect dimensions perceivers seem to utilize to make action characterizations related to the Big Five and HEXACO personality dimensions. To do so, a set of 150 situation-action scenarios were generated from actions suggestive of conscientiousness-related characteristics (Study 1), and of characteristics in other HEXACO domains (Study 2). Participants then characterized each action on a range of bipolar dimensions (e.g., assertive vs. submissive). A separate group of raters coded the expected effects of performing these actions on 21 different outcomes (e.g., effort expenditure; achievement of career goals). Action characterizations were highly predicted by expected effect dimensions in ways that matched provisional hypotheses and were consistent across studies. Furthermore, actions characterizations tended to be highly diagnostic of self-reported individual differences in the same characteristics. We discuss implications for a range of phenomena, such as understanding the relations between behaviors and traits, integrating trait models and decision-making models, and understanding the effect of situational features on personality traits. [ABSTRACT FROM AUTHOR]

This article reflects on the family as a value, and on the worth of the family. The piece is divided into six themes with an introduction and a conclusion. The six nuclei are essential, simple and clear considerations of several fundamental coordinates of the reality of the family today, recalling both the values and the reality, the present and prospects for the family, anthropological aspects, as well at legal orientations of the Lobbies which really obscure the rights and duties of those directly involved, both parents and children. The educational and formative attention runs throughout the whole reflection, pointing out some practical issues that can be the object of working with sound humanistic discernment. [ABSTRACT FROM AUTHOR]

Purpose – The objective of this study is to develop and test a specific conceptualization and modeling specification of web site service quality and the influence web site service quality has on e-service provider loyalty and word-of-mouth. Design/methodology/approach – Data from an online survey of consumers across two studies were obtained. Findings – The results support the conceptualization and operationalization of web site service quality and its effects on word of mouth and loyalty. Research limitations/implications – Implications for the measurement of consumer perceptions of web site service quality are highlighted for researchers, and directions for future research are discussed. Practical implications – E-service providers should carefully consider customer experiences on their web sites, and their management efforts to ensure they design and deliver the appropriate levels of service quality derived through the specific components of web site service quality. Originality/value – The findings are of value to e-service practitioners managing web site service quality and offer empirical support to its formative conceptualization. The focus here on formative modeling will stimulate discussion among e-service researchers and practitioners, which could result in richer e-service quality measures leading to improved marketing decision making and web site quality management. [ABSTRACT FROM AUTHOR]

This paper aims to discuss the different results of factor analysis used to test construct validity and reliability in SERVQUAL studies, considering the fact that our study measuring service quality of Turk Eximbank, export credit agency in Turkey, resulted in more different -but not unusual- findings than those proposed by the designers of the model. These controversial results observed in both our study and various service quality investigations directed us to look more closely into this issue. As Collier and Bienstock (2009) claimed, service quality can be well represented as a formative construct although it is treated as a reflective construct by the traditional approach. It makes us consider again on the SERVQUAL model specification since the results of factor analysis in service quality studies are diverse from originally designed construct proposed by Parasuraman et al. (1985/1988). Hence, this paper entails a discussion of factor analysis an d its results in SERVQUAL studies. Perhaps, it might be beneficial to ruminate whether SERVQUAL model is specified by a formative construct or not. Accordingly, alternative validity tests other than, or contributing to, factor analysis can be more appropriate for SERVQUAL model. [ABSTRACT FROM AUTHOR]

Copyright of Portuguese Journal of Marketing / Revista Portuguesa de Marketing is the property of Revista Portuguesa de Marketing / Portuguese Journal of Marketing and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Research on the effects of adversity has led to mounting interest in examining the differential impact of adversity as a function of its timing and type. The current study examines whether the effects of different types (i.e., physical, sexual, and emotional abuse) and timing (i.e., early, middle childhood, adolescence, or adulthood) of adversity on maternal mental and physical health outcomes in pregnancy, are best accounted for by a cumulative model or independent effects model. Women from a prospective pregnancy cohort (N =3,362) reported retrospectively on their experiences of adversity (i.e., physical, sexual, and emotional abuse) in early childhood (0-5 years], middle childhood (6-12 years], adolescence (13-18 years], and adulthood (19+ years]. Measures of overall health, stress, anxiety, and depression were gathered in pregnancy. Results showed that a cumulative formative latent model was selected as more parsimonious than a direct effects model. Results also supported a model where the strength of the effect of adversity did not vary across abuse timing or type. Thus, cumulative adversity resulted in greater physical and mental health difficulties. In conclusion, cumulative adversity is a more parsimonious predictor of maternal physical and mental health outcomes than adversity at any one specific adversity timing or subtype.