There is no way to decide how to combine these elements; instead it is necessary to develop a logical "inference chain" (Lesh, Lovitts and and Kelly, 2000). This set of reasoning should be consistent, clear (can be copied by other researchers), and must be persuasive to skeptical readers (eg, such counter-facts are solved).
Both quantitative and qualitative exact studies embody the same reasoning logic (King, Keohane and Verba, 1994). The inference of this reasoning is supported by a clear statement of how to reach the conclusion of this study. How can I determine if the evidence is valid? How to consider or abandon alternate descriptions How is the relationship between data and conceptual or theoretical framework formed?
The nature of this series of reasoning depends on the design of the research, and the design of the research also depends on the problem to be investigated. Does the research develop, extend, modify or test the hypothesis? Its purpose is to decide. How does it work? Under what circumstances is it effective? If the purpose of the study is to validate the hypotheses mentioned in the form of "if-then" rules, the success of reasoning may depend on the extent to which the measurement rules predict the outcome under various conditions there is. If the objective is to create a description of a complex system such as an intracellular organelle or a hierarchical social organization, the success of reasoning is dependent on the fidelity and internal consistency of the observation method applied to the various components, It depends on the reliability of it. collected. Inference of research design and its reasoning must demonstrate the subtle nature of the questions asked and a thorough understanding of the procedures used to answer them.
For example, Muller (1994) collected data on orbital sinks of the earth over 100,000 years and combined it with the emergence of ice age. However, we excluded the possibility of orbital eccentricity as a cause of ice age and estimated the orbit of the earth. Rebound can lead to the Ice Age (see box 3-2). Putnam finds that various methods are used to rigorously test his hypothesis about the impact of democratic system success in rigorous testing in various social environments and the importance of evidence is supported It was.
Many of the disputes surrounding educational research design have focused on whether the research is scientific, as determined by the National Research Council (NRC) Science Education Research (2002). Discussions will center on the use of scientific methods to study educational problems. The position of the NRC is generally supported by some people (Feuer, Towne, & Shavelson, 2002; Slavin, 2002) and has been warned or rejected (Berliner, 2002; Erickson & Gutierrez, 2002; Olson, 2004; St. Pierre , 2002). Research design, research funding and politics are interrelated (Burkhardt & Schoenfeld, 2003). The Obama administration has done a lot of work to restore the importance of scientific knowledge in policy making, but people can never think that this change is permanent.
Universality is the central foundation of scientific research in teaching methods. In the report of the National Research Council in 2002, the editor said that the "regularity of intergroup models and time crossing models" is the cause of generalization rather than replication itself. Of course, the purpose of this scientific approach has not changed. It is to specify a generalized model (p. Generalization is a powerful statistical tool that enables researchers to predict patterns of behavior within groups based on measures of behavior obtained from demographic samples, such as the percentage of people voting as independent groups is. It is attractive to policy makers because it shows that specific solutions are ubiquitous in the population. The value of generalization is more limited as related actions become more complex as the way students learn ethical behavior. Lincoln and Guba (1985) explained well about this limitation.