【正文】 S ) D. Procedures2) Guard against threats to reliability and validity (Nunan. 16) A. Reliability A) Definition: refers to the consistency of the results obtained from a piece of research internal B) Types: external a. Internal reliability —— consistency of data collection, analysis, and interpretation b. A case of high external reliability (Nunan, 1992, ) c. External reliability —— the extent to which independent researchers can reproduce a study and obtain results similar to those obtained in the original study.B. Validity A) Definition: the extent to which a piece of research actually investigates what the researcher purports to investigate. internalB) Types of validity validity externalb. Factors affecting internal validity ① Subject variability (Samp。S. ) A. Determine the data collection methods B. Define the variables (Samp。 the .01 level indicates a less than l percent chance.( 3) Select an appropriate statistical method A. T—test a. one treatment with 2 groups or one test with 2 groups b. one group with two treatments or one group with two tests B. Analysis of variance (ANOVA) Comparing more than two means, or more than two groups C. Correlation D. Chi—square (4) Compare the observed statistics and critical values A. Observed statistics: The result of the calculations, the actual figures found for the sample (Brown, 1988: 117) B. Degree of freedom: the number of observations that are free to vary, used in determining levels of probability (Charles, C. M. amp。（見馬廣惠，2003：79） B. Alpha level: The probability level (significance level) that a researcher establishes as acceptable before conducting a test of significance.C. P＜.05。 is used to establish levels of confidence related to individual measures and hypothesis testing (Charles, C. M. amp。 they participate in the very act that they are describing.3) Open observation and structured observation n In open observation, the data being recorded are broad and more general. n In structured observation, the researcher uses tools (instruments) to collect data.3. Uses of observation1) To observe how learners acquire or use language. (龔少瑜、陳永培，1997）2) To study the interaction between teachers and students.3) To find out how an experiment is going on.4) To find out the effect of different classroom practices on learning oute.4. Observation scheme COLT 5. Case studies n the systematic investigation of an individual ‘case’, whether that refers to one teacher, one learner, one group, one class, or whatever.6. Methods to collect data in case studiesn a wide range of techniques can be used.1) Observation 2) Questionnaire and interview 3) Verbal report 4) Tests Ⅶ. Experimental Method 1. Definitionn an approach to educational research in which an idea or hypothesis is tested or verified by setting up situations in which the relationship between different subjects or variables can be determined. The description of the purposes of the research, its plan, the statistical procedures used etc., in an experimental study is called the experimental design ( Richards et al, 1998)2. The major ponents of experimental research1) The type and number of groups(1) “Natural” groups(2) Groups formed specially for an experiment2) The treatmentn This refers to anything done to groups in order to measure its effect, e. g. a language teaching method, teaching materialsn Treatments are the independent variables in the research 3) Measurement or observation n refers to how the effects of the treatment will be evaluated or observed. X= treatment O= observation or measurement R= randomization3. Single group design 1) One shot design: XO2) One group pretest + posttest design n O1 X O23) Time series designn O1, O2, O3, O4, X, On+1, On+24. Design using control groups 1) Static group parison design: X O1 O12) Non – equivalent control group design: O1 O2 O1 O23) Pretest / posttest with randomization of groups: R O1 X O2 R O1 O2 ↓ R X O2 R O24) Variation of 3) R O2 R O25. Factorial Designs（因子設(shè)計）beginner group R O1 Xc O2 R O1 O2 beginners1—O1—Xd—O2 R beginners2—O1—Xc—O2 R O1 Xd O2 beginners3—O1 —— O2 R O1 O2advanced group R O1—Xc—O2 R O1 O2 advanced1—O1—Xd—O2 R advanced2—O1—Xc—O2 R O1—Xd—O2 advanced3—O1 —— O2 R O1 O26. Types of experiments 1) Pre—experiment n May have pre and post treatment tests, but lacks a control group2) Quasi—experiment n has both pre and post tests and experimental group and control group, but no random assignment of subjects3) True experimentn has both pretest and posttest, experimental and control groups, and random assignment of subjectsⅧ. Action Research1. Definition Action research is a form of selfreflective enquiry undertaken by participants in social (including educational) situation in order to improve the rationality and justice of (a) their own social or educational practices, (b) their understanding of these practices, and (c) the situations (and institutions in which these practices are carried out. (Kemmis and Mc Taggart, 1982:5)2. Characteristics of action research (Nunan, 1992:1718)(1) It is carried out by practitioners. (2) It is collaborative. (3) It is aimed at changing things.3. Steps in the action research cy