Informetric theories and methods for exploring the Internet: an analytical survey of recent research literature

Library Trends, Wntr, 2002 by Judit Bar-Ilan, Bluma C. Peritz

ABSTRACT

THE INTERNET, AND MORE SPECIFICALLY the WorldWide Web, is quickly becoming one of our main information sources. Systematic evaluation and analysis can help us understand how this medium works, grows, and changes, and how it influences our lives and research. New approaches in informetrics can provide an appropriate means towards achieving the above goals, and towards establishing a sound theory. This paper presents a selective review of research based on the Internet, using bibliometric and informetric methods and tools. Some of these studies clearly show the applicability of bibliometric laws to the Internet, while others establish new definitions and methods based on the respective definitions for printed sources. Both informetrics and Internet research can gain from these additional methods.

INTRODUCTION

Tague-Sutcliffe (1992) defined Informetrics as "the study of the quantitative aspects of information in any form ... and in any social group," and Brookes (1990) characterized it as "a generic term that embraces both biblio- and scientometrics." Along the lines of Tague-Sutcliffe, informetrics investigates: Characteristics and measurements of persons, groups, institutions, countries; publications and information sources; disciplines and fields; and information retrieval processes.

When the above definitions were offered, the World Wide Web was still non-existent, but today it is quickly becoming a major information source. Informetric methods can be and are applied to the Web, and new methods are being developed for this medium. This paper presents a selective review of research based on the Internet, using bibliometric and informetric methods and tools. The review is organized according to the following methods:

* Data collection methods

* Informetric analysis

* Citation analysis

* Cocitation and coword analysis

* Content analysis

* Evaluation using existing/new measures

* Identifying and calculating indicators

* Models

* Fitting existing models and bibliometric laws

DATA COLLECTION METHODS

Data collection from the Web is far from trivial, due to its size and its extremely dynamic nature. There are no methods to enumerate the "whole Web" (the total population under study) or to enable us to get a truly random sample of Web pages. When studying Web documents, sites, or the structure of parts of the Web, data collection is often carried out using the currently existing information retrieval tools, mainly the search engines, which are far from perfect. Bar-Ilan (2000a) discusses problems related to this type of data collection. Use studies rely mainly on surveys, interviews, and log analysis.

Surveys

Surveys on the Internet are employed mainly to receive information on the use of technology. Savolainen (1998) analyzes use studies of electronic networks. A considerable number of the reviewed studies collect data through quantitative surveys. Questionnaires can be sent out by regular mail or e-mail, can be filled out on Web pages, or can use a combination of these methods.

Lazinger, Bar-Ilan, & Peritz (1997) carried out an extensive survey on Internet use of the faculty members of Hebrew University of Jerusalem. The questionnaire was sent out by regular mail, in order to reach also faculty members who did not use e-mail. A follow up was sent to non-respondents. The overall response rate was 59.4 percent. More than 80 percent of the respondents used some Internet services, with e-mail being the most popular one (the questionnaires were sent in spring 1995, when the graphic browsers to the Web were just being introduced [Life on the Internet, n. d.]). Significant differences were found in the use patterns between the Humanities and Social Sciences faculty and the Science and Agriculture faculty.

Kovacs, Robinson, & Dixon (1995) investigated the use of discussion lists by library and information science professionals. The questionnaire was sent out to the participants of fifty-seven library and information science related discussion groups--approximately 10,000 participants. Filled out questionnaires were returned by e-mail. Only 576 responses were received. The majority of these respondents stated that discussion groups enhanced other sources of professional information. However, the majority also stated that discussion groups did not replace other sources of information.

The purpose of the survey conducted by Zhang (2000)--was to enhance understanding of the scholarly use of Internet based e-sources among LIS researchers and to evaluate the potential of Web-based surveys. The population of the survey was 201 researchers with in-press publications in eight LIS journals. An e-mail was sent to these researchers requesting they participate in the survey. The respondents could either fill out a Web-based questionnaire or request a printed copy and return it by mail or fax. Only 10 percent of the researchers requested printed copies, and 20 percent of the researchers returned the questionnaires by regular mail or fax (some of them printed out the Web-based questionnaires by themselves). Three follow-ups were sent out, and the total response rate was 89.1 percent.