解释性项目反应理论模型：理论与应用

doi:10.3724/SP.J.1042.2019.00937

摘要/Abstract

摘要：

解释性项目反应理论模型(Explanatory Item Response Theory Models, EIRTM)是指基于广义线性混合模型和非线性混合模型构建的项目反应理论(Item Response Theory, IRT)模型。EIRTM能在IRT模型的基础上直接加入预测变量, 从而解决各类测量问题。首先介绍EIRTM的相关概念和参数估计方法, 然后展示如何使用EIRTM处理题目位置效应、测验模式效应、题目功能差异、局部被试依赖和局部题目依赖, 接着提供实例对EIRTM的使用进行说明, 最后对EIRTM的不足之处和应用前景进行讨论。

关键词: 解释性项目反应理论, 广义线性混合模型, 非线性混合模型, 测量不变性, 解释性测量

Abstract:

Explanatory item response theory models (EIRTM) refer to a family of item response theory (IRT) models that are constructed based on the generalized linear mixed models and nonlinear mixed models. EIRTM can be utilized to address various measurement problems by incorporating predictors into IRT models. First, the relevant concepts and parameter estimation methods of EIRTM are introduced in this paper, followed by the procedures regarding how to use EIRTM to account for the item position effect, test mode effect, differential item functioning, local person dependence, and local item dependence. Next, an example is provided to illustrate the use of EIRTM. Finally, the shortcomings and potential applications of EIRTM are discussed.

Key words: explanatory item response theory, generalized linear mixed models, nonlinear mixed models, measurement invariance, explanatory measurement

中图分类号:

B841

陈冠宇, 陈平. (2019). 解释性项目反应理论模型：理论与应用. 心理科学进展 , 27(5), 937-950.

CHEN Guanyu, CHEN Ping. (2019). Explanatory item response theory models: Theory and application. Advances in Psychological Science, 27(5), 937-950.

图/表 3

参考文献 76

[1]	刘红云, 骆方 . ( 2008). 多水平项目反应理论模型在测验发展中的应用. 心理学报, 40( 1), 92-100.
[2]	聂旭刚, 陈平, 张缨斌, 何引红 . ( 2018). 题目位置效应的概念及检测. 心理科学进展, 26( 2), 368-380.
[3]	詹沛达, 王文中, 王立君 . ( 2013). 项目反应理论新进展之题组反应理论. 心理科学进展, 21( 12), 2265-2280.
[4]	Adams R. J., Wu M. L., & Wilson M. R . ( 1988). ACER ConQuest: Generalised item response modelling software [Computer software]. Melbourne, Victoria, Australia: Australian Council for Educational Research.
[5]	Baghaei P., Ravand H., . ( 2016). Modeling local item dependence in cloze and reading comprehension test items using testlet response theory. Psicologica: International Journal of Methodology and Experimental Psychology, 37( 1), 85-104.
[6]	Bates D., Mächler M., Bolker B. M., & Walker S. C ( 2015). Fitting linear mixed-effects models using LME4. Journal of Statistical Software, 67( 1), 1-48.
[7]	Bechger T. M Maris G ., ( 2015). A statistical test for differential item pair functioning. Psychometrika, 80( 2), 317-340. doi: 10.1007/s11336-014-9408-y URL
[8]	Binet A., & Simon T. , ( 1904). Méthodes nouvelles pour le diagnostic du niveau intellectuel des anormaux. L'année Psychologique, 11( 1), 191-244.
[9]	Birnbaum A., , ( 1968). Some latent trait models and their use in inferring an examinee’s ability. In F. M. Lord & M. R. Novick (Eds.), Statistical theories of mental test scores( pp. 392-479). Reading, MA: Addison-Wesley.
[10]	Bock R. D., & Aitkin M. , ( 1981). Marginal maximum likelihood estimation of item parameters: Application of an EM algorithm. Psychometrika, 46( 4), 443-459.
[11]	Bock R. D., & Lieberman M. , ( 1970). Fitting a response model for n dichotomously scored items. Psychometrika, 35(2), 179-197.
[12]	Bolker B. M., Brooks M. E., Clark C. J., Geange S. W., Poulsen J. R., Stevens M. H. H & White J. S. S ., ( 2009). Generalized linear mixed models: A practical guide for ecology and evolution. Trends in Ecology & Evolution, 24( 3), 127-135.
[13]	Bolt D. M . ( 2002). A Monte Carlo comparison of parametric and nonparametric polytomous DIF detection methods. Applied Measurement in Education, 15( 2), 113-141. doi: 10.1207/S15324818AME1502_01 URL
[14]	Cosgrove J., & Cartwright F. , ( 2014). Changes in achievement on PISA: The case of Ireland and implications for international assessment practice. Large Scale Assessments in Education, 2( 2), 1-17.
[15]	Debeer D. & Janssen R. , ( 2013). Modeling item-position effects within an IRT framework. Journal of Educational Measurement, 50( 2), 164-185. doi: 10.1111/jedm.2013.50.issue-2 URL
[16]	Debeer D., Buchholz J., Hartig J., & Janssen R . ( 2014). Student, school, and country differences in sustained test-taking effort in the 2009 PISA reading assessment. Journal of Educational and Behavioral Statistics, 39( 6), 502-523. doi: 10.3102/1076998614558485 URL
[17]	De Boeck P., Bakker M., Zwitser R., Nivard M., Hofman A., Tuerlinckx F., & Partchev I . ( 2011). The estimation of item response models with the lmer function from the lme4 package in R. Journal of Statistical Software, 39( 12), 1-28.
[18]	De Boeck P., & Wilson M. , ( 2004). Explanatory item response models: A generalized linear and nonlinear approach. New York, NY: Springer.
[19]	De Boeck P., Wilson M. R . ( 2016). Explanatory response models. In W. J. van der Linden (Ed.), Handbook of Item Response Theory, Volume One: Models( pp. 565-580). New York, NY: Chapman and Hall/CRC.
[20]	Eyre J., Berg M., Mazengarb J., & Lawes E . ( 2017). Mode equivalency in PAT: Reading comprehension. Wellington: NZCER.
[21]	Fujimoto K. A . ( 2018). A general Bayesian multilevel multidimensional IRT model for locally dependent data. British Journal of Mathematical and Statistical Psychology, 71( 3), 536-560. doi: 10.1111/bmsp.2018.71.issue-3 URL
[22]	Fukuhara H. & Kamata A. , ( 2011). A bifactor multidimensional item response theory model for differential item functioning analysis on testlet-based items. Applied Psychological Measurement, 35( 8), 604-622. doi: 10.1177/0146621611428447 URL
[23]	Gamerman D., Gonçalves F. B., Soares T. M . ( 2018). Differential item functioning. In W. J. van der Linden (Ed.), Handbook of Item Response Theory, Volume Three: Applications( pp. 67-86). New York, NY: Chapman and Hall/CRC.
[24]	Gill J . ( 2000). Generalized linear models: A unified approach (Vol. 134). Thousand Oaks, CA: Sage Publications.
[25]	Hartig J., & Buchholz J. , ( 2012). A multilevel item response model for item position effects and individual persistence. Psychological Test and Assessment Modeling, 54( 4), 418-431.
[26]	Hohensinn C., Kubinger K. D., Reif M., Schleicher E., & Khorramdel L . ( 2011). Analyzing item position effects due to test booklet design within large-scale assessment. Educational Research and Evaluation, 17( 6), 497-509. doi: 10.1080/13803611.2011.632668 URL
[27]	, , Hoskens M., & De Boeck P. , ( 1997). A parametric model for local dependence among test items. Psychological Methods, 2( 3), 261-277.
[28]	Ip E. H . ( 2000). Adjusting for information inflation due to local dependency in moderately large item clusters. Psychometrika, 65( 1), 73-91. doi: 10.1007/BF02294187 URL
[29]	Janssen R .( 2016). Linear Logistic Models. In W. J. van der Linden (Ed.), Handbook of Item Response Theory, Volume One: Models ( pp. 211-224). New York, NY: Chapman and Hall/CRC.
[30]	Jeon M., Rijmen F., & Rabe-Hesketh S . ( 2013). Modeling differential item functioning using a generalization of the multiple-group bifactor model. Journal of Educational and Behavioral Statistics, 38( 1), 32-60. doi: 10.3102/1076998611432173 URL
[31]	Jeon M., Rijmen F., & Rabe-Hesketh S . ( 2014). Flexible item response theory modeling with FLIRT. Applied Psychological Measurement, 38( 5), 404-405.
[32]	Jerrim J .( 2016). PISA 2012: How do results for the paper and computer tests compare? Assessment in Education: Principles, Policy & Practice, 23( 4), 495-518.
[33]	Jerrim J., Micklewright J., Heine J. H., Salzer C., & McKeown C . ( 2018). PISA 2015: How big is the ‘mode effect’ and what has been done about it? Oxford Review of Education, 44( 4), 476-493.
[34]	Jiao H., Kamata A., Wang S., & Jin Y . ( 2012). A multilevel testlet model for dual local dependence. Journal of Educational Measurement, 49( 1), 82-100. doi: 10.1111/jedm.2012.49.issue-1 URL
[35]	Jiao H., Kamata A. & Xie C. , ( 2015). Multilevel cross-classified testlet model for complex item and person clustering in item response data analysis. In J. R. Harring, L. M. Stapleton & S. N. Beretvas (Eds.), Advances in multilevel modeling for educational research: Addressing practical issues found in real-world applications (pp. 139-161). Charlotte, NC: Information Age Publishing Inc.
[36]	Jiao H., Wang S. D., & Kamata A . ( 2005). Modeling local item dependence with the hierarchical generalized linear model. Journal of Applied Measurement, 6( 3), 311-321.
[37]	Jiao H.,Zhang Y , ( 2015). Polytomous multilevel testlet models for testlet-based assessments with complex sampling designs. British Journal of Mathematical and Statistical Psychology, 68( 1), 65-83. doi: 10.1111/bmsp.2015.68.issue-1 URL
[38]	Jin Y.,Kang M , ( 2016). Comparing DIF methods for data with dual dependency. Large-scale Assessments in Education, 4( 1), 18. doi: 10.1186/s40536-016-0033-3 URL
[39]	Kamata A. , ( 2001). Item analysis by the hierarchical generalized linear model. Journal of Educational Measurement, 38( 1), 79-93. doi: 10.1111/jedm.2001.38.issue-1 URL
[40]	Kang C. , ( 2014). Linear and nonlinear modeling of item position effects (Unpublished master’s thesis). University of Nebraska-Lincoln.
[41]	Klein Entink R. H., Kuhn J. T., Hornke L. F., & Fox J. P . ( 2009). Evaluating cognitive theory: A joint modeling approach using responses and response times. Psychological methods, 14( 1), 54-75.
[42]	Koziol N. A . ( 2016). Parameter recovery and classification accuracy under conditions of testlet dependency: A comparison of the traditional 2PL, testlet, and bi-factor models. Applied Measurement in Education, 29( 3), 184-195. doi: 10.1080/08957347.2016.1171767 URL
[43]	Lee Y .( 2004). Examining passage-related local item dependence (LID) and measurement construct using Q3 statistics in an EFL reading comprehension test. Language Testing, 21( 1), 74-100. doi: 10.1191/0265532204lt260oa URL
[44]	Logan T . ( 2015). The influence of test mode and visuospatial ability on mathematics assessment performance. Mathematics Education Research Journal, 27(4), 423-441. doi: 10.1007/s13394-015-0143-1 URL
[45]	Mislevy R. J . ( 2016). How developments in psychology and technology challenge validity argumentation. Journal of Educational Measurement, 53( 3), 265-292.
[46]	OECD. ( 2017a). PISA 2015 technical report. Pairs: OECD Publishing.
[47]	OECD. ( 2017b). PISA 2015 assessment and analytical framework: Science, reading, mathematic, financial literacy and collaborative problem solving, Paris: OECD Publishing. Retrieved from http://dx.doiorg/10.1787/9789264281820-en.
[48]	Osterlind S. J., & Everson H. T . ( 2009). Differential item functioning (Vol. 161). Thousand Oaks, CA: Sage Publications.
[49]	Paek I., Fukuhara H . ( 2015). Estimating a DIF decomposition model using a random-weights linear logistic test model approach. Behavior Research Methods, 47( 3), 890-901. doi: 10.3758/s13428-014-0512-9 URL
[50]	Plummer M . ( 2017). JAGS version 4. 3.0 user manual [Software manual]. Retrieved from
[51]	Rabe-Hesketh S., Skrondal A . ( 2016). Generalized linear latent and mixed modeling. In W. J. van der Linden (Ed.), Handbook of Item Response Theory, Volume One: Models( pp. 503-526). New York, NY: Chapman and Hall/CRC.
[52]	Rabe-Hesketh S., Skrondal A.Pickles, & Pickles A., , ( 2004). GLLAMM manual [Software manual]. (U. C. Berkeley Division of Biostatistics Working Paper Series, 160)
[53]	Raudenbush S. W., Bryk A. S., Cheong Y. F., Congdon Jr R. T., & Toit M. D . ( 2011). HLM7 hierarchical linear and nonlinear modeling manual [Software manual]. Lincolnwood, IL: SSI Scientific Software International Inc.
[54]	Ravand H . ( 2015). Assessing testlet effect, impact, differential testlet, and item functioning using cross-classified multilevel measurement modeling. SAGE Open, 5( 2).
[55]	Rijmen F . ( 2006). BNL: A Matlab toolbox for Bayesian networks with logistic regression( Tech. Rep.). Amsterdam, the Netherlands: VU University Medical Center.
[56]	Rijmen F., Tuerlinckx F., De Boeck P., & Kuppens P . ( 2003). A nonlinear mixed model framework for item response theory. Psychological Methods, 8( 2), 185-205. doi: 10.1037/1082-989X.8.2.185 URL
[57]	SAS Institute . ( 2015). SAS/STAT 14.1: user's guide [Software manual]. Cary, NC: SAS Institute Inc.
[58]	Spiegelhalter D., Thomas A., Best N., & Lunn D. ( 2014). OpenBUGS (Version 3.2.3) [Software manual]. Retrieved from, .
[59]	Stroup W. W . ( 2012). Generalized linear mixed models: Modern concepts, methods and applications. Boca Raton, FL: CRC press.
[60]	Su Y, Yajima M ( 2015). R2jags: A Package for Running JAGS from R [Computer software]. Retrieved from
[61]	Teker G. T Dogan N ., ( 2015). The Effects of testlets on reliability and differential item functioning. Educational Sciences: Theory and Practice, 15( 4), 969-980.
[62]	Thissen D ., ( 1991). MULTILOG [Software manual]. Lincolnwood, IL: Scientific Software.
[63]	Trendtel M., Robitzsch A ., ( 2018). Modeling item position effects with a Bayesian item response model applied to PISA 2009-2015 data. Psychological Test and Assessment Modeling, 60( 2), 241-263.
[64]	Tutz G., Berger M ., ( 2016). Item-focussed trees for the identification of items in differential item functioning. Psychometrika, 81( 3), 727-750. doi: 10.1007/s11336-015-9488-3 URL
[65]	Tutz G., Schauberger G ., ( 2015). A penalty approach to differential item functioning in Rasch models. Psychometrika, 80( 1), 21-43. doi: 10.1007/s11336-013-9377-6 URL
[66]	van der Linden W.J, . ( 2016). Handbook of Item Response Theory, Volume One. New York, NY: Chapman and Hall/ CRC.
[67]	van der Linden W.J, . ( 2018). Handbook of Item Response Theory, Volume Three: Applications. New York, NY: Chapman and Hall/CRC.
[68]	Vansteelandt K, .( 2000). Formal models for contextualized personality psychology (Unpublished doctoral dissertation). K.U. Leuven, Belgium.
[69]	Wainer H., & Lukhele R. , ( 1997). How reliable are TOEFL scores? Educational and Psychological Measurement, 57( 5), 741-758. doi: 10.1177/0013164497057005002 URL
[70]	Wainer H., Sireci S. G., & Thissen D . ( 1991). Differential testlet functioning definitions and detection (Research Rep. 91-21). Princeton NJ: ETS.
[71]	Wang W. C., & Wilson M. ,( 2005). Assessment of differential item functioning in testlet-based items using the Rasch testlet model. Educational and Psychological Measurement, 65( 4), 549-576.
[72]	Weirich S., Hecht M., Böhme K . ( 2014). Modeling item position effects using generalized linear mixed models. Applied Psychological Measurement, 38( 7), 535-548. doi: 10.1177/0146621614534955 URL
[73]	Weirich S., Hecht M., Penk C., Roppelt A., Böhme K . ( 2017). Item position effects are moderated by changes in test-taking effort. Applied psychological measurement, 41( 2), 115-129. doi: 10.1177/0146621616676791 URL
[74]	Wilson M., Zheng X. H., & McGuire L . ( 2012). Formulating latent growth using an explanatory item response model approach. Journal of Applied Measurement, 13( 1), 1-22.
[75]	Xie C . ( 2014). Cross-classified modeling of dual local item dependence (Unpublished doctoral dissertation). University of Maryland, College Park, MD.
[76]	Xie C., & Jiao H. , ( 2014, April). Cross-classified modeling of dual local item dependence. Paper presented at the Annual Meeting of the American Educational Research Association, Phliadelphia, PA.

题目	行为模式	情境类型	行为类型
一辆公交车没有进站停靠, 我想诅咒。	想	他人责任	诅咒
一辆公交车没有进站停靠, 我想责备。	想	他人责任	责备
一辆公交车没有进站停靠, 我想怒骂。	想	他人责任	怒骂
因为工作人员给我错误的信息, 我错过了火车, 我想诅咒。	想	他人责任	诅咒
因为工作人员给我错误的信息, 我错过了火车, 我想责备。	想	他人责任	责备
因为工作人员给我错误的信息, 我错过了火车, 我想怒骂。	想	他人责任	怒骂
当我刚进入商店, 商店就关门了, 我想诅咒。	想	自己责任	诅咒
当我刚进入商店, 商店就关门了, 我想责备。	想	自己责任	责备
当我刚进入商店, 商店就关门了, 我想怒骂。	想	自己责任	怒骂
我与对方的通话断了, 因为我用完了话费, 我想诅咒。	想	自己责任	诅咒
我与对方的通话断了, 因为我用完了话费, 我想责备。	想	自己责任	责备
我与对方的通话断了, 因为我用完了话费, 我想怒骂。	想	自己责任	怒骂
一辆公交车没有进站停靠, 我会诅咒。	做	他人责任	诅咒
一辆公交车没有进站停靠, 我会责备。	做	他人责任	责备
一辆公交车没有进站停靠, 我会怒骂。	做	他人责任	怒骂
因为工作人员给我错误的信息, 我错过了火车, 我会诅咒。	做	他人责任	诅咒
因为工作人员给我错误的信息, 我错过了火车, 我会责备。	做	他人责任	责备
因为工作人员给我错误的信息, 我错过了火车, 我会怒骂。	做	他人责任	怒骂
当我刚进入商店, 商店就关门了, 我会诅咒。	做	自己责任	诅咒
当我刚进入商店, 商店就关门了, 我会责备。	做	自己责任	责备
当我刚进入商店, 商店就关门了, 我会怒骂。	做	自己责任	怒骂
我与对方的通话断了, 因为我用完了话费, 我会诅咒。	做	自己责任	诅咒
我与对方的通话断了, 因为我用完了话费, 我会责备。	做	自己责任	责备
我与对方的通话断了, 因为我用完了话费, 我会怒骂。	做	自己责任	怒骂

题目	行为模式	情境类型	行为类型
一辆公交车没有进站停靠, 我想诅咒。	想	他人责任	诅咒
一辆公交车没有进站停靠, 我想责备。	想	他人责任	责备
一辆公交车没有进站停靠, 我想怒骂。	想	他人责任	怒骂
因为工作人员给我错误的信息, 我错过了火车, 我想诅咒。	想	他人责任	诅咒
因为工作人员给我错误的信息, 我错过了火车, 我想责备。	想	他人责任	责备
因为工作人员给我错误的信息, 我错过了火车, 我想怒骂。	想	他人责任	怒骂
当我刚进入商店, 商店就关门了, 我想诅咒。	想	自己责任	诅咒
当我刚进入商店, 商店就关门了, 我想责备。	想	自己责任	责备
当我刚进入商店, 商店就关门了, 我想怒骂。	想	自己责任	怒骂
我与对方的通话断了, 因为我用完了话费, 我想诅咒。	想	自己责任	诅咒
我与对方的通话断了, 因为我用完了话费, 我想责备。	想	自己责任	责备
我与对方的通话断了, 因为我用完了话费, 我想怒骂。	想	自己责任	怒骂
一辆公交车没有进站停靠, 我会诅咒。	做	他人责任	诅咒
一辆公交车没有进站停靠, 我会责备。	做	他人责任	责备
一辆公交车没有进站停靠, 我会怒骂。	做	他人责任	怒骂
因为工作人员给我错误的信息, 我错过了火车, 我会诅咒。	做	他人责任	诅咒
因为工作人员给我错误的信息, 我错过了火车, 我会责备。	做	他人责任	责备
因为工作人员给我错误的信息, 我错过了火车, 我会怒骂。	做	他人责任	怒骂
当我刚进入商店, 商店就关门了, 我会诅咒。	做	自己责任	诅咒
当我刚进入商店, 商店就关门了, 我会责备。	做	自己责任	责备
当我刚进入商店, 商店就关门了, 我会怒骂。	做	自己责任	怒骂
我与对方的通话断了, 因为我用完了话费, 我会诅咒。	做	自己责任	诅咒
我与对方的通话断了, 因为我用完了话费, 我会责备。	做	自己责任	责备
我与对方的通话断了, 因为我用完了话费, 我会怒骂。	做	自己责任	怒骂

题目	模型1	模型2		模型3			模型4
题目	β_q	β_q	行为模式	β_q	DIF	95%置信区间	β_q
1	-1.162	-1.148		-1.196	-0.101	(-0.723, 0.549)	-1.248
2	-0.546	-0.531		-0.574	-0.104	(-0.717, 0.505)	-0.584
3	-0.091	-0.074		-0.134	-0.171	(-0.777, 0.431)	-0.101
4	-1.657	-1.641		-1.727	-0.261	(-0.934, 0.449)	-1.800
5	-0.681	-0.667		-0.729	-0.182	(-0.800, 0.433)	-0.746
6	-0.026	-0.011		-0.184	-0.684	(-1.293, -0.070)	-0.031
7	-0.512	-0.496		-0.495	0.103	(-0.507, 0.721)	-0.617
8	0.630	0.643		0.751	0.535	(-0.067, 1.151)	0.689
9	1.430	1.451		1.338	-0.455	(-1.153, 0.240)	1.610
10	-1.014	-0.998		-1.071	-0.221	(-0.853, 0.415)	-1.221
11	0.312	0.329		0.362	0.231	(-0.376, 0.826)	0.354
12	0.963	0.982		0.866	-0.454	(-1.104, 0.185)	1.132
13	-1.145	-1.580	-0.465	-1.066	0.426	(-0.251, 1.108)	-1.225
14	-0.383	-0.820	-0.465	-0.215	0.792	(0.156, 1.420)	-0.412
15	0.820	0.381	-0.465	0.786	-0.133	(-0.767, 0.487)	0.885
16	-0.822	-1.260	-0.465	-0.618	1.006	(0.352, 1.706)	-0.895
17	0.035	-0.404	-0.465	0.263	1.019	(0.409, 1.648)	0.042
18	1.372	0.933	-0.465	1.422	0.222	(-0.417, 0.879)	1.498
19	0.200	-0.240	-0.465	0.393	0.864	(0.280, 1.481)	0.199
20	1.390	0.956	-0.465	1.579	0.750	(0.093, 1.390)	1.563
21	2.711	2.277	-0.465	2.775	0.244	(-0.615, 1.062)	3.034
22	-0.660	-1.106	-0.465	-0.548	0.568	(-0.068, 1.205)	-0.801
23	0.363	-0.080	-0.465	0.488	0.546	(-0.059, 1.146)	0.416
24	1.867	1.427	-0.465	1.799	-0.359	(-1.138, 0.375)	2.202

题目	模型1	模型2		模型3			模型4
题目	β_q	β_q	行为模式	β_q	DIF	95%置信区间	β_q
1	-1.162	-1.148		-1.196	-0.101	(-0.723, 0.549)	-1.248
2	-0.546	-0.531		-0.574	-0.104	(-0.717, 0.505)	-0.584
3	-0.091	-0.074		-0.134	-0.171	(-0.777, 0.431)	-0.101
4	-1.657	-1.641		-1.727	-0.261	(-0.934, 0.449)	-1.800
5	-0.681	-0.667		-0.729	-0.182	(-0.800, 0.433)	-0.746
6	-0.026	-0.011		-0.184	-0.684	(-1.293, -0.070)	-0.031
7	-0.512	-0.496		-0.495	0.103	(-0.507, 0.721)	-0.617
8	0.630	0.643		0.751	0.535	(-0.067, 1.151)	0.689
9	1.430	1.451		1.338	-0.455	(-1.153, 0.240)	1.610
10	-1.014	-0.998		-1.071	-0.221	(-0.853, 0.415)	-1.221
11	0.312	0.329		0.362	0.231	(-0.376, 0.826)	0.354
12	0.963	0.982		0.866	-0.454	(-1.104, 0.185)	1.132
13	-1.145	-1.580	-0.465	-1.066	0.426	(-0.251, 1.108)	-1.225
14	-0.383	-0.820	-0.465	-0.215	0.792	(0.156, 1.420)	-0.412
15	0.820	0.381	-0.465	0.786	-0.133	(-0.767, 0.487)	0.885
16	-0.822	-1.260	-0.465	-0.618	1.006	(0.352, 1.706)	-0.895
17	0.035	-0.404	-0.465	0.263	1.019	(0.409, 1.648)	0.042
18	1.372	0.933	-0.465	1.422	0.222	(-0.417, 0.879)	1.498
19	0.200	-0.240	-0.465	0.393	0.864	(0.280, 1.481)	0.199
20	1.390	0.956	-0.465	1.579	0.750	(0.093, 1.390)	1.563
21	2.711	2.277	-0.465	2.775	0.244	(-0.615, 1.062)	3.034
22	-0.660	-1.106	-0.465	-0.548	0.568	(-0.068, 1.205)	-0.801
23	0.363	-0.080	-0.465	0.488	0.546	(-0.059, 1.146)	0.416
24	1.867	1.427	-0.465	1.799	-0.359	(-1.138, 0.375)	2.202