PDF pending restoration
ERIC Number: ED022143
Record Type: RIE
Publication Date: 1967
Subvalidation of Language Programmes, or How to Avoid the Reinforcement of Errors.
Lack of agreement among linguists on basic points in language makes it difficult for the programer of self-instructional language texts to make definite statements concerning his subject. Programed instruction has been "too much concerned with the efficiency of the technique but it has not offered sufficient means of checking the validity of the assertions made in the program." In pointing out some of the mistakes which exist in German textbooks, the author feels that there are certain stimulus/response relationships which are particularly difficult to clarify. What is needed is basic linguistic research of a most extensive kind, and to which the language programers will have access. Good programs are needed but the commercial enterprises developing programs cannot be stopped to wait for research. As a solution to this problem, the author suggests that each program should go through a "subvalidation" process--a kind of "ad hoc minimum research." The test subjects in the validation procedure for a "German for non-Germans" program would be non-Germans; the test subjects in the subvalidation procedure for the same program would be Germans. The same applies "mutatis mutandis" for other language programs. (AMM)
Descriptors: Course Evaluation, Evaluation Methods, German, Language Programs, Material Development, Predictive Validity, Programed Instruction, Programing Problems, Stimulus Devices
Copyprints, Ltd., 87 Borough High Street, London SE 1.
Publication Type: N/A
Education Level: N/A
Authoring Institution: N/A
Note: From Programmed Learning and the Language Laboratory 2, Longmac, London.