Background

The study in this chapter concerns the task of retrieving and analysing ESD by casual users of Statistical Information Systems (SIS). Both these key concepts will now be described.

Statistical Information Systems (SIS)

The term SIS is not widely used in the literature and systems described as SIS often differ quite significantly from one another. Malmborg and Sundgren (1994) provide a description that is consistent with the general usage, namely: an SIS is an information system that takes statistical data as its inputs and produces statistical information as its outputs. The tasks associated with SIS are to locate, organize and present statistical data in ways that are meaningful to users. This description is consistent with that found in the literature (Gassler, Frohlich & Kopcsa, 1996; Rogers, 1987).

The term SIS has not been widely used in the literature because the similarities that exist between SIS were often overshadowed by the differences that exist between the functionality and technologies found in different types of SIS. A Geographical Information System (GIS), for example, presents the results of statistical calculations on a map of a geographical region. Such a GIS takes statistical data as its inputs and produces statistical information as its outputs and, hence, satisfies the definition of an SIS. Similarly, software like SAS, Minitab or JMP can be shown to meet the definition of an SIS but are usually described as "statistics packages" or "statistical software." Finally, statistical databases could also be described, quite appropriately, as SIS but this is seldom, if ever, the case. Despite these differences, all of these types of systems do share the defining characteristics of SIS. Two more recent technologies that have been used to disseminate ESD are Online Analytical Processing (OAP) and the World Wide Web (Web). It is these two technologies that were integrated to produce the Abacus prototype. Both of these will be described in more detail.

Online Analytical Processing (OLAP)

Thomsen (1997, p. xvii) describes OLAP as "the process of creating and managing multi-dimensional enterprise data for analysis and viewing by the user who seeks an understanding of what the data is really saying." Three aspects of Thomsen's description need further explanation. First, Thomsen says OLAP systems create and manage multi-dimensional data because, typically, OLAP systems generate either a permanent or temporary multi-dimensional database from original source data. Second, he notes that OLAP systems do more than simply allow the user to view that multi-dimensional data set, they allow the user to analyse it. They do this by allowing the user to select the dimensions that will be involved, the level of aggregation of each of the dimensions and the members of dimensions to be included in their custom view of the data. Finally, Thomsen says that OLAP allows the user to see what the data is really saying. By this, he means that OLAP allows the user to bring together disparate sets of data, to detect trends in that data and to explore the underlying causes of those trends by drilling down into the data.

The main features associated with OLAP in the literature include:

  • Providing a simple means of analysing multi-dimensional data, which is usually historical data drawn from a number of sources.

  • Supporting complex analysis such as calculations and modelling applied across dimensions, through hierarchies and/or across members.

  • Carrying out analyses in an interactive setting and with a response time of five to 20 seconds.

Providing several common functions including being able to:

  • Add or remove entire dimensions from an analysis of a population (rotation).

  • Aggregate (roll up) or disaggregate (drill down) data for any dimension having a hierarchical structure.

  • Filter particular members of a dimension out of an analysis.

  • View a sub-population of the data (slicing) (Thomsen, 1997; Pendse, 1999).

An initial set of requirements for the manipulation of ESD, by casual users of ESD, was gathered via pilot studies conducted by Hyland and Hasan (1997). The functions provided by OLAP appear to match these requirements very closely. For this reason, OLAP would seem to be a useful technology to incorporate in a system for casual users of ESD.

World Wide Web and the Internet

One problem associated with the retrieval and analysis of ESD is the difficulty in getting access to sources of data in electronic form. This problem has been significantly addressed in recent years by the use of the Internet and the Web. The Web has been described as a global information service that is available on almost all hardware platforms and which subsumes virtually all the previous Internet tools (Ford, 1995). The Web provides a common interface to virtually the entire Internet and in which the technological aspects of navigating the network are virtually invisible to the user. From a user's perspective, the Web provides a method for moving from one resource to another by simply clicking on a link.

There are a growing number of references in the literature to the use of the Web and the Internet by ESD providers. The growing use of the Web by community organizations and local councils (Latamore, 1996; McQueen, 1996; Zorn, 1996) would strongly support the idea that the Web could play an important part in the provision of ESD to the community at large. In regard to the casual users of ESD, the Web provides an excellent means of accessing the data because the hypertext interface is straightforward and unintimidating (Ford, 1995).

The Tasks Carried Out by Users of ESD

The literature does not appear to provide any model of the task of using ESD. This is not surprising and supports the proposition that many of the problems encountered by users of ESD stem from the lack of adequate models of the tasks users perform. Malmborg and Sundgren (1994) imply an initial set of sub-tasks in their definition of an SIS, namely, to locate, organise and present statistical data in ways that are meaningful to users.

Task models do exist for related tasks, such as information gathering for managerial decision-making (Goodhue, 1998) and these models may serve as a useful starting point for a task model for the use of ESD. There is a close parallel between the three sub-tasks suggested by Goodhue and those suggested by Malmborg and Sundgren. Goodhue (1998) posits that the task of decision-making consists of three sub-tasks and suggests some of the factors that may affect the success of these three sub-tasks. These are summarized below:

  1. Identification of the data needed:

    • Identification of the dimensions, structure and level of detail required.

    • Selection of a source or sources from which the data can be acquired.

  2. Acquisition of data:

    • Query formulation, selection of useful data, paring of unwanted data.

  3. Interpretation of that data:

    • Assessment of the currency, credibility and compatibility of the acquired data.

The models suggested by Goodhue and by Malmborg and Sundgren are shown in Figure 1.

click to expand
Figure 1: Comparison of Goodhue's Model and a Proposed Model for ESD Use

Figure 1 shows a possible relationship between these sub-tasks and the subtasks suggested for ESD use by Malmborg and Sundgren (1994). The use of ESD is frequently associated with decision-making, which accounts for the similarity between the proposed model of ESD use and Goodhue's model of information gathering to support managerial decision making.

At this stage, it is unclear exactly what is involved in the three subtasks proposed by Malmborg and Sundgren (1994). However, the availability of the Abacus prototype provides an excellent opportunity to observe typical users of ESD carrying out the task of manipulating ESD and so add further detail to the skeleton model proposed in Figure 1. The results of user observation and the resultant model of the task are discussed later in the chapter.



Computing Information Technology. The Human Side
Computing Information Technology: The Human Side
ISBN: 1931777527
EAN: 2147483647
Year: 2003
Pages: 186

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net