- Problem selection and definition is the responsibility of the individual doing the research
- Tentative answers (hypotheses) to the problem may be derived by deduction from theories or by induction from an organized body of knowladge.
- Value judgments by the researcher are limited to those implicit in the selection of the problem
- Given the statement of the problem and the hypothesis, the research can be replicated
- The data to be collected are determined largely by the problem and the hypothesis
- Relevant variable can be controlled or manipulated, and systematic effects of other variables can be eliminated by randomization
- The problem is almost completely determined by the situation in which the study is conducted. Many people may be involved in its definition and , because of its comlexity, the problem initially is difficult to define.
- Precise hypotheses usually cannot be generated ; rather the task becomes one of testing generalizations from a variety of research studies, some of which are basically contradictory. There are many gaps which in the absence of verified knowledge must be filled by reliance on judgment and experience
- Value judgments are made explicit in the selection and the definition of problem as well as in the development and implementation of the procedures of the study.
- The study is unique to a situation and seldom can be replicated, even approximately
- The data to be collected are heavily influenced if not determined by feasibility. choices, when possible, reflect value judgments of decision-makers or of those who set policy. There are often large differences between data for which the collection is feasible and data which are of most value to the decision-makers.
- Only superficial control of a multitude of variables important to interpretation of results is possible. Randomization to eliminate the systematic effects of these variables is extremely difficult or impractical to accomplish.
Makalah dari : John K. Hemphill
Halaman. 190 s.d. 191
