Transfer of LOGO Debugging Skill: Analysis, Instruction, and Assessment.

The goal of thls dissertation is to determine the extent to which learning debugging in the context of LOGO programming improves children's debugging in other programming and nonprogramming contexts. The approach involves detailed task analysis of debugging (in the form of a computer simulation model), development of model-based instructional guidelines for teaching children debugging skills they do not learn "by discovery." and assessment of the debugging skills children are able to transfer to other programming and non-programming tasks. Twenty-tWo 8to I1-year-old students took two 25 hour LOGO courses. Halt of the students were taught debugging In the context of a LOGO graphics course first and then a LOGO list-processing course. The other half were taught debugging in the same two minicourses but In the reverse order. Debugging skills were tested at three times during each mini-course. The performance of children taking tests In the first mfrtl-course was compared with the performance of children taking the same tests in the seconFrmlni-course to reveal the transfer from one LOGO domain to the other. Debugging on no44-pregramming tasks was assessed prior to the first mini-course, between minl-courses, and after the last minicourse to assess more remote transfer of debugging skills. Assessments of students' debugging skill revealed large savings from the first to the second mini-course. Students' Increasing use of selective search strategies increased the accuracy. efficiency, and :peed of their debugging. Corresponding Improvements were demonstrated on a variety of tasks requiring debugging of non-computer directions. Children shifted from exhaustive to selectiVe search strategies which increased the accuracy. efficiency, and speed with which they debugged written directions; Thus; the debugging strategies learned from explicit Instruction in the the first computer programming mini-course had a positive impact on children's debugging in both new programming and non-programming situations.

[1]  Gary M. Olson,et al.  Comprehension differences in debugging by skilled and novice programmers , 1986 .

[2]  Allen Newell,et al.  Human Problem Solving. , 1973 .

[3]  Diane O. Cuneo.  Young Children and Turtle Graphics Programming: Understanding Turtle Commands. , 1985 .

[4]  Ellin Kofsky Scholnick,et al.  Initial Mastery of the Syntax and Semantics of Logo Positioning Commands , 1986 .

[5]  Seymour Papert,et al.  Mindstorms: Children, Computers, and Powerful Ideas , 1981 .

[6]  K. Holyoak,et al.  Pragmatic versus syntactic approaches to training deductive reasoning , 1986, Cognitive Psychology.

[7]  John Seely Brown,et al.  Diagnostic Models for Procedural Bugs in Basic Mathematical Skills , 1978, Cogn. Sci..

[8]  K. A. Ericsson,et al.  Protocol Analysis: Verbal Reports as Data , 1984 .

[9]  Elliot Soloway,et al.  Analyzing the high frequency bugs in novice programs , 1986 .

[10]  Douglas Degelman,et al.  Concept Learning in Preschool Children: Effects of a Short-Term Logo Experience , 1986 .

[11]  D. Gentner,et al.  Flowing waters or teeming crowds: Mental models of electricity , 1982 .

[12]  John R. Anderson,et al.  Production systems, learning, and tutoring , 1987 .

[13]  E. Thorndike.  Educational psychology, Vol 2: The psychology of learning. , 1913 .

[14]  D. Clements,et al.  Effects of Computer Programming on Young Children's Cognition , 1984 .

[15]  John R. Anderson,et al.  Cognitive principles in the design of computer tutors , 1984 .

[16]  M. C. Linn,et al.  The Gap between Promise and Reality in Computer Education: Planning a Response. , 1984 .