Search

The relationship between learning to program at a construct level and learning and study strategies was studied for college students enrolled in a beginning Pascal programing course and a calculus course (four sections of the programing course and two of the mathematics course). For both the experimental group (n=42) and the first control group (n=51), the programing course consisted of 150 minutes of lecture and 150 minutes of laboratory each week. Both groups were given instruction in programing and Pascal, but only the experimental group was taught the loop and nested loop constructs and the corresponding Pascal codes whose outputs embodied the combination and intersection schema of Piaget. The second control group (n=38) was given instruction in calculus with 200 minutes of lecture per week for 7 weeks. The 77-item Learning Strategies and Study Skills Test, a transfer test, and an achievement test were used. Results show significant effects of learning and study strategies on the learning of complex programing constructs. A greater degree of programing skills increased the capability of selecting main ideas, but had no effect on transfer of programing skills to solving analogous problems in other domains. Results also indicate that learning complex programing constructs led to greater levels of anxiety, but had no effect on other affective factors of motivation, attitude, and concentration. Six tables present study findings. Contains 33 references. (SLD)

Performances of 45 individuals with varying degrees of formal and informal training in machining and programming were compared on tasks designed to tap intellectual changes that may occur with the introduction of computer numerical control (CNC). Participants--30 machinists, 8 machine operators, and 7 engineers--were asked background questions and then presented with a series of paper-and-pencil tasks corresponding to the phases of the machining process. Participants' histories demonstrated many different routes into developing skills at work. Those who began careers by learning on the job were less likely to be involved in programming. Younger workers worked with CNC machines more often. The majority felt that hands-on experience supplemented by class work for certain basic topics was the most efficient way to develop skills in machining. Learning and teaching on the job were commonplace. The experimental tasks revealed distinctions in thinking patterns related to hands-on versus programming experience and to machining versus engineering training. They tapped experiential differences and yielded some evidence of mental restructuring due to learning programming. Participants were convinced of the power of hands-on learning and would not recommend a future work force deprived of traditional practical knowledge. Recommendations for CNC machining training included hands-on experience plus class work in basics and up-to-date techniques and continued training and mentoring among employees. (Contains 25 references.) (YLB)

In August 1992, the Eastern Iowa Community College District convened a 2-day Developing a Curriculum (DACUM) workshop involving a panel of six computer programmer/analysts (CP/A's) in the identification of specific competencies and tasks expected of CP/A's. To validate the entry-level skills identified, a survey was conducted of 96 businesses employing CP/A's. The survey instrument listed 106 specific job skills in the following categories: communication skills (e.g., display a positive attitude and demonstrate good listening skills); working as part of a team (e.g., follow directions and brainstorm ideas); utilizing organizational skills (e.g., develop task lists and manage multiple projects); designing/writing programs/systems (e.g., test and debug programs/systems); changing/modifying programs/systems (e.g., modify job control structure); creating documentation (e.g., identify level of documentation being created); continuing education (e.g., share information with colleagues and utilize available manuals); operating personal computers (e.g., follow shutdown and security procedures); and utilizing mainframe resources (e.g., follow log on/off procedures and perform file maintenance). Respondents were asked to identify the skills they expected entry-level CP/A's to perform and to indicate the importance of each skill. A total of 22 usable surveys were returned for a 25% response rate. Only two of the 106 skills were not considered to be entry level by a majority of the respondents: selecting the language to be utilized and conducting training. The survey instrument is provided. (PAA)

Teaching of computer programming in elementary and secondary schools has become a very popular activity. The purpose of this paper is to review and analyze the empirical research on the benefits of learning computer programming on students' cognition abilities. The paper focuses on the question: Do the skills of computer programming transfer to students' cognition? The paper is organized into three sections based on the measurement outcomes of each study. Section 1 contains seven studies that investigated the relationship of programming with the following general cognitive outcomes: metacognitive ability, cognitive development, cognitive growth, general ability, creativity, and achievement. Section 2 contains eight studies that analyze the relationship between learning computer programming and problem solving. Section 3 contains six studies that examine learning computer programming with the following specific cognitive skills: rule-learning tasks, development of thinking skills, reasoning skills, general variable skills, and planning skills. In the 21 studies reviewed one half showed learning computer programming did not affect student cognition, and the other half showed some positive effects. However, the author concludes that all of the studies were flawed. For example only one study reported the reliability and validity of its instruments. Based on the studies, the author concludes that there is no relationship between learning computer programming and student cognition. (PR)

This study investigated the cognitive benefits of learning how to program by determining the degree of cognitive transfer of programming skills at a construct level to solving analogous problems in other domains. Subjects, who were students enrolled in four sections of the beginning Pascal programming course and two sections of a calculus course, were assigned to the experimental group or one of two control groups. For both the experimental group (n=42) and the first control group (n=51), the programming course consisted of 150 minutes of lecture and 150 minutes of laboratory per week. Both groups were given instruction in programming and in the language Pascal through lectures and laboratory activities lasting 6 weeks and 4 weeks respectively. Only the experimental group was taught loop and nested loop constructs and the corresponding Pascal codes whose outputs embodied the combination and intersection schema of Piaget. The second control group (n=38) was given instruction in calculus consisting of 200 minutes of lectures per week for 7 weeks. Results showed significant effects of learning the nested loop construct on solving analogous problems in other domains. By using a tracer, the degree of learning of the nested loop construct was positively related to the degree of transfer. Students who scored higher on the achievement test, which was used in determining understanding and mastery of the nested loop construct, scored higher on the transfer posttest than those who scored lower on the achievement test. Students who learned the nested loop construct and who did not perform well on the transfer posttest showed significant improvement after a hint was given. (Contains 7 tables and 32 references.) (Author/ALF)

This paper presents the results of a study on the effects of using 3 simulators with 14 computer science majors enrolled in an undergraduate computer architecture course. Three ability groups (high, medium, and low) were formed based on the students' quality point averages for prior computer science courses. Analyses of data from a pretest, a posttest, instructor-prepared achievement tests following each of the simulators, programming assignments, and survey results showed: (1) no significant differences among the three groups on the pretest and posttest scores; (2) only the low and medium ability groups showed significant performance gains from pretest to posttest; (3) the use of the simulators equalized the student performance in the three ability groups; (4) nonsignificant differences on both programming assignments for all ability groups; (5) positive student attitudes toward and perceptions of the use of simulators; and (6) student perceptions that the simulators made learning about computer architecture more concrete, and helped in writing and understanding programs and microsequences of instructions. Suggestions for further research are presented including using larger samples, different methodology for defining the ability groups, and the use of qualitative interviews to ascertain student perceptions. Six figures present the results of the various statistical analyses. (Contains 5 references.) (ALF)

This paper describes three computer-based simulators developed to support an undergraduate computer architecture course. Using a hypermedia authoring package, a simulator was created for each of the three course content areas: (1) introduction of the hypothetical computer structure--contains a tutorial, a quiz, and a simple program that presents background information and introduces the architecture of the hypothetical machine and the first 11 fundamental machine instructions; (2) presentation of the microsequence operations--includes a quiz section which tests student understanding of concepts; and (3) actual programming of the hypothetical computer--allows students to write and execute programs using a graphical presentation of the program. Results from research on the class indicated that students' responses were favorable to the incorporation of the simulators in the class; students showed lower levels of frustration and a higher level of understanding of abstract materials; students shifted from lecture and classroom questioning on specific items as a means of obtaining information to the consultation of the appropriate simulator to obtain the answer; and the class showed significant improvements. Two figures illustrate sample screens from Simulators One and Two. (ALF)

In October 1991, a labor market assessment was conducted to provide information and direction for the curriculum development and evaluation efforts of the Eastern Iowa Community College District's (EICCD's) Business Computer Programming (BCP) Program. The study focused on employment opportunities for and educational needs of personnel performing business computer programming functions. Surveys were sent to 96 local businesses believed to be employing business computer programmers, requesting information on the functions and training needs of computer personnel; minimum educational level and salary of entry-level computer programmers; immediate and projected job openings; and emerging trends. Study findings, based on a 35% response rate (N=33), included the following: (1) 92% of the respondents reported using microcomputers, with most (85%) using them for word processing; (2) 64% reported that they used mainframe computers, and 36% reported using mid-range computers; (3) 57.1% reported that an associate degree was the minimum educational level for full-time mainframe computer programmers; (4) respondents utilizing mainframe computers projected 111 job openings through 1994, while those using mid-range computers projected only 12 openings; (5) respondents using mainframe computers who have hired EICCD graduates recommended more emphasis on accounting, communication skills, and human relations skills in the BCP program; and (6) 85% of the respondents estimated that their company would send employees to community colleges for training. Data tables and the survey instrument are included. (PAA)

The Computer Rehabilitation Training Program for the Severely Disabled is a job-oriented training program to prepare physically handicapped persons to become computer programmers and analysts. The program is operated by: a nonprofit organization of Baton Rouge-area business people interested in data processing; the Department of Social Services, Louisiana Rehabilitation Services; the Louisiana State University (LSU) System Network Computer Center; and the LSU Division of Continuing Education. The program objectives include: provide training to enable students to develop skills necessary to be successful in a computer programmer/analyst position; provide a pseudo-work environment to aid the student to succeed in a working business environment; assist the graduate in acquiring his/her first position as a computer programmer/analyst; and develop independent living skills. The program seeks to maintain a highly qualified and complete curriculum designed to meet the needs of both the students and their future employers. The curriculum, which includes over 1,800 hours of lab experience during 11 months of instruction, encompasses such subjects as accounting, C programming, COBOL programming, operating systems, spreadsheet applications, and relational databases. This pamphlet describes the role of each agency helping to operate the program and lists admissions policies, academic and program policies, causes for dismissal, formal evaluation and review procedures, certification policy, awards, and benefits. (JDD)

This paper presents the results of a study designed to determine the computer anxiety, attitudes, and confidence levels of students enrolling in preservice teacher education computing courses. Specific questions compared these students with national norms, and looked to see whether sex, age, or past computer experiences had any effect on measures of student enjoyment of using computers, their computer anxiety and confidence levels, and their perceptions of the usefulness of computers. Data were collected from 518 college students (394 female, 108 male, and 16 not reported). Results indicated that prior knowledge and experience with computers correlate highly with student attitudes toward computer use. This relationship indicates that these earlier experiences were generally successful, even though the majority of the experiences consisted of recreational computer games and simple word processing. Approximately 32% of the variance in anxiety scores, 30% of the variance in confidence scores, and 10% of the variance in usefulness scores were shared by previous word processing, recreation, programming, and database experience. Removing sex, age, and experience using spreadsheets did not significantly change the predictive value of the equation for the anxiety, confidence, or usefulness subscale scores. It is concluded that student attitudes toward working with computers are important indicators of the ways in which students will use computers when they become teachers in their own classroom settings. (9 references) (DB)

The Knowledge Engineering for Young Scholars (KEYS) Program was a National Science Foundation (NSF) program conducted at Louisiana State University during 1989 and 1990. The program's goals were to increase 8th-12th grade students' exposure to science, acquaint them with university research, stimulate interest in science, and build their confidence to pursue science careers. To achieve this goal students were introduced to the concepts of artificial intelligence, expert systems, and knowledge-based systems. This document presents the evaluation report of the KEYS Program in six sections. The first section identifies the goal of the program. The second section provides information about the purpose of the program, a program description, student recruitment and selection criteria, program activities, career awareness activities, philosophy and ethics of science activities, follow-up activities, the project setting, a staff profile, and cost considerations. The third section describes the evaluation study of the program that included NSF pre- and post-participation survey forms, and interviews with KEYS program managers (n=2), faculty mentors (n=2), program teachers (n=2), counselors (n=2), and random samples of student participants (n=10), parents of KEYS student participants (n=6), and applicants who did not participate (n=10). The fourth section provides brief summaries of each groups' responses to specific interview questions, and results of the NSF surveys. The fifth section concludes that: (1) since the participants were already highly motivated and interested in science careers, little change in reported interest was evidenced; (2) student participants' self report on their confidence level for their first choice of a major in college was not reliable; and (3) students did report that they did work with university faculty, develop expert systems, and learned about careers in science. The sixth section presents recommendations concerning program length, effective activities, participant recruitment, participation fees, and program assessment techniques. An appendix provides sample abstracts of KEYS students' expert systems. (MDH)

A study involving 120 undergraduate students attending the University of St. Thomas in St. Paul (Minnesota) was undertaken to test the interactive effect of instructional strategy (structured versus constructed) with learner's prior domain knowledge in concept acquisition. Previous instructional design research on concept learning has focused on structured strategies (i.e., expository and practice presentations) for initial learning of concepts. The assumption has been that learners had no prior domain knowledge of the concepts to be learned. The propose of this study was to extend that research by investigating the interaction of instructional strategies with students who have prior domain knowledge. Six abstract programming concepts from the domain of structured languages were selected as the content for the learning program. Results indicate that learners with no prior domain knowledge learned concepts better with a structured strategy than with a strategy that forced learners to construct the necessary conceptual knowledge. In contrast, learners with prior domain knowledge performed better when required to construct the knowledge than did learners who experienced the structured strategy. The importance of learner prior domain knowledge is discussed as it pertains to the selection of instructional strategies. Two tables are included. (TJH)

A computerized grammar checker was developed to assist teachers of English as a Second Language in editing student compositions. The first stage of development consisted of an error analysis of 125 writing samples collected from students. The 1,659 errors found were classified into 14 main types and 93 subtypes. This analysis served as the basis for constructing a taxonomy of mistakes and ranking the categories according to frequency of occurrence and comprehensibility. The grammar checker was then designed with a small electronic dictionary containing, 1,402 word stems and necessary features, and a suffix processor to accommodate morphosyntactic variants of each word stem. An augmented transition network parser equipped with phrase structure rules and error patterns was then constructed. In addition, a set of disambiguating rules for multiple word categories was designed to eliminate unlikely categories, increasing the parser's efficiency. The current implementation detects seven types of errors and provides corresponding feedback messages. Future research will focus on detecting more kinds of mistakes with greater precision and on providing appropriate editing strategies. (Author/MSE)

Computer assisted instructional environments offer unique opportunities for the analysis of learning and problem solving processes. In this paper, partially automated techniques for process analysis are applied in a study of the problem solving strategies of a group of gifted students learning the Logo programming language. The objectives of the paper are to develop strategy specifications, construct a model of the students' programming process, and illustrate the application of partially automated interaction process analysis. The computer software programs included tools for collecting, displaying, encoding, and dynamically representing the students' problem solving in the Logo environment. Two tasks were administered: a graphics task midway through the program, and a word-and-list task at the end. Problem solving strategies were identified in each task, and it was found that they differed across tasks and across subjects. Results indicate that students' strategies are a function of specific task demands and the students' representations of the problem, their knowledge of program design, and their knowledge-base of specific methods for implementing their designs in code. Recommendations for future behavioral science research in the area of problem solving include routine use of microcomputers to monitor students' performances and provide appropriate individualized assistance instantaneously. (66 references) (Author/DB)

This paper presents a four-component instructional design model for the training of complex cognitive skills. In the analysis phase, each skill is broken down into a set of recurrent skills that remain highly consistent over various problem situations, and a set of non-recurrent skills that require variable performance over different situations. Components 1 and 2 relate to analyses that are conducted to describe the recurrent skills in terms of specific rules or procedures and the non-recurrent skills in terms of heuristics or systematic problem approaches. Components 3 and 4 relate to the knowledge analyses that are conducted to describe the facts and concepts that are prerequisite to perform recurrent skills or the complex schemata that are helpful to perform non-recurrent skills. In the design phase, instructional tactics are selected for each of the four components. Components 1 and 2 relate to the design of practice and the conditions leading to rule automation or schema acquisition. Components 3 and 4 relate to the design of information presentation. Finally, the composition of the training strategy is governed by selected instructional tactics, which pose constraints on the design of an interactive learning environment. The basic prediction of this model is that its application leads to "reflective expertise" and increased transfer performance, i.e., the ability to solve new problems on the basis of domain-specific procedures and heuristics. Applications of the model that support this prediction are discussed for the training of fault management, computer programming, and statistical analysis skills. (74 references) (Author/DB)

The Nihongo tutorial system is an intelligent tutorial system designed to use a computer to assist scientists and engineers in developing reading competence in technical Japanese. It consists of three applications: the Nihongo Tutor, which provides useful information about an article (translation, syntax, pronunciation) to help understand the text of a personalized lesson, and records student performance; the Parse Tree Editor, used to prepare technical articles for the tutorial system; and the Administrator, which uses student data to assign an article that is technically and pedagogically appropriate. The program integrates the three functions. This report details the technical specifications for system implementation, including both the data structures and algorithms used. The section on the Administrator looks at the attributes of the system files, technical (discipline) fields included, user records, user article selection, and user database. Information on the Nihongo Tutor covers text data structures, student database management, and electronic dictionaries. An explanation of the Parse Tree Editor offers specifics on the four stages of parsing: segmentation, syntax, semantics, and mapping. Some screens are used for illustration. (MSE)

Advantages and disadvantages of standard Rasch analysis computer programs are discussed. The unconditional maximum likelihood algorithm allows all observations to participate equally in determining the measures and calibrations to be obtained quickly from a data set. On the advantage side, standard Rasch programs can be used immediately, are debugged and accurate, and can report statistics that are difficult to calculate. On the disadvantage side, the user must have the correct hardware, conclusions based on numbers appearing in the output are not necessarily valid, and the effort involved in producing one's own program can be considerable. Guidelines for writing programs, including those that surpass standard programs, are provided; and sample output from a number of standard programs is examined for strong and weak points. Emphasis is on adequate and useful statistics presented as easily comprehended graphical output. Topics discussed include: estimated measure, standard error, goodness of fit, plotting fit, presenting the variable, and differential item functioning or bias. Eleven figures and a BASIC program for performing unconditional maximum likelihood estimations are provided. (TJH)

Students learning to program recursive LISP functions in a typical school-like lesson on recursion were observed. The typical lesson contains text and examples and involves solving a series of programming problems. The focus of this study is on students' learning strategies in new domains. In this light, a Soar computational model of self-explanation strategies of instructional text was examined. The subjects, 12 students with little or no computer programming experience, proceeded through a series of lessons in the LISP Tutor program on recursion. Subjects, who were asked to think aloud, were videotaped as they read through an instructional booklet prior to solving problems using the Tutor. Subjects were also asked to explain examples to themselves. After reading their texts, subjects worked through 12 recursion problems with the LISP Tutor. Subjects were then divided into groups of "good" and "poor" performers. The verbal protocols were segmented into elaborations, which are pause-bound utterances and are not a first reading of the text. Each elaboration was categorized in terms of the instructional content to which it referred and to one of six types of comments. Sequences of related elaborations were coded into episodes (or macro-codings). Results indicate a strong correlation between types and contents of elaborations made and subsequent problem-solving performance. Self-explanations of "good" students were much more structured into goal-based episodes than were those of "poor" students, indicating more persistence on the part of the "good" students. Three data tables and three flowcharts are included. (TJH)

The development of a powerful learning environment aiming at the acquisition and transfer of general thinking skills through learning to program in LOGO was studied. The following conditions were sought: (1) sufficient knowledge of the LOGO primitives and concepts; (2) mastery of the thinking skills within the programming context; and (3) explicit training for transfer. A systematic experiment was conducted in 3 6th-grade classes with 72 12-year-old students, using a pretest-posttest design with a non-treatment control group. A LOGO course, involving the training of a systematic programming strategy, was taught to two of the classes. In one of these classes, explicit instruction for transfer was also applied. At the end of the school year, the fulfillment of the conditions was tested. Findings reveal that the first two transfer conditions were fulfilled in both experimental groups; results with respect to explicit training for transfer were less positive. A series of transfer tests was administered, and data analysis revealed that transfer was obtained in both experimental groups. This finding implies that fulfillment of the first two conditions is sufficient for realizing transfer of thinking skills. Three tables present study data, and 13 figures illustrate items from the tests. (Author/SLD)

Project TACKLE (Together, Addressing the Challenge of Knowledge and Literacy for Employees) was a cooperative project, jointly developed to address employees with the inadequate basic skills necessary to operate modernized, technical equipment and maintain job security. Approximately 500 current employees of General Motors, Flint, Michigan, used the Learning Lab in some capacity throughout the 27-month project period. Project TACKLE brought together a local educational agency (Carman-Ainsworth Community Schools) and the area's predominant industry (General Motors Corporation) to meet the demands of the work force and the challenge of new technology. Specific outcomes were provided through participation in various components of the Learning Lab: (1) adult literacy and basic skills information; (2) adult high school completion; (3) structured classes; (4) limited English proficiency training; (5) basic computer literacy and computer programming; (6) problem solving skills training through the PALS (Principal of the Alphabet Literacy Systems) program; (7) educational and career counseling; and (8) an interpreter for hearing impaired employees. This project met the needs of special populations: people with hearing impairments, those for whom English is a second language, and individuals from diverse backgrounds. An ongoing evaluation was performed for different aspects of the program. At the culmination of each module, the Learning Lab was modified. In addition, each participant was given an exit survey, including program dropouts. (Four appendices are included in this evaluation report: partnership agreement; training needs assessment; dropout questionnaire; and quarterly reports.) (NLA)

This is the first issue in a series of reports on a national survey conducted to determine the number and primary uses of microcomputers in schools. The survey sample consisted of 2,209 public, private, and parochial elementary and secondary schools in the United States. Findings of the survey (conducted in 1982/83 by means of mailed questionnaires and telephone contact to the schools) indicate that: (1) secondary schools are the largest pre-college users of microcomputers, with emphasis on the teaching of BASIC language programming; (2) 53 percent of schools now have at least one microcomputer for student instruction; (3) the use of microcomputers is growing faster in secondary schools than in elementary schools; (4) elementary schools tend to have only cassette-based microcomputer equipment; (5) small parochial elementary schools and public schools in poorer districts are less likely to have microcomputers; (6) microcomputers are used for both drill-and-practice and for programming instruction, with greater emphasis placed on drill-and-practice by elementary schools and on programmming instruction by secondary schools; (7) social effects outweigh the academic effects of microcomputers; and (8) above-average students benefit more than average or below-average students from the use of microcomputers. (AOS)