This text aims to help researchers and students to understand the purpose and presentation of multivariate statistical techniques. The most commonly used techniques are described in detail, such as multiple regression and correlation and path analysis
Panel data - information gathered from the same individuals or units at several different points in time - are commonly used in the social sciences to test theories of individual and social change. This book highlights the developments in this technique in a range of disciplines and analytic traditions. Providing an overview of models appropriate for the analysis of panel data, the book focuses specifically on the area where panels offer major advantages over cross-sectional research designs: the analysis of causal interrelationships among variables. Finkel demonstrates how panel data offer multiple ways of strengthening the causal inference process. He also explores how to estimate models that contain a variety of lag specifications, reciprocal effects and imperfectly measured variables.
Designed to help beginners estimate and test structural equation modelling (SEM) using the EQS approach, this book demonstrates a variety of SEM/EQS applications that include both partial factor analytic and full latent variable models. Beginning with an overview of the basic concepts of SEM and the EQS program, the author works through applications starting with a single sample approach through to more advanced applications, such as a multi-sample approach. The book concludes with a section on using EQS for modelling with Windows.
Through the use of careful explanation and examples, Berry demonstrates how to consider whether the assumptions of multiple regression are actually satisfied in a particular research project. Beginning with a brief review of the regression assumptions as they are typically presented in text books, he moves on to explore in detail the substantive meaning of each assumption, for example, lack of measurement error, absence of specification error, linearity, homoscedasticity and lack of autocorrelation.
It is often necessary for social scientists to study differences in groups, such as gender or race differences in attitudes, buying behaviour, or socioeconomic characteristics. When the researcher seeks to estimate group differences through the use of independent variables that are qualitative, dummy variables allow the researcher to represent information about group membership in quantitative terms without imposing unrealistic measurement assumptions on the categorical variables. Beginning with the simplest model, Hardy probes the use of dummy variable regression in increasingly complex specifications, exploring issues such as: interaction, heteroscedasticity, multiple comparisons and significance testing, the use of effects or contrast coding, testing for curvilinearity and estimating a piecewise linear regression.
This successful book, now available in paperback, provides academics and researchers with a clear set of prescriptions for estimating, testing and probing interactions in regression models. Including the latest research in the area, such as Fuller's work on the corrected/constrained estimator, the book is appropriate for anyone who uses multiple ......
This collection of essays brings together the current work on the concept of audience in written communication. The historical views of audience are first examined, then current theories are explored and sythesized. Finally, the contributors report on new qualitative and quantitative research on audience in written discourse.
