臺灣博碩士論文加值系統

測驗公平性是測驗施測基礎，也是檢視測驗品質的重要指標，因此差異試題功能（differential item functioning, DIF）的檢測就顯得格外重要。當進行DIF檢測時，如果資料部分遺失，DIF檢測方法便會受到影響；如果遺失比率過高，甚至會使DIF檢測無法運作，因此如何處理遺失值以確保DIF檢測結果是可信賴的，這是研究者們所必須面對的重要課題；再者，由於不同遺失值處理方法能夠回復資料的狀態也不盡相同，後續也可能影響DIF檢測結果。Finch (2011)指出，採用多重插補法處理遺失資料具有優勢，然而該研究只考慮第一道試題遺失的情境，未考慮DIF試題比率、遺失比率等因素對於DIF檢測結果的影響，故本研究目的欲瞭解這些因素如何影響遺失值對於DIF檢測的結果，期能為資料處理時可能會遇到的問題提供較好的處理方法。本研究分為兩部分探討，研究一是探討不同遺失值處理方法在面對不同遺失比率、遺失機制、DIF比率和DIF量時資料回復的效果，研究二是探討經不同遺失處理方法處理後，比較不同DIF檢測方法的差異。研究結果顯示，通過觀察準確率可以發現，操弄遺失機制對資料回復的影響最大，其中又以完全隨機遺失機制在資料回復表現最好，隨機遺失與非隨機遺失機制表現較差，除此之外，遺失機制的不同也會造成遺失比率對資料回復有所影響，在遺失資料處理方法方面，分類與回歸樹、預測平均數匹配與多重插補法三種方法在資料回復的效果上表現最佳，並且在不同遺失機制上保有一定的資料回復能力；在DIF檢測方面，Lord法在DIF比率10%至20%時型一錯誤率符合模式預期，MH法則是在DIF比率10%以下時型一錯誤率符合模式預期，兩種方法皆會隨著DIF比率上升，而使統計檢定力下降且型一錯誤率上升，兩種方法差異僅在適用的DIF比率並不相同，但在統計檢定力上，Lord法表現較低，未來研究建議使用分類與回歸樹、預測平均數匹配與多重插補法三種方法，DIF比率10%以下時使用MH法，DIF比率10%至20%時使用Lord法。

The fairness of a test is the basis for administering a test, and also a critical indicator for examining test quality. Therefore, the detection of differential item functioning (DIF) is particularly crucial. Missing data may affect the DIF detection method. If the data missing rate is too high, the DIF detection cannot work. Therefore, how to handle missing values to ensure that the DIF detection results are reliable is a crucial topic for researchers. Because different methods of handling missing values result in different state of restored data, they may subsequently affect the DIF detection results. Finch (2011) reported that using multiple imputation to handle missing data is advantageous. However, that research only considered the scenario of first item missing, not considering the factors of DIF item ration or data missing ratio on DIF detection results. Therefore, this study aimed to understand how these factors affected missing values on the DIF detection results. The results of this study may provide a more favorable handling method for problems encountered when handling data. This study was divided into two parts. Part I investigated the data restore effects using various methods of handling missing values when facing different missing ratios, missing mechanisms, DIF ratios, and magnitudes of DIF. Part II investigated the differences in using different DIF detection methods after undergoing different missing handling methods. Observations of data accuracy revealed that manipulating missing data mechanisms had the greatest effect on data restoration. Among them, the missing completely at random mechanism demonstrated optimal performance in data restoration, superior to that of missing at random and not missing at random mechanisms. In addition, different missing mechanisms resulted in influenced on missing ratio on data restore. Regarding missing data handling methods, three methods, namely classification and regression tree, predictive-mean matching, and multiple imputation, demonstrated optimal performance in data restoration. These methods maintained certain data restoration functions under different missing mechanisms. Regarding DIF detection, Lord method met the model expectation of type I error when the DIF ratio was 10%–20%. The MH method met the model expectation of type I error when the DIF ratio was 10% or lower. As DIF ratio increased, the statistical power of both methods decreased, and their type I error rates increased. These two methods only differed in the DIF ratio to which they were suitable. Regarding statistical power, Lord method was inferior. Future studies are advised to use classification and regression trees, predictive-mean matching, and multiple imputation. If the DIF ratio is 10% or lower, the MH method should be used; if it is 10%–20%, the Lord method should be applied.

摘　　要
Abstract
目次
表次
第一章　緒論
第一節　研究動機與目的
第二節　研究問題
第二章　文獻探討
第一節　遺失機制
第二節　遺失資料處理方法
第三節　DIF檢測方法
第四節　影響DIF檢測因素
第三章　研究方法
第一節　操弄變項
第二節　研究程序
第四章　研究結果與討論
第一節、不同遺失處理方法是否在不同模擬情境下有所差異
第二節、不同遺失處理方法是否影響DIF檢測效果
第五章　結論與建議
第一節　結論
第二節　研究限制與建議
參考文獻

Ackerman, T. A. (1992). A didactic explanation of item bias, item impact, and item validity from a multidimensional perspective. Journal of Educational Measurement, 29(1), 67–91. https://doi.org/10.1111/j.1745-3984.1992.tb00368.x
Allison, P. D. (2000). Multiple imputation for missing data: A cautionary tale. Sociological Methods & Research, 28(3), 301–309. https://doi.org/10.1177/0049124100028003003
Allison, P. D. (2002). Quantitative applications in the social sciences: Missing data. Thousand Oaks, CA: SAGE Publications.
Awuor, R. (2008). Effect of unequal sample sizes on the power of DIF detection: An IRT-based monte carlo study with SIBTEST and Mantel-Haenszel procedures (Unpublished doctoral dissertation). Virginia Polytechnic Institute and State University, U.S.A.
Banks, K. (2015). An introduction to missing data in the context of differential item functioning. Practical Assessment, Research & Evaluation. 20(12), 1-10. https://doi.org/10.7275/fpg0-5079
Batista, G. E., & Monard, M. C. (2003). An analysis of four missing data treatment methods for supervised learning. Applied Artificial Intelligence, 17, 519-533. https://doi.org/10.1080/713827181
Birnbaum, A. (1968). Some latent trait models and their use in inferring an examinee's ability. In Lord, F.M. & Novick, M.R., Statistical theories of mental test scores (pp. 397–479). Reading, Mass.: Addison-Wesley.
Breiman, L., Friedman, J. H., Olshen, R. A., & Stone, C. J. (1984). Classification and regression trees. Monterey, CA: Wadsworth and Brooks.
Burgette, L. F., & Reiter, J. P. (2010). Multiple imputation for missing data via sequential regression trees. American Journal of Epidemiology, 172(9), 1070–1076. https://doi.org/10.1093/aje/kwq260
Dempster, A., Laird, N., & Rubin, D. (1977). Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society. Series B (Methodological), 39(1), 1-38.
Finch, H. (2011). The use of multiple imputation for missing data in uniform DIF analysis: Power and type I error rates. Applied Measurement in Education, 24(4), 281-301. https://doi.org/10.1080/08957347.2011.607054
Fix, E., & Hodges, J. L. (1951). Discriminatory analysis: nonparametric discrimination, consistency properties. Randolph Field, Tx.: USAF School of Aviation Medicine.
Gower, J. (1971). A general coefficient of similarity and some of its properties. Biometrics, 27(4), 857-871. https://doi.org/10.2307/2528823
Hambleton, R. K., & Rogers, H. J. (1989). Detecting potentially biased test items: Comparison of IRT area and Mantel-Haenszel methods. Applied Measurement in Education, 2(4), 313-334. https://doi.org/10.1207/s15324818ame0204_4
Holland, P. W., & Thayer, D. T. (1988). Differential item performance and the Mantel-Haenszel procedure. In H. Wainer & H. I. Braun (Eds.), Test validity (pp. 129–145). Hillsdale, NJ: Lawrence Erlbaum Associates.
Holland, P. W., & Wainer, H. (1993). Differential item functioning. Hillsdale, NJ: Lawrence Erlbaum Associates.
Jönsson, P., & Wohlin, C. (2006). Benchmarking K-nearest neighbour imputation with homogeneous Likert data. Empirical Software Engineering, 11(3), 463-489. https://doi.org/10.1007/s10664-006-9001-9
Kleinke, K. (2017). Multiple imputation under violated distributional assumptions: A systematic evaluation of the assumed robustness of predictive mean matching. Journal of Educational and Behavioral Statistics, 42(4), 371–404. https://doi.org/10.3102/1076998616687084
Little, R. J. A. (1988). A Test of Missing Completely at Random for Multivariate Data with Missing Values. Journal of the American Statistical Association, 83(404), 1198-1202. https://doi.org/10.2307/2290157
Little, R. J. A., & Rubin, D. B. (2002). Statistical analysis with missing data. Hoboken, NJ: John Wiley & Sons.
Lord, F.M. (1980). Application of item response theory to practical testing problems. Hillsdale, NJ: Lawrence Erlbaum Associates.
Magis, D., & De Boeck, P. (2014). Type I error inflation in DIF identification with Mantel-Haenszel: An explanation and a solution. Educational and Psychological Measurement, 74(4), 713-728. https://doi.org/10.1177/0013164413516855
Marshall, A., Altman, D.G. & Holder, R.L. (2010). Comparison of imputation methods for handling missing covariate data when fitting a cox proportional hazards model: A resampling study. BMC Medical Research Methodology, 10, 112. https://doi.org/10.1186/1471-2288-10-112
Mantel, N., & Haenszel, W. (1959). Statistical aspects of the analysis of data from retrospective studies of disease. Journal of the National Cancer Institute, 22(4), 719-748. https://doi.org/10.1093/jnci/22.4.719
Mellenbergh, G. J. (1982). Contingency table models for assessing item bias. Journal of Educational Statistics, 7(2), 105–118. https://doi.org/10.3102/10769986007002105
Morris, T. P., White, I. R., & Royston, P. (2014). Tuning multiple imputation by predictive mean matching and local residual draws. BMC Medical Research Methodology, 14(1), 75. https://doi.org/10.1186/1471-2288-14-75
Narayanan, P., & Swaminathan, H. (1994). Performance of the Mantel-Haenszel and simultaneous item bias procedures for detecting differential item functioning. Applied Psychological Measurement, 18(4), 315–328. https://doi.org/10.1177/014662169401800403
Peng, C. Y., Harwell, M. R., Liou, S. M., & Ehman, L. H. (2006). Advances in missing data methods and implications for educational research. In S. S. Sawilowsky (Ed.), Real Data Analysis (pp. 31-78). Charlotte, NC: New Information Age.
Peng C.-Y. J., & Zhu J. (2008). Comparison of two approaches for handling missing covariates in logistic regression. Educational and Psychological Measurement, 68(1), 58-77. https://doi.org/10.1177/0013164407305582
Peugh, J. L., & Enders, C. K. (2004). Missing data in educational research: A review of reporting practices and suggestions for improvement. Review of Educational Research, 74(4), 525–556. https://doi.org/10.3102/00346543074004525
Rasch, G. (1960). Studies in mathematical psychology: I. Probabilistic models for some intelligence and attainment tests. Oxford, EN: Nielsen & Lydiche.
Robitzsch, A., & Rupp, A. A. (2009). Impact of missing data on the detection of differential item functioning: The case of Mantel-Haenszel and logistic regression analysis. Educational and Psychological Measurement, 69(1), 18-34. https://doi.org/10.1177/0013164408318756
Rubin, D. B. (1976). Inference and missing data. Biometrika, 63(3), 581-592. https://doi.org/10.1093/biomet/63.3.581
Rubin, D. B. (1987). Multiple imputation for nonresponse in surveys. New York, NY: John Wiley & Sons.
Schafer, J. L. (1997). Analysis of incomplete multivariate data. London, UK: Chapman & Hall.
Schafer, J. L. (1999). Multiple imputation: A primer. Statistical Methods in Medical Research, 8(1), 3-15. https://doi.org/10.1177/096228029900800102
Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art. Psychological Methods, 7(2), 147–177. https://doi.org/10.1037/1082-989X.7.2.147
Shealy, R., & Stout, W. (1993). A model-based standardization approach that separates true bias/DIF from group ability differences and detects test bias/DTF as well as item bias/DIF. Psychometrika, 58(2), 159–194. https://doi.org/10.1007/BF02294572
Swaminathan, H., & Rogers, H. J. (1990). Detecting differential item functioning using logistic regression procedures. Journal of Educational Measurement, 27(4), 361-370. https://doi.org/10.1111/j.1745-3984.1990.tb00754.x
Thissen, D., Steinberg, L., & Wainer, H. (1993). Detection of differential item functioning using the parameters of item response models. In P. W. Holland & H. Wainer (Eds.), Differential item functioning (pp. 67–113). Hillsdale, NJ: Lawrence Erlbaum Associates.
Troyanskaya, O., Cantor, M., Sherlock, G., Brown, P., Hastie, T., Tibshirani, R., …Altman, R. (2001). Missing value estimation methods for DNA microarrays. Bioinformatics, 17(6), 520–525. https://doi.org/10.1093/bioinformatics/17.6.520
Wang, W., & Su, Y. (2004). Effects of average signed area between two item characteristic curves and test purification procedures on the DIF detection via the Mantel-Haenszel method. Applied Measurement in Education, 17(2), 113 - 144.