"A comparative analysis of performance and behavioral outcomes in technology-based learning"
Katia Passerini, 
Doctoral Candidate in the Information and Decisions Systems, pkatia@gwu.edu
Department of Management Science, School of Business and Public Management
The George Washington University
Introduction and Summary
A systematic understanding of the impact of different technological media on the achievement of instructional goals enables educators to deliver the subject matter more effectively.This research focuses on the impact of selected instructional delivery modes in supporting the learning of business management topics. Among the various instructional technologies that advance learning, interactive multimedia is recurrently identified as a very powerful tool. The dissertation measures the effects of multimedia technology, compared to both traditional classroom and text-based instruction. It analyzes learners’ performance within a knowledge representation framework that looks at recall and application[1] of facts, concepts, principles, or procedures, as representations of instructional outcomes. The research uses a pretest-posttest quasi-experimental design to address concerns expressed by earlier comparison analyses. The design supports the examination of selected instructional objectives achieved within a specific timeframe, and isolates the effect of the medium on students’ (graduate and undergraduate) learning and satisfaction outcomes.
Prior Research
There is an on-going controversy in the literature dealing with the question of the learning effectiveness of instructional technology. In the last three decades, every time a new instructional technology has come about, researchers have tried to assess its impact on learning. Some scholars (Kozma, 1991) claim that media influences learning. Many others (Clark, 1983, 1985, 1991, 1994; Souder, 1993; Martin & Rainey, 1993; Barry & Runyan, 1995; McClure, 1996) claim that there is no difference in learning outcomes based on the medium. Few scholars (Moore & Kearsley, 1996) argue that researchers should not be asking effectiveness questions because “for any group of students, the environment in which learning occurs and the medium of communication between the teacher and the learners are not significant as predictors of achievement” (pg. 65). More recently, others (Jones & Paolucci, 1998) are calling for further research. There is no final answer yet. In spite of the several studies (Nugent, 1982; Mayer & Anderson,1991) and meta-analyses (Hartley, 1977; Burns & Bozeman, 1981; Kulik & Bangert-Drowns, 1983, Bosco, 1986; Kulik, Kulik & Shwalb, 1986; Kulik & Kulik, 1989, 1991; Roblyer, Castine & King, 1988; Kulik, 1994; Khalili & Shashaani, 1994; Fletcher-Flinn & Gravatt, 1995;) competing to answer the question: “do computers and other related technologies make a difference in learning?” this controversy is still alive.
Theoretical Background
The most relevant learning theories supporting the model tested in this research are related to studies on interactive multimedia and its impact on the cognitive system (figure 1).
Research Goals and Methodology
There are three main goals associated with this research:
1.The elaboration of a simplified theoretical model to assess learning.
2.The application of a research methodology that overcomes limitations – such as temporal and instructor-based biases - of earlier studies.
3.The delineation of a task-driven matrix on the comparative advantages of different instructional environments.
 
To achieve these goals, the dissertation applies a quasi-experimental design, which involves one factor (learning project management topics such as scheduling tools and change control) with three treatments (learning from the in-class lecture, reading the textbook, or using the interactive multimedia CD-ROM). The participants in the study are graduates and undergraduates enrolled in business administration degree programs at a major University in the East Coast. The study is replicated, under comparable conditions, with the two groups. The purpose of the replication is to investigate whether effectiveness (in recall and application tasks) and satisfaction vary by age (graduate and undergraduate), and by subject topic characteristics (complexity). The latter include topics such as ‘scheduling tools’ (higher complexity topic, or ‘hard-topic’) and ‘change control” (lower complexity topic, or ‘soft-topic’) – where complexity definition is based on a well know cognitive taxonomy (Bloom 1953).

The main and secondary research questions focus on performance and behavioral outcomes:

Question 1: Which instructional delivery environment best supports learning: multimedia, textbooks, or in-class lectures?
§Question 1.1: Which environment is more effective at achieving the learning objectives of recall or application?
§Question 1.2: Which environment is more effective at delivering a low-complexity and high-complexity topic?
Question 2: Which learning environment is more appealing: multimedia, textbooks, or in-class lectures?
§Question 2.1: Which environment, if any, is more appealing to graduates and undergraduates?
 
Figure 2 presents the learning model tested in the dissertation, and summarizes the expected media relationships developed on the basis of an extensive review of the literature.
To test the model presented in Figure 2, several univariate and multivariate data analysis techniques are employed. In addition, exploratory factor analysis is used to identify the underlying model, the number of factors in the model, and the variables associated with each factor. The expectation is that the analysis supports at least a 2-factor model (the recall and application groupings), although it may support other combinations and explain knowledge hierarchies. 
Expected Contributions
This research contributes to research on technology effectiveness in several ways:

§Offering an actionable theoretical framework that integrates different learning models, enabling the research to draw conclusions on the comparative effects of the medium as a whole, rather than on the individual components, methods, or specific media characteristics. 

§Presenting a unique comparative situation that enables the researcher to control for content-differences biases found in earlier studies. The research design overcomes several limitations of these earlier studies (controll