Editorial

Abstract

Introduction

Software

Software Evaluations
  Functionality
  Ease of Use
  Implementation
  Methodology

Assessment

Summary

Conclusions

Acknowledgements

References

Case Studies Index

3.2 Ease of use

Ease of use is fundamental for no matter how good a software system is, if the system is difficult to use this will discourage many from persevering with it. This assessment criterion includes the quality of the user interface and the quality of the support documentation.

The quality of the user interface refers to the organisation of the available options and the extent to which this organisation helps or hinders, in the first instance, the process of learning to use the software and subsequently, the process of continuing to use the software. Are the packages easy to come back to after a period of not using them? Windows interfaces place a barrier between user and system simply because they do not correspond to the way in which we are used to accessing tools in more general circumstances. Individual menu items are hidden and no matter how logical the structure of the menu, the user is inevitably confronted with the need to retain a mental map of how to access the required tools. This is true whether the system is essentially a cookbook compendium of tools or whether it is a system that provides the ingredients which the user must then select from and adapt in order to build their analysis.

The quality of the documentation - its comprehensiveness, how easy it is to use when needing to deal with problems - will also be assessed. Is there a tutorial guide with a trial dataset provided to help the user get started and appreciate the range of tools provided?

3.3 Implementation

The implementation criterion is a technical criterion and includes how easy a package is to install, its portability, how easily it can be extended and the kinds of data formats that it supports.

The installation criterion includes how easy the package is to download or install on the users system. Portability includes the range of platforms on which the software will operate. Extensibility covers how easy it is for the user to add new facilities and whether this has to be done via the source code or whether facilities to attach macros are provided. Finally the issue of data formats refers to both inputs and outputs. Does the software read in data output by other packages and write out data that can be read in by other packages - and what are the packages it has such interfaces with?

3.4 Methodology

Given the constraints on time it was not possible to make a full assessment of the four packages under all these criteria. The main focus of the assessment was to:

• Review the packages in relation to the broad criteria defined in sections 3.2 to 3.4.
• Make an initial assessment of the extent to which such software could be useful to social scientists.

Therefore the approach which was used involved:

• Identification of the facilities offered by each package from the documentation (including Web pages, as listed in the appendix to this report).
• Dowloading and installing each package (except SAGE).
• Use of a range of facilities from each package. The inital intention had been to attempt to load a set of data into each package, and carry out a simple program of ESDA involving the display of maps and graphs, and the fitting of a simple regression model. In the event, the time it took to learn even the rudiments of each system, using the various sets of manuals, help systems and tutorials, meant that all that was possible was a general review of the facilities in each of the categories listed in Table 2. The fact that this is all that could be done is testament to the fact that even highly visual systems, which have been designed to have an `intuitive' interface, require an investment of time before they can be used effectively.