Premium
Tests and prediction: A systems analysis approach
Author(s) -
Shooster Charles N.
Publication year - 1974
Publication title -
behavioral science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.371
H-Index - 45
eISSN - 1099-1743
pISSN - 0005-7940
DOI - 10.1002/bs.3830190205
Subject(s) - variance (accounting) , causality (physics) , test (biology) , path (computing) , computer science , path analysis (statistics) , line (geometry) , econometrics , mathematics , machine learning , economics , physics , paleontology , geometry , accounting , quantum mechanics , biology , programming language
Some basic principles of general systems theory were applied to the problem of test measurement and prediction. Although test measurement in the past has been primarily concerned with the contribution of person variance, it was suggested that there can be no artificial separation between the person and the operational system in which the person is imbedded, so that the contribution of system variance influencing person performance requires equal measurement emphasis. Following also from a systems analysis view, it was observed that causal effects do not follow the traditionally assumed straight‐line unidirectional path, put rather causality evidently occurs in a circular chain direction or bidirectional form. That is, there are circular cause‐effect influences coming from both the person as well as the system. This led to the conclusion that it is just as reasonable to make predictive statements from the system as from the person.