New Technology in Sociology

Practical Applications in Research and Work

Grant Blank, James L. McCartney, Edward Brent

• Publisher: Routledge
• Year: 2019
• eBook ISBN:
  9781000678383
• Print ISBN:
  9780887387692
• Subtopic: Social Science Research & Methodology

When technology has been applied in business environments, its justification has usually been cast in terms of saving time or saving money. In the social sciences, the justification must be different; the viability of sociology as a profession, for example, will not be enhanced by cost reductions. The focus in this volume is on a different bottom l

Introduction: New Technology and the Nature of Sociological Work

GRANT BLANK

No sociologist. . . should think himself too good ... to make tens of thousands of quite trivial computations in his head and perhaps for months at a time (Weber 1946, p. 135).

Max Weber's advice illustrates how much sociology has changed. Few sociologists would be willing to spend months making "thousands of trivial computations. . . Few have done so since counter-sorters came into wide use in the 1940s. Now counter-sorters are gone, replaced by another technology, the computer. Computers are a tremendously flexible and versatile tool, and they raise the question of the impact of new technology on sociological work with unprecedented force. The impact of technology is an important scholarly topic because of the recurring suspicion that technical tools influence the content and quality of scholarly work.

The articles in this volume examine the social, cultural, and political influence of technologies ranging from microcomputers to photography. To unify the presentation we need a common ground from which we can assess these divergent influences and technologies. This is provided by Charles Ragin's (1987) fundamental distinction between case-oriented and variable-oriented research. Ragin's distinction identifies two poles of a continuum on which most research may be arranged. At one pole, case-oriented research focuses on understanding the specific characteristics of individual cases. Cases are treated as wholes, and the distinctive combination of characteristics in each case is used to understand the context in which events occur.

By contrast, variable-oriented research focuses on making generalizations about populations. Cases become interchangeable observations that are characterized by the values of independent and dependent variables. Individual observations are analyzed only if they deviate from the major pattern of the population (e.g., as outliers). Variables are incorporated into a database of information used to test hypotheses. This is clearly a radically analytic strategy in that the observations are separated into their component parts as variables, and the main effort is devoted to examining the relationships among the variables.

The point of this distinction is to draw attention to the ways in which the organization of research and the requirements of data analysis techniques influence how the investigator thinks about the subject. The papers in this issue address that problem through examination of a variety of technologies. The first four papers by Lyman, Ragin and Becker, Walsh and Gordon, and Danziger offer a broad overview of the changes that microcomputer technologies may cause in our analytic habits, writing, and patterns of scholarly communication. Next, four pairs of papers examine the impact of technology on four topics: visual sociology, artificial intelligence, qualitative research, and teaching.

Cultures of Knowledge

The characteristics of each technology influence sociological work by making it easier to organize the research differently, or by making other kinds of data analysis easier (or even possible). Through these mechanisms, new technology may induce researchers to shift to a more case-oriented or more variable-oriented research strategies, or to move back and forth between the two. This view of technology is exemplified in the article by Peter Lyman, "Sociological Literature in an Age of Computerized Texts."

Lyman compares two approaches to text, reading, and writing which he calls "cultures of knowledge." In one, the structure and sequence of printed text are defined by authors as they write. This empowers the author, since the reader cannot alter the text, but encounters it in a single sequence. Thus, the context of the words and paragraphs in the text are preserved as they were set down by the original author. The typical text with these characteristics is a narrative, and Lyman calls this a "narrative culture."

The other culture of knowledge is characterized by the fact that the reader can, in principle, alter the structure and sequence of the text. This empowers the reader, because both the reader and author can determine the sequence of the text. Thus, the context of individual paragraphs and pages changes as readers link them together in different ways. This culture of knowledge emerges from social groups that build technical objects; for example, software and hardware designers, like Papert (1980) or Newell and Simon (1972). Lyman calls it a "technical culture."

An example of the latter approach is the way software reference manuals are used. Readers do not typically approach reference documentation with the intention of reading it from cover to cover like a novel. More likely, they use it to help solve a problem that can often be expressed as a question like: "How can I do X using this software?" This focus on problem solving is characteristic of a technical approach to knowledge. The text is treated as a database that contains information to aid in constructing the solution. Lyman argues that this has become a "paradigm of knowledge in the technical community and politically a model for reforming the knowledge base of every intellectual discipline."

In its extreme form, a text that gives readers the capability to read in sequences that were not specified by the author is called "hypertext." This concept, developed by Nelson (1965, 1987), is often referred to as nonsequential text. Such a characterization may be confusing since what Nelson means is not that hypertext is nonsequential in any strict sense—it is still read linearly—but the sequence in which a reader approaches it is not predetermined. The idea of hypertext is familiar in printed text, although not by that name. Devices like tables of contents, lists of figures, footnotes, and indexes give readers access to printed text in a hypertext-like fashion. Concordances and dictionaries are designed to be used as hypertext. The difference is that the flexibility of electronic text allows expansion of such capabilities far beyond what could be done using print.

Lyman's distinction between narrative culture and technical culture is analogous to Ragin's distinction between case-oriented and variable-oriented research. The primary characteristic of narrative text is that it is structured by the author. To understand narrative as the author intended involves a holistic understanding of the entire text as the context within which individual portions of the text acquire meaning and significance. Similarly, case-oriented research strategies try to grasp social processes and events as wholes. These approaches typically focus on an interpretive understanding of social life, often trying to understand the meaning of social events as the participants themselves understand it

An alternative research strategy removes events from their original, idiosyncratic contexts. Rather than understanding events in their own context, or as the participants understand them, the researcher imposes a conceptual ordering by classifying them into general categories (i.e., variables). The significance of the variables is defined by the questions that the researcher asks. Because researchers impose their own conceptual structure, variable-oriented research is characteristic of a technical culture approach to scientific inquiry. These two ways of examining text are summarized in Table 1.¹

Lyman uses this distinction to discuss the changing nature of scientific communication. Hypertext has the potential to undermine the concept of author because readers can annotate or change electronic text at will, much like text in a word processor. Advocates of hypertext argue that this will introduce a new style of collective writing in which electronic documents grow and change as new readers add their thoughts and commentary. This, they point out, is a democratic vision of writing because anyone can be published electronically and anyone can comment without restrictions.

Yet, the possibility of collective writing does not mean that it will be explored. To illustrate one problem, consider a public, electronic text of Max Weber's Protestant Ethic, to which anyone, in the hypertext spirit of collective writing, could add their annotations and marginalia. Would many sociologists find this an attractive prospect? I suspect most sociologists would prefer that people think they were not the second Weber, rather than give their colleagues an opportunity to read their annotations and remove all doubt. This suggests a real problem— maintaining quality.

How important is quality? Scientists need verified knowledge, and the peer review process is designed to check for errors and omissions. Beyond accuracy, high-quality work is well organized and contains intriguing ideas, vividly expressed. Ensuring quality is an important editorial function. As editor, my overriding concern in the preparation of this volume was to encourage authors and referees to write in the most thoughtful, careful, and sophisticated way that they could. Lyman points out that the editing function has been reinvented in several electronic contexts precisely to filter the dross from the valuable material. Further, the collective writing style of hypertext makes identifying the originators of ideas difficult. Recognition for promotion or tenure purposes is more difficult to obtain. In short, a fundamental weakness of the concept of collective writing as an approach to scholarly publishing is that it abandons mechanisms in the current publication system that encourage and provide recognition for higher-quality texts.

TABLE 1
Summary of Narrative and Technical Cultures

Many of these issues of publication and recognition reflect concerns with the reward system of science and, at least from a Mertonian perspective (Merton 1973), are at the heart of science as an institution. Given the importance of winning professional recognition, it does not seem to me probable that any technological capability that compromises the possibility of recognition will be adopted. Thus, collective writing, at least for professional purposes, is likely to remain an unexplored capability of hypertext.

Hypertext is valuable for other purposes. Some of those are explored by John P. Walsh and Andrew C. Gordon in their paper, "Intellectual Property and New Technology: Some Possible Futures in Sociology." One of several examples they discuss is the combination of electronic communications with hypertext on electronic bulletin boards. Hypertext would allow the original text of a paper to remain intact while the author and others add comments and revisions. The names of commentators can be retained in the hypertext document, so their contributions could be recognized and rewarded.

Many scholars may first encounter hypertext in their libraries. Libraries now have not only card catalogs in on-line databases, but also tables of contents of journals and edited books, abstracts of articles, and, in some cases, the full text of articles or books (Wheeler 1987). Librarians face the question of how to give scholars access to this mass of material. The software and hardware for scholars' workstations can be designed to retrieve text in many ways, but what design is appropriate? What capabilities should be given to users? For example, implementing a system that would give patrons the ability to retrieve abstracts and the results sections of articles without retrieving the methodology or theory sections is relatively simple, but is it a good idea? Some of this on-line system may look a great deal like hypertext. In fact, Lyman's job at the University of Southern California is to rethink how library patrons should be given access to different kinds of texts in preparation for designing just this sort of system. Understanding the impact of technologies like hypertext will become increasingly important as systems like this are implemented.

The Microcomputer Revolution

We're clearly in transition as new technology is first applied to old problems. Charles C. Ragin and Howard S. Becker's article, "How the Microcomputer Is Changing Our Analytic Habits," compares microcomputers to mainframes. In contrast to mainframes, the simple, highly interactive user interface of a microcomputer coupled with the low marginal cost of additional analysis encourages "deeper, more continuous, and more intensive dialogue between investigators, their ideas, and their data." For variable-oriented research, microcomputers facilitate analysis of individual cases (like outliers), and the use of graphical displays, which show the idiosyncracies in subsets of their data. At the same time, for case-oriented researchers microcomputers can automate some of the mechanical tasks of text processing. Thus, Ragin and Becker argue, microcomputers may provide a technological basis that will bridge some of the split between qualitative and quantitative research methods.

James Danziger's article, "Waiting for the Revolution: The Use of Microcomputers by Social Scientists," offers an all-too-rare opportunity to compare the often-repeated claims of a microcomputer revolution with reality. Danziger has surveyed social science microcomputer users repeatedly over the past four years, and his article contains the first results from this longitudinal project. Among many questions, Danziger asks, does microcomputer use encourage variable-oriented work? His answer is provocative; the social scientists in his study did not use more statistical or formal models after they received microcomputers than before. Of course, variable-oriented research strategies are broader than these models, but a limited answer is an important beginning.

Danziger's work reminds us that sociology as a profession can make a strong contribution to the discussion of the impact of computers. No other profession offers the combination of substantive focus on social phenomena with systematic methodologies able to measure social change. Sociologists need not rely on armchair theorizing or attention to leading cases; we have the methodological tools to be more thorough and systematic, to investigate the range of computer impacts, and discover whether the projections of commentators are true.

Visual Sociology

Douglas Harper's paper, "Visual Sociology: Expanding Sociological Vision," reports the status and future prospects for the use of visual tools. He begins by tracing the history of sociological photography from the photographs published during the pre-World War I, muckraking days of AJS to the radical sociology of the 1960s. He develops a typology of four categories of visual methods based on how sociologists understand information in the world around them: scientific, narrative, reflexive, and phenomenological. These categories organize Harper's digest of current work using visual methods.

Harper's work shows that technology does not necessarily promote variable-oriented research strategies. Visual sociol...

Table of contents

• Cover
• Half Title
• Title
• Copyright
• Contents
• Preface
• Acknowledgments
• Introduction: New Technology and the Nature of Sociological Work
• Index