Summary
Contents
Subject index
Making Sense of Statistical Methods in Social Research is a critical introduction to the use of statistical methods in social research. It provides a unique approach to statistics that concentrates on helping social researchers think about the conceptual basis for the statistical methods they're using.
Whereas other statistical methods books instruct students in how to get through the statistics-based elements of their chosen course with as little mathematical knowledge as possible, this book aims to improve students' statistical literacy, with the ultimate goal of turning them into competent researchers.
Making Sense of Statistical Methods in Social Research contains careful discussion of the conceptual foundation of statistical methods, specifying what questions they can, or cannot, answer. The logic of each statistical method or procedure is explained, drawing on the historical development of the method, existing publications that apply the method, and methodological discussions. Statistical techniques and procedures are presented not for the purpose of showing how to produce statistics with certain software packages, but as a way of illuminating the underlying logic behind the symbols.
The limited statistical knowledge that students gain from straight forward ‘how-to’ books makes it very hard for students to move beyond introductory statistics courses to postgraduate study and research. This book should help to bridge this gap.
Linear Regression Models and Their Generalizations
Linear Regression Models and Their Generalizations
Often social researchers are interested in the relationships of several variables in a single analysis. In this chapter, I account for the variation of one response variable with the added effects of a set of explanatory variables. Such distinction between the explanatory and the response will apply to all the models that I shall introduce below.1 We shall start with the multiple linear regression models with one metrical response variable and two or more explanatory variables that can be either metrical or ...
- Loading...