A Guide to Quantitative History

Synopsis

This book introduces statistical techniques for the particular needs of historians. The approach is practical and applied; alternative ways of resolving methodological problems are worked through. The emphasis is on applying the fundamentals of statistical theory to the specific research interests of historians and available historical sources. A special feature of the book is a discussion of the historical evolution of statistical techniques as a means of understanding their current applications and interpretations.

Excerpt

The first step of any analysis is to specify the unit of analysis, the variables, and the population to be studied. the unit of analysis is the case that is undergoing analysis. For example, in a study of 500 females, the unit of analysis is a person; in a study of thirty religious denominations, the unit of analysis is a religious denomination; in a study of 400 families, the unit of analysis is a family.

At times, multiple units of analysis are required. Consider a study of marriage. That subject may require two units of analysis: the individual and the couple. Structuring data such that cases (particular instances of the unit of analysis) are grouped into couples with one person first and the other second can accomplish this. a computer can then structure the data two ways: as couples, by organizing the two cases together; or as individuals, by organizing them separately.

Variables are characteristics of the unit of analysis. If the unit of analysis is a county, variables might be the state in which it is located, its total population in 1920, and its largest religious denomination. Some variables, such as the population, are already in numerical form. Others are not. For a variety of reasons it is useful to convert non-numerical variables, such as the name of the state and the county's largest religious denomination, into numbers. Conversion into numbers simply facilitates computer processing. the assignment of these numbers can be arbitrary and requires the preparation of a code book that identifies the meaning of numerical codes.

The population is the set of cases to be studied. It is important that we define the population explicitly; the researcher must determine in advance which cases are eligible for inclusion and which are not. Difficulties arise when the population is not properly defined or when one studies one popu-

Search by...
Show...

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.