1. Theory is an invention aimed at organizing and explaining specific aspects of the environment. A major characteristic of a scientific theory----one that distinguishes it from other forms of explanantions---is that testable hypotheses may be derived from it. 

2. Popper argued persuasively that hypotheses and theories cannot be confirmed or proven, no matter how much evidence is marshaled in support of their assertions. Instead, they may be falsified. 

3. Lakatos distinguished between the "hard core" of a theory, consisting of a set of centrat principles impervious to challenge, and the "protective belt" surrounding it. Progressive adjustments are made in the theory's protective belt in attempts to withstand the onslaught of anomalies and counterinstances. 

4.Speaking of amassing data and of using fance methodology in support of trivial hypotheses, McGuire(1973) paraphrased Maslow, saying:" what is not worth doing, is not worth doing well"(p.450). 

5. In wiew of the foregoing, what is the researcher to do? To begin with, he or she would do well to heed Lynd's(1939) advice to the scientist to "continually ask himself...'Why do I pose a given problem and ask the questions I do regarding it?'"...Such self-questioning may serve as an antidote to becoming so involved with the research,the data, the fancy analyses carried out by powerful computers, so as to be blind to the triviality or vacuousness of the problem. 

6. A problem refers to relations among variables. ...Some questions refer to measurement of specific variables, not to relations between variables.

 7. It is a truism that past research on the phenomenon under investigation should play a key role in the process of problem formulation and the design of a study_---hence, the importance of a thorough review of the relevant literature. Unfortunately, what is served in the guise of a literature review is, in many instances, little more than a tiresome listing of studies and findings without even a hint that the author has been thinking when reading the literature, not to mention critically evaluating what he or she has been reading. One frequently gets the strong impression that the "review" was done after the study was completed. 

8. At the very least, a review of the literature should serve a s a safeguard against attempts to reinvent the wheel. To this end, it is crucial to break out of the almost inevitable parochillism that training and specialization engender. 

9. It would suffice to point out that, loosely speaking, meta-analysis refers "to the analysis of analyses," or "to the satistical analysis of a large collection of analysis results from individual studies for the purpose of integrating the findings. 

10. It is utmost importance to be familiar, albeit superficially, with what is going on in areas other than one's own. We recommend that you begin by systematically following annual reviews and handbooks in the anthropology. 

11. Among other things, Mcguire faulted the training of sociobehavioral scientists venturing a guess that"at least 90% of the time in our current courses on methodology is devoted to presenting ways of testing hypotheses and that little time is spent on the prior and more important process of how one creates these hypothesis in the first place. 

12. Tukey's dictum:"the important question about methods is not "How" but "Why'". we would like to caution against a conclusion that methods are of no use at the problem-formulation stage. the stage. The choice of a method(e.g., analyhsis, measurement )is obviously determined largely by one's familiarity with it.  "the man of one method or one instrument...tends to become method oriented rather than problem-oriented"(Platt). the benefits of exposure to a wide range of methods can, therefore, not be overestimated. 

13. If the independent variable consists of only two categories(e.g., Conservative, We make hypotheses about what we are going to search for, otherwise we would not find anything. Eaually undeniable is the force of hypotheses to misguide by serving as blinders, so to speak, on one's eyes and mind. Broading speaking, an hypothesis is rejected when the evidence is inconsistent with it. 

14. A P value refers to the probability of the evidence having arisen as a result of sampling errors. given that the null hypothesis is true. A P value that is deemed "small" would lead to the rejection of the null hypothesis. thus, significance testing is used to asses the evidence against the null hypothesis. 

15. A very common error, committed despite numerous warnings in the literature, is the confounding of statistical significance with substantive meaningfulness.

                            -----Measurement, Design, and analysis: an integrated approach

                                      Elazar j. Pedhazur, Liora Pedhazur Schmelkin

转载本文请联系原作者获取授权，同时请注明本文来自王芳科学网博客。
链接地址：https://blog.sciencenet.cn/blog-38036-32716.html

上一篇：国内外研究生教育的差距
下一篇：别处的宁静