and his colleagues Remmers and Brandenburg 1927). Centra (1993) divided the body of work on student evaluation of college teaching (SET) into the following four historical time frames:1. 1927-60 was dominated by the work of Remmers and his colleagues at Purdue University who developed the Purdue Rating Scale for Instructors and conducted early psychometric analyses to support what was, at the time, quite an innovative method of evaluation (Remmers 1927).2. During the 1960s, the use of student evaluations was almost entirely voluntary.3. The 1970s were considered the "golden age of research" on student evaluations and extended the earlier work to demonstrate the validity and utility of student ratings for both formative and summative purposes. 134 COLLEGE TEACHING T Downloaded by [Nova Southeastern University] at 23:15 27 December 2014 Vol. 52/No. 4 Further research on contextual variables and SETs: What do written comments reveal about the SETs? Do students comment on factors that are out of control of the instructor (e.g., climate of the room, time of class meeting, technical problems, etc.)? Downloaded by [Nova Southeastern University] at 23:15 27 December 2014

The World Wide Web (WWW) is a universal medium that has revolutionized the dissemination and gathering of information; however, on-line barriers limit the accessibility of the WWW for individuals with disabilities. Federal legislation directs that the provision of assistive technology devices and services enable individuals with disabilities to benefit from the opportunities available to their neighbors and peers without disabilities. The purposes of this study were to evaluate the accessibility of university special education programs' home pages and discuss accessibility recommendations. Eighty-nine special education Web sites were evaluated for accessibility errors. Most (73%) special education home pages had accessibility problems, and the majority of these errors (71%) severely limited access for individuals with disabilities. The good news is that the majority (83%) of the errors can easily be corrected. Recommendations and methods for improving accessibility to the WWW for individuals with disabilities are discussed.

No abstract

The purpose of this experimental study was to investigate the comparative perceptibility of hypertext navigation buttons in three configurations: buttons with both pictorial symbols and text labels, with text labels only, and with pictorial symbols only. An instructional HyperCard stack was created in three versions, each differing only in the type of buttons used. Subjects were given typical situated tasks which required them to interpret navigational functions of various buttons. Findings indicated that buttons with both pictorial symbols and text labels resulted in significantly less user confusion than did buttons with pictorial symbols only. Buttons with text labels only also produced significantly less confusion, compared to those with pictorial symbols only. These findings have practical implications for hypertext designers. Many extant stacks typically use buttons with pictorial symbols only, which may create user confusion during stack navigation.

migrated and evolved from those principles. Gustafson and Branch (1997) state that ''the role of models in instructional design is to provide us with conceptual and communication tools that we can use to visualize, direct, and manage processes'' (p. 18). The key concept here is the ability of the individual, when looking at any complex activity, to conceptualize a myriad of causal relationships and chart them in some manner that can be communicated to others. A given model's criterion must enable HPT analysts to accurately conceptualize a suspected performance problem in a given business environment. The ability to visualize and then communicate the process logic to others will be the true measurement of any HPT model's effectiveness and suitability for use.Stolovitch and Keeps (1992) report that early HPT practitioners attempted to use linear instructional design models to describe performance technology processes. These linear models did not always accurately describe the environment or inter-relationships in sophisticated, multifaceted business processes. As a result, the early pioneers in the HPT field began to develop their own unique models. The diversity and complexity of the analyzed environments, coupled with different perspectives and backgrounds of the profession's pioneers, have created a large number of models, many of which are still emerging and evolving. HPT PIONEERSThe works of Gilbert, Harless, Mager, and Rummler became the principles of the foundations for performance analysis and HPT modeling theory (Rosenberg, Coscarelli, & Hutchinson, 1992). Many have acknowledged Thomas Gilbert to be the ''father of performance technology'' (Dean, 1998). Gilbert felt that improving the performance of people must begin with identifying and resolving the environmental barriers, thus enabling the people (performers) to achieve maximum performance (Dean, 1997).Another performance technology pioneer who continued with Gilbert's diagnostic approach was Joe Harless. Harless believed that understanding the cause of a problem should drive any solution (Ripley, 1997). This belief would eventually become the process of front-end analysis as reflected in his first performance technology process model (Figure 1.1). This model had a clear focus on the early determination of goals and performance during the analysis phase. Later, Harless revised his original model so that it included the four phases of analysis, design, development and testing, and implementation and evaluation, which became well known by its abbreviation, ADDIE (Figure 1.2). Harless proposed to the performance technology disciples that a partnership and business focus should exist in order to apply the most costeffective intervention.