Beyond Broadband Access: Developing Data-Based Information Policy Strategies

CHAPTER 2 Understanding Digital Gaps

23 CHAPTER 2 Understanding Digital Gaps A Quartet of Empirical Methodologies bin zhang and richard d. taylor Since the 1960s, the field of information studies has had a tradition of trying to understand the role of information in society by measurement, typically by counting things: media, words, bits, and so on. This reflects an intuitive sense that something important is happening. Because of the intangible nature of the subject “information,” however, its role has been hard to grasp. This goal manifests itself in current times most often as the study of the so-called digital divide (or e-readiness). Approaches to this matter have grown more sophisticated over time, and now use complex statistical methodologies to parse huge databases for lessons from the past for potential gains from shaping the future. This chapter provides a detailed exercise in four of those methods, as applied to one country, China, but with the view that their underlying principles may be of broad application and of substantial use to planners and policy makers.1 These are not the only approaches—there are many— indeed, so many that the very idea of finding coherence in the field is challenging. While much progress has been made, it is argued that going forward there needs to be a new way of thinking about this field, and new emphasis on theory and testing as a way of developing analyses that are both explanatory and predictive. This study is the product of an international collaboration (United States and China) to advance the thinking in this field, which in China is broadly referred to as informatization, a concept that resonates with industrialization— a sweeping industrial and social change affecting all aspects of life and society. China has highly prioritized informatization, and created a top-level state b. zhang and r. d. taylor 24 council Leading Group to coordinate and promote the concept. The United States has no such grand metaphor, focusing instead on universal service and broadband access, which are much more limited concepts. Continued advancement of theory and practice in this field will be of greatest benefit to whoever advances informatization the most. Our intention with this study is to encourage scholars and policy practitioners to give informatization greater consideration. We chose China as a case study because it has pursued the empirical study of informatization for some time, and has collected extensive relevant data. This chapter provides examples of four methods for analyzing that data, with the hope that their underlying principles may be of general application and useful to policy makers. These methods are: • Static and dynamic analysis/analytic hierarchy process (AHP) • Hierarchical clustering analysis (HCA) • Time distance analysis (TDA) • Data envelopment analysis (DEA) The examples are different in name, but similar in purpose. They have the same starting point, the Informatization Level index, described below. They use different assumptions, and the calculations are different. Their common purpose is to objectively describe and analyze the digital divides within China. Based on the particular situation and the data available, one or several methods may fit better than others. Rarely will one method describe all the relevant considerations. In the examples in this chapter, the results are in some ways different, providing different insights; but also similar, in that their respective conclusions converge. Which method to choose depends both on the researcher’s goals and the philosophy. case 1: using static and dynamic analysis ⁄ analytic hierarchy process Based on a review and analysis of the structure (inductive/deductive), statistical methods, conclusions, and comparative strengths of twenty-eight existing digital divide index systems, a comprehensive index system was developed for the measurement of regional digital divides in thirty-one regions of China. Using the AHP model, rankings of indicators were obtained from the work of experts using factor analysis, and indicators measuring more effective access were given higher weights. Index weights were then determined for twenty-nine factors by computing a comparison matrix. Using regional data in China from 2002 to 2007, normalized by means of equalization, the un d e r s tan d i n g d i gi tal gaps 25 index values and ranks of informatization levels for each region for each year were obtained; then mean deviation was used to analyze the changing trends in the digital divides during those years. Attention was focused on five core factors affecting digital divides in China: technology, economy, government, education, and society. Case...