The design and implementation of intelligent tutoring systems requires the close interaction, cooperation and communication of a number of disciplines. Among these disciplines are cognitive scientists, educators and computer scientists. Trans-boundary communication of concepts, terminology, and implementation procedures among the practitioners of these specialties is difficult and leads to confusion and time-consuming efforts to clarify these concepts. Therefore, a communication tool is needed to alleviate these problems. / Many intelligent tutoring systems rely upon elements of user feed-back to more precisely tailor the tutoring session to individual needs and requirements. Many significant interrelationships are generated within the intelligent tutoring systems and are available for analysis but are usually obscure. Therefore, a tool is needed to allow visualization of these interrelationships. / This dissertation explores the potential for using both precise and imprecise decision tables as not only software engineering tools but also as communication tools. These devices graphically represent interrelationships among elements of information supplied by the user, the system, and the researcher. Further, it shows that the employment of imprecise decision tables in conjunction with an intelligent tutoring system provides an effective method for dealing with, and making decisions about, many of the real-world situations and characteristics which are not readily represented by crisp methods. / Source: Dissertation Abstracts International, Volume: 53-09, Section: B, page: 4784. / Major Professor: Lois W. Hawkes. / Thesis (Ph.D.)--The Florida State University, 1992.

Educators' and psychologists' interests in individualization of instruction are founded in the intuitively logical assertion that every individual has unique learning or cognitive characteristics which combine to require unique instruction. However, that intuitive assertion has met with mixed results when tested in empirical investigations. Research has moved from the exclusive manipulation of various components of the instruction to broader issues of matching these various components of instruction with various learner characteristics. / This study inspects the specific instructional component of learner control of sequencing of instruction and the specific learner characteristic of field dependency. Although both sequencing and field dependency have been studied for some time, a recently introduced technological innovation of computer-based instruction via hypertext programming allowed the collection of learner sequential pattern data that was previously not available. This study correlates those patterns with a field dependency measure and posttest achievement. / A hypertext-based computer assisted instructional program was used to present a lesson in Bloom's taxonomy to trainers from private and public institutes in Singapore. The hypertext program was designed to allow the subjects to explore the lesson in a flexible manner. The computer recorded each subject's unique sequential pattern through the lesson. / Subjects' patterns were compared to each other with the SPSS Runs and Logistic Regression procedures to determine pattern types. These pattern types were then compared to subjects' Group Embedded Figures Test of field dependency and posttest scores using the SPSS Analysis of Covariance and Analysis of Variance Least Significant Difference procedures. / The study found that there were significant relationships between the field dependency measure and pattern types. However, there were no significant relationships between the pattern types and posttest scores. Only subjects' level of education had a relationship to posttest scores. / Source: Dissertation Abstracts International, Volume: 53-09, Section: A, page: 3149. / Major Professor: Marcy Driscoll. / Thesis (Ph.D.)--The Florida State University, 1992.

The purpose of this study was to ascertain E-mail users' assessments of two communication modes: computer-mediated communication (CMC) and face-to-face meetings (FTF). E-mail users' judgements regarding the freedom, frequency, and equality of interactions made possible with CMC vs. FTF were analyzed first. Subsequently, the two modes' perceived usefulness, effectiveness, and convenience were evaluated. Finally, the respondents' self-reported usage patterns were examined according to specific communication tasks and selected personality variables. / Data were collected from 202 E-mail users. Paired t-tests for the dependent samples and student t-tests for the independent samples were employed for testing hypotheses. / The major findings were as follows: For E-mail users, CMC was perceived as allowing for freer, more frequent, and more equal communication processing than FTF. For E-mail users, CMC was perceived as more useful, more effective, and more convenient than FTF. In addition, introverts perceived CMC as allowing more equal communication than did extroverts, whereas extroverts perceived FTF as allowing freer and more equal communication than did introverts. There were no differences between introverts and extroverts in the way E-mail users perceived CMC and FTF in regards to frequency of communication. For the E-mail users, CMC was regarded as more useful and more effective than for task-related communication, while FTF was regarded as more useful and more effective than CMC for nontask-related communication. / The results of this study showed that CMC and FTF were perceived differently among E-mail users. Furthermore, the notion of social presence was confirmed. This theory states that media have different degrees of presence and that users have different perceptions of the media according to the degree to which the latter contain social and nonverbal cues. / Source: Dissertation Abstracts International, Volume: 55-03, Section: A, page: 0408. / Major Professor: John K. Mayo. / Thesis (Ph.D.)--The Florida State University, 1994.

The purpose of this study was to investigate the relative effectiveness and efficiency of systematically designed materials using text alone, text augmented with normal speech (175 words per minute), and text augmented with compressed speech (262 words per minute). The study was designed to determine if systematically designed text augmented with compressed speech could increase the number of objectives achieved and reduce the actual amount of instructional time needed for mastery of the objectives. / The 78 subjects were from Florida vocational education programs located in five geographically dispersed schools. The materials were previously published as a part of a job skills training program. The research was conducted in individualized settings in the school media centers. The results indicated that the text augmented with compressed speech was equally effective and as efficient as text alone, and more efficient than text augmented with normal speech. A post-hoc analysis was conducted using instructional time as a variable. Systematically designed text augmented with compressed speech took significantly less time to complete than text augmented with normal speech. / Source: Dissertation Abstracts International, Volume: 51-01, Section: A, page: 0143. / Major Professor: Walter W. Wager. / Thesis (Ph.D.)--The Florida State University, 1989.

Of 736 reference librarians at four-year colleges and universities in the United States surveyed about their use of CD-ROM databases for reference purposes, 633 (86%) responded. Almost half (308) are using CD-ROM technology and collectively they own 809 work-stations and use 961 CD-ROM databases. Most (70%/221) not using CD-ROM have plans to acquire it in the future. By 1990, 63% (399) expect to be using CD-ROM for reference purposes. / Student enrollment is the most reliable predictor of CD-ROM use in a surveyed library, although the number of reference librarians, respondent age and size of the reference budget also detected significant differences. About 25% (38) of the institutions with fewer than 1,000 students are using CD-ROM compared to 82% (37) with enrollments exceeding 15,000 students. / The popularity of CD-ROM causes many libraries to regulate its use by such measures as requiring appointments and limiting use by length of time or frequency. Measures such as limiting patron use of printers and a policy of encouraging downloading of data for post-processing outside the library are often encountered. Despite such cost-reduction measures, most (60%/141) believe the annual overhead cost per work-station exceeds $250. Barely 2\% (7) find it necessary to have user fees to recover CD-ROM expenses. / Although 65% (200) of the libraries employ special security measures to protect the CD-ROM work-stations, fewer than 1% (2) believe they have serious security problems. Most (83%/520) respondents believe CD-ROM can be a cost-effective alternative to heavily used online databases. Over 80% (508) believe use of CD-ROM in a library has a positive effect on patron image of academic librarians. Significantly, 75% (238) of the librarians not now using CD-ROM share this belief. A majority of librarians (59%/182) using CD-ROM expect it will lessen future demands for printed reference sources. / Source: Dissertation Abstracts International, Volume: 50-03, Section: A, page: 0565. / Major Professor: Ronald Blazek. / Thesis (Ph.D.)--The Florida State University, 1989.

This experimental study explored the optimal use of hypertext for instruction by investigating the following questions: What are the effects of cognitive load of learning on learning outcomes and efficiency? Are there interactive effects between cognitive load and aptitude on students' learning outcomes and enjoyment of instructional method? / Seventy-nine ninth-grade students from a north Florida urban research school participated in this research. Six students participated in one-to-one evaluation of the materials, and 73 participated in the experiment. / The independent variables were cognitive load of learning and learning aptitude. Cognitive load was manipulated with learning guidance and access constraint to create two treatment groups--Hierarchical hypertext and prototype hypertext. Learning aptitude was chosen to be prior achievement in science, the subject area of the instructional treatment. Subjects' science scores on the Comprehensive Tests of Basic Skills were used as prior achievement data. / The dependent variables were general, intentional, and incidental learning outcomes, learning efficiency, and enjoyment of instructional method. Learning outcomes were measured with a multiple choice posttest, and enjoyment was measured with a questionnaire with an eight-point scale. / The experiment was a posttest-only, equivalent-groups design. Subjects of the same sex were pair-matched based on prior achievement and randomly assigned to the treatment groups. After subjects finished the hypertext instruction, they took the posttest and then answered the questionnaire. The results were analyzed with interval estimation. / Except results on enjoyment, most results are not statistically significant at the.05 alpha level. But the pattern of most results is consistent with theoretical predictions. Hierarchical hypertext was more effective for all categories of learning outcomes than prototype hypertext, but the advantage decreased with increasing prior achievement. Also hierarchical hypertext was more efficient for low aptitude students, but was less efficient for high aptitude students. Low aptitude students preferred hierarchical hypertext, and high aptitude students preferred prototype hypertext. Implications of learning path data collected for educational research are also discussed. / Source: Dissertation Abstracts International, Volume: 50-03, Section: A, page: 0670. / Major Professor: Marcy Perkins Driscoll. / Thesis (Ph.D.)--The Florida State University, 1989.

Learner characteristics (like previous training and experience on learner control, knowledge of the subject matter, and maturity level) and instructional variables (like statements of encouragement, advice, feedback, advance organizers, and monitoring strategies) have been found to affect what an individual learns when given control. / Instructional Curriculum Maps (ICMs) (Wager, 1978) are among the potential tools for improving learner control of instruction. ICMs provide information about learning objectives, which are important to self-monitoring of study behavior, and information on prerequisite and functional relationships among objectives, which can support decisions about sequencing instruction. / This study tests if: (1) Students who use a lesson map (LM) to guide study get better scores in a performance test than those students who use only a list of objectives (LI); or those who use a table of contents (TC); and (2) There is a positive correlation between ratings of student understanding of lesson content and performance for each treatment. / Two computer-based lessons on the topic of heat energy were programmed using a hypertext environment. The lessons were administered on two consecutive days to 81 ninth graders randomly assigned to one of three experimental conditions. A retention test was administered one week later. / The degree to which the students followed the prerequisite or functional relationships was analyzed using chi-square. A 3 x 2 x 2 factor analysis of variance with repeated measures on the last two factors was used to evaluate the effect of treatments on performance over time (posttest and retention test) and level of learning skills (high and low intellectual skills), and the Pearson correlation was used to test the relation between self assessment of student learning and performance. / Results show that students follow diagrammed sequences. Data do not support significant differences of performance for the three groups, neither for high and low intellectual skills, neither for posttest and retention test. Use of LM and CT affects positively self assessment of learning understanding and these effects last after a week. / Source: Dissertation Abstracts International, Volume: 50-06, Section: A, page: 1545. / Major Professor: Walter W. Wager. / Thesis (Ph.D.)--The Florida State University, 1989.

The purpose of this study was to examine the potential of computer assisted instruction for teaching problem solving skills. It was conducted in three phases. During the first phase, two pieces of problem solving software, The King's Rule and Safari Search, were identified and analyzed in light of the research and theoretical literature in the fields of problem solving and computer assisted instruction. During the second phase, two groups of six fourth grade students were each observed using one piece of software for seven 30-minute sessions. Think-aloud protocols were collected at the beginning and end of the observational period. Pretests and posttests were administered to assess problem solving ability and transfer. In the third phase, these data were first analyzed separately by software, then the results were compared. While the students used limited versions of the strategies the software claimed to teach, students were also found to have developed several strategies that allowed them to succeed in the program without using the desired strategies. No transfer of the problem solving strategies was observed. This study provides specific examples of how problem solving software affects the acquisition of problem solving strategies, student learning behaviors, and transfer. Fifteen factors that should affect the acquisition of problem solving ability with computers are also provided. This study has provided a detailed picture of how students interact with problem solving software and how the nature of that interaction may vary from what was intended. This type of research will provide valuable clues for the design of effective problem solving software. / Source: Dissertation Abstracts International, Volume: 50-12, Section: A, page: 3925. / Major Professor: Robert Allan Reiser. / Thesis (Ph.D.)--The Florida State University, 1989.

The main purpose of this study was to examine the effects of different levels of textual display and of timed versus untimed practice, using passages containing inserted adjunct questions, upon student performance on intentional and incidental learning. In this study, selected features of textual display were, based on previous research, expected to improve learner performance. On the other hand, variations in study time have also accounted for a sizeable amount of differences between groups of students. The second purpose of this study was to examine student motivation in relation to studying the instructional materials. / Subjects were 99 students assigned to four treatment groups. Learning materials presented to these treatment groups were under one of two levels of textual display (conventional textual display and enhanced textual display) and, in either of two time modes (time control and time free). / Student performance on the repeated and unrelated questions were measured by the posttest materials which consisted of two types of multiple-choice questions: 10 repeated questions and 10 unrelated questions. Student motivational reaction to instructional materials was measured by items from the Instructional Materials Motivation Scale (IMMS). / Data were analyzed using analysis of variance. Results indicated that there was no interaction between time and textual display on performance for both repeated and unrelated questions, and preference. However, time free treatment produced significantly higher scores for both repeated and unrelated questions. Results of the test for preference showed no strong choice by students for the materials exhibiting enhanced textual display. / These results reveal that time free treatment improves learners' performance. In contrast to earlier studies, the text display variations used in this study did not have any effect on their performance. Additional studies can help clarify the text display features that do and do not affect learner motivation and performance. / Source: Dissertation Abstracts International, Volume: 51-12, Section: A, page: 4098. / Major Professor: John M. Keller. / Thesis (Ph.D.)--The Florida State University, 1990.

A primary purpose of this study was to investigate causal relationships among monitoring strategy, knowledge, general problem-solving strategies, and problem-solving performance. In order to achieve these purposes, a theoretical model was hypothesized. A hypothetical, theoretical causal model was constructed. / Subjects were provided with three problems, and were asked to talk aloud during problem solving. Monitoring strategy and general problem-solving strategies were measured by protocol analysis of subject's verbalization of the problem solving process. The problem-solving performance was measured by the criterion of correct solution and solving speed for each problem. The knowledge variable was measured by a test developed to measure the subject's possession of the knowledge required to solve a problem. The data were analyzed by LISREL VI, a recent version of the LISREL subprogram package. / The following findings of the study were resulted in: (1) Monitoring strategy significantly affected planning strategy, means-ends analysis, representation strategy. The monitoring strategy had indirect effects on problem-solving performance. (2) Representation strategy failed to show a direct effect on performance measures. (3) Means-ends analysis did not directly affect work problem solving, but did affect inferential problem solving. (4) The direct effect of planning strategy was the most significant determinant for problem-solving performance. (5) Knowledge was a direct cause of means-ends analysis for word problem solving. / Source: Dissertation Abstracts International, Volume: 51-02, Section: B, page: 1019. / Major Professor: Robert M. Morgan. / Thesis (Ph.D.)--The Florida State University, 1990.