Marina Doucerain, Norman Segalowitz, and Andrew G. Ryder
This article discusses the importance of clear and precise conceptualizations of acculturation as well as the need for consistencies in definition, operationalization, and measurement. More specifically, it argues for an expanded acculturation research toolkit that does not rely too heavily on self-report acculturation scales. The article begins with an overview of the state of affairs with respect to acculturation conceptualizations and methods, paying particular attention to the unidimensional, bidimensional, and multidimensional frameworks of psychological acculturation. It then considers ways in which commonly used definitions and methods of acculturation can be used more intelligently. It also describes alternative methods for researchers interested in moving beyond self-report rating scales, a tiered approach to acculturation research, and method-specific health considerations. Finally, it offers some recommendations aimed at helping the field of acculturation and health research move forward.
Elaine D. Pulakos, Rose A. Mueller-Hanson, and Johnathan K. Nelson
In this chapter, we examine issues relevant to incorporating trainability and adaptive performance into selection research. We adopt the definition of adaptive performance suggested by Pulakos et al. (2000) that specified eight dimensions defining this construct. One of these dimensions, leaning new tasks, technology, and procedures, was used to define trainability. We then examine recent models of adaptive performance and training to identify likely predictors of adaptability and trainability and propose a method for determining when and where these criteria should be included and explicitly predicted in selection research. We examine the pros and cons associated with different criterion measures and recommend that typical rating measures potentially supplemented by lower-fidelity work sample measures be incorporated in selection research. Finally, we discuss gaps in the current literature and recommend areas for future research.
Anthony J. Onwuegbuzie and John H. Hitchcock
Because of the complexity of mixed analysis, several authors recently have written methodological works that provide either an introductory- or intermediate-level guide to conducting such analyses. Although all of these works have been useful for beginning and emergent mixed researchers, what is lacking are works that describe and illustrate advanced mixed analysis approaches. Thus, in this chapter, we provide a compendium of advanced mixed analysis approaches. In particular, we outline three advanced quantitative-dominant crossover mixed analyses, three advanced qualitative-dominant crossover mixed analyses, and one advanced equal-status crossover mixed analysis. Most of these advanced crossover mixed analyses previously have not been described as a mixed analysis technique in any published work, illustrating the significance and innovation of our chapter.
Larry R. Price
A brief history of imaging neuroscience is presented followed by an introduction to data acquisition using positron emission tomography (PET)and functional magnetic resonance imaging (fMRI). Next, statistical parametric mapping is introduced in conjunction with random field theory as being fundamental to identifying sites of neural activation. The general linear model is discussed as being foundational for all imaging analyses. Finally, methods for studying functional and effective connectivity such as eigenimage analysis, partial least squares, multivariate autoregressive models, structural equation models, and dynamic causal models are reviewed in light of deterministic and stochastic analytic approaches.
The chapter gives an instruction to event history analysis. The central goals are first to justify why what perhaps must be considered an unusual modeling approach is needed and next to explicate in some detail what the key ideas from probability theory are and how these ideas solve the problems that arise when using more standard techniques such as regression analysis for continuous dependent variables or logit analysis for binary dependent variables. Elaborations for how to take account of measured variables are given. It elaborates on what the dependent variable is in event history analysis, on the framework for repeated event processes, multi-state processes, and continuous-state space processes.
Stephen W. Gilliland and Dirk D. Steiner
Applicant reactions to selection and assessment have developed into a theoretically grounded and productive body of research over the past 20 years. Organizational justice theories provide a valuable foundation for much of this research, but important models have also been developed from test motivation and social psychological perspectives. Research indicates that applicant reactions are strongly related to prehire attitudes and applicant self-perceptions, but not related to most behaviors. Research has also demonstrated substantial consistency in applicant reactions across gender, race, and cultures. Generally, applicants react most favorably toward work sample tests and interviews, negatively toward graphology and honesty tests, and moderately toward cognitive ability tests, biodata, and personality inventories. We conclude by highlighting a number of areas for future research, suggesting that with broader perspectives applicant reactions research can continue to be as productive as it has been in the past.
Multimethod and mixed methods are well suited to prevention research in global health; however, their application has not yet been adequately discussed or demonstrated. This chapter illustrates key opportunities and challenges through focusing on using multimethod and mixed methods for investigating prevention involving migration. It summarizes one large study focused on labor migrants and HIV/AIDS risk and protection to illustrate how innovative strategies combining different forms of knowledge in multimethod and mixed methods can generate more robust and useful findings. Multimethod and mixed methods in prevention research in global health should strategically utilize multiple study elements (investigators, theories, methods, and data) that are most responsive to the central research problems and questions, through existing and new synergies, so as to most appropriately address the key preventive intervention characteristics and contribute to the overall completeness of the knowledge.
Jonathan P. Schwartz, Michael Waldo, and Margaret Schwartz Moravec
Assessment is critical to understanding the outcomes and processes inherent in group counseling. However, assessment in groups is often ignored or attempted utilizing measures with poor psychometrics. The purpose of this chapter is to explore the various purposes of assessment in group counseling, followed by a summary of different types of assessment that may be used. Strengths and weaknesses of various assessments and research designs will also be discussed, along with implications for best practice.
Randall T. Salekin, Matthew A. Jarrett, and Elizabeth W. Adams
This chapter focuses on a range of measurement issues including traditional test development procedures and model testing. Themes of the chapter include the importance of integrating research disciplines and nosologies with regard to test development, the need for test development and measurement in the context of idiographic and nomothetic treatment designs, the need for change sensitive measures, and the possible integration of both idiographic and nomothetic treatment designs. Finally, the chapter emphasizes the importance of exploring novel areas of measurement (e.g., biological functioning, contextual factors) using emerging technologies. It is thought that innovative test development and use will lead to improved intervention model testing.
Todd A. Baker and Deborah L. Gebhardt
The world of work has many arduous jobs that require the worker to possess greater levels of physical ability than found in the normal population. This chapter provides an overview of the underlying physiological principles associated with physical performance and methods to assess arduous jobs in the workplace. It includes an overview of test development and validation of physical tests and litigation related to their use in job selection and retention. The benefits of physical testing and the methods for reducing adverse impact are highlighted.
Assessment of Voluntary Turnover in Organizations: Answering the Questions of Why, Who, and How Much
Sang Eun Woo and Carl P. Maertz
This chapter reviews recent research related to the assessment of three questions: why people quit, who is more or less likely to quit, and how much do instances of quitting cost the organization. Assessments of the “why” and the “who” questions together inform the assessment of turnover functionality (i.e., the “how much” question). Furthermore, the “why” assessment indicates which specific interventions are most likely to succeed with employees in the organization, and the “who” assessment helps determine for which employees the interventions should be implemented. Therefore, answers to these three interrelated questions ultimately converge as the critical inputs for practitioners who seek to determine whether organizations should (and are able) to intervene effectively to influence specific turnover instances. Following discussions of the complexity involved in these assessments, a person-centered, clustering approach was recommended as an effective strategy of accounting for the uniqueness of turnover incidents and prevention strategies.
Michael D. Mumford, Jamie D. Barrett, and Kimberly S. Hester
Background data, or biodata, measures are widely applied in personnel selection. In the present effort, it is argued that background data measures reflect the recall of differential experiential, or case-based, knowledge. The techniques for developing and scaling background data measures are described and evidence bearing on the reliability and validity of these measures is discussed. Critical contingencies bearing on the application of these measures in personnel selection are described. Potential directions for future research are examined along with issues bearing on the application of background data measures in personnel selection.
David Kaplan and Sarah Depaoli
This chapter provides a general overview of Bayesian statistical methods. Topics include the notion of probability from a Bayesian perspective, Bayesian inference and hypothesis testing, and Bayesian computation. Three examples are provided to demonstrate the utility of Bayesian methods: simple linear regression, multilevel regression, and confirmatory factor analysis. Throughout the chapter, references are made to the epistemological differences between Bayesian theory and classical (frequentist) theory.
Andrew Christopher, Pam Marek, and Kevin Zabel
We provide insights into writing for publication in peer-reviewed journals. We stress the need to make scientific writing part of one’s daily routine. We also include counsel on writing manuscripts, including information on how to write each of the major sections of an empirical manuscript. Finally, we conclude with insights into navigating the peer-review process to get a manuscript accepted for publication.
Carol M. Woods
Categorical methods refer to statistical procedures for analyzing data when the outcome variable is binary, nominal, or ordinal, according to Stevens’ (1946) popular taxonomy of scale types. Myriad such methods exist. This chapter is focused specifically on methods for a single binary or ordinal outcome. The methods addressed are: the Pearson χ2 statistic, the mean score statistic, the correlation statistic, midrank scores, odds ratios, differences between proportions, the γ-family measures of ordinal association, the Mantel-Haenszel test, the Mantel test, average conditional effect sizes, the Breslow-Day-Tarone test, binary logistic regression, and the proportional odds logistic regression model.
Robert K. Yin
Mixed methods research (MMR) has struggled to bridge paradigms that differ starkly on two central concepts—causality and generalizability. This chapter depicts the differences and the efforts made to bridge them. However, MMR can go beyond sheer bridging and strive to create an integrated craft. In particular, both the qualitative and quantitative camps have left four procedures underspecified: triangulating, examining plausible rival explanations, analyzing mixed methods data, and making analytic generalizations. For instance, no criteria exist to show whether a study has sufficiently triangulated or examined rivals. If MMR developed operational benchmarks, its studies could then use all four procedures in a compelling manner. In so doing, MMR might become a truly blended craft—not just one that bridges existing paradigms.
This chapter reviews the problem of finding the 'right level' of causal explanation in psychiatry. This is not a purely philosophical problem, but one that frequently arises in practice for psychiatrists. For most scientists, experiment is the crucial test of a causal hypothesis: for X to cause Y is for intervention on X to be reflected in a change in the value of Y. But this kind of approach cannot tell us the right "level" at which to specify the causes of a particular outcome. The natural idea is the right level is one that specifies the 'mechanism' by which Y is produced. But the notion of a "mechanism" in psychiatry is obviously problematic. This chapter attempts to locate the sources of the difficulty here, looking at both a priori and empirical views as to when a "mechanism" has been correctly specified.
Simon L. Albrecht
This chapter addresses the issue of the creation and maintenance of a climate for employee engagement in organizations. Employee engagement has been receiving increased attention in the past 5 to 10 years and is increasingly recognized as a crucial source of competitive advantage. This chapter offers a definition of “a climate for engagement,” locates climate for engagement in a taxonomy of “climates for something,” offers items by which to measure a climate for engagement, and offers an integrated model showing how climate for engagement mediates the influence of antecedents (e.g., organizational leadership, culture, human resource management systems, organizational climate) on psychological-motivational factors (e.g., need satisfaction, engagement) and downstream attitudes, behaviors and organizational level effectiveness outcomes.
Frederick T. L. Leong and Zornitsa Kalibatseva
The increasing population of racial and ethnic minorities calls for more attention to cultural diversity in clinical research. This chapter starts with a definition of culture and a brief discussion of the two parallel approaches to culture within psychology, cross-cultural psychology and racial and ethnic minority psychology. Subsequently, the chapter reviews cross-cultural issues in clinical research along two dimensions, namely the methodological strategies used to undertake clinical research and the methodological challenges encountered in clinical research. The reviewed methodological strategies include clinical case studies, analogue and simulation studies, randomized clinical trials, archival research and secondary data analysis, culture-specific approaches to treatment research, and meta-analysis. Lastly, the chapter discusses five methodological challenges for clinical research with culturally diverse populations, such as sample selection, measurement equivalence, race and ethnicity as demographic versus psychological variables, confounds and intersectionality, and differential research infrastructure.
André A. Rupp
In this chapter I first describe core terminology, notation, and related readings for certain core clustering and classification techniques. I then discuss the theoretical underpinnings and practical applications of nonparametric techniques that do not require distributional assumptions on outcome variables followed by parametric/model-based techniques that do require such assumptions. In the former set, I specifically discuss hierarchical clustering techniques and K-means clustering techniques. In the latter set I specifically discuss univariate and multivariate finite mixture models, unrestricted latent class models, and restricted latent class models. I further show how so-called diagnostic classification models are a particularly useful class of restricted latent class models for calibration and scaling purposes in educational and psychological measurement.