Premium
XLII. The architecture of cross‐sectional research (Part I)
Author(s) -
Feinstein Alvan R.
Publication year - 1978
Publication title -
clinical pharmacology and therapeutics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.941
H-Index - 188
eISSN - 1532-6535
pISSN - 0009-9236
DOI - 10.1002/cpt197823181
Subject(s) - computer science , experimental data , observer (physics) , test (biology) , management science , data science , statistical hypothesis testing , statistics , mathematics , engineering , paleontology , physics , quantum mechanics , biology
Despite the many mathematical ‘models’ that have been developed for the statistical analysis of a research project, relatively little attention has been given to the basic scientific ‘models’ on which the work depends. For most people who have learned about science and the ‘scientific method’, the fundamental ‘model’ of scientific research is an experiment. This ‘model’ has many obvious advantages, but it is not applicable for many types of research that appear in clinical and epidemiologic journals. The investigator may be interested in studying observer variability or diagnostic test procedures for which an experimental model is inappropriate. Alternatively, the investigator may want to explore etiologic possibilities or therapeutic events for which an experimental model is unfeasible.