中国人文社会科学核心期刊

中文社会科学引文索引(CSSCI)来源期刊

中文核心期刊

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

非连续性与异质性——多阶段混合增长模型在语言发展研究中的应用

刘源 刘红云

刘源, 刘红云. 非连续性与异质性——多阶段混合增长模型在语言发展研究中的应用[J]. 华东师范大学学报(教育科学版), 2018, 36(1): 137-148+166. doi: 10.16382/j.cnki.1000-5560.2018.01.017
引用本文: 刘源, 刘红云. 非连续性与异质性——多阶段混合增长模型在语言发展研究中的应用[J]. 华东师范大学学报(教育科学版), 2018, 36(1): 137-148+166. doi: 10.16382/j.cnki.1000-5560.2018.01.017
LIU Yuan, LIU Hongyun. Non-Continuity and Heterogeneity: Application of Piecewise Growth Mixture Model in Language Development Study[J]. Journal of East China Normal University (Educational Sciences), 2018, 36(1): 137-148+166. doi: 10.16382/j.cnki.1000-5560.2018.01.017
Citation: LIU Yuan, LIU Hongyun. Non-Continuity and Heterogeneity: Application of Piecewise Growth Mixture Model in Language Development Study[J]. Journal of East China Normal University (Educational Sciences), 2018, 36(1): 137-148+166. doi: 10.16382/j.cnki.1000-5560.2018.01.017

非连续性与异质性——多阶段混合增长模型在语言发展研究中的应用

doi: 10.16382/j.cnki.1000-5560.2018.01.017
基金项目: 

中央高校基本科研业务费专项资金资助 SWU1709379

Non-Continuity and Heterogeneity: Application of Piecewise Growth Mixture Model in Language Development Study

  • 摘要: 多阶段混合增长模型(Piecewise Growth Mixture Modeling,PGMM)是近几年新兴的同时关注群体的发展阶段非连续性和潜在异质性的统计模型。它将多阶段增长模型和潜类别增长模型进行整合,可以描述同时存在发展转折点和不同发展类别的描述群体增长趋势的数据。文章以早期儿童的追踪研究(幼儿园版)为例,运用PGMM模型探索其增长趋势,得出:(1)两阶段混合增长模型能最有效地描述学生阅读能力的发展,转折点在一年级,随着年龄的增加,发展速度变慢;(2)发展趋势分为三类,大部分个体起点低、发展快,小部分个体起点高、发展慢,到三年级以后两个类别差距越来越小,另一部分整体发展都比较缓慢;(3)教师对学生行为的评价比父母的评价更能有效预测学生阅读成绩的类别和趋势。
  • 图  1  两阶段混合增长模型

    图  2  不同类别混合增长模型BIC的变化

    图  3  阅读能力发展类别及趋势

    表  1  六次阅读成绩以及协变量的相关矩阵与均值、标准差

    阅读1 阅读2 阅读4 阅读5 阅读6 阅读7 SES 父母评价 教师评价
    阅读2 .793**
    阅读4 .669** .777**
    阅读5 .603** .657** .762**
    阅读6 .584** .639** .721** .852**
    阅读7 .542** .559** .604** .746** .795**
    SES .147** .132** .129** .157** .162** .178**
    父母评价 -.019** -.014** -.028** -.012** -.027** -.024** .540**
    教师评价 .269** .264** .244** .240** .233** .204** .092** .050**
    M -1.277 -0.715 0.128 0.798 1.046 1.293 0.014 3.492 3.086
    SD 0.506 0.491 0.453 0.315 0.299 0.383 0.789 1.505 0.960
    ** p<.01
    下载: 导出CSV

    表  2  四个备选模型与数据的整体拟合

    模型 χ2 df CFI RMSEA AIC BIC ABIC
    模型1 52797.438 16 0.000 0.612 56593.443 56670.016 56635.060
    模型2 18906.476 15 0.433 0.449 26187.379 26291.796 26244.129
    模型3 11728.063 12 0.648 0.354 17145.410 17249.827 17202.160
    模型4 4415.809 7 0.868 0.284 8426.164 8565.386 8501.830
    下载: 导出CSV

    表  3  零模型(模型4)参数估计结果

    固定部分估计值 SE t 随机部分方差 SE χ2
    截距 -1.715 0.008 -226.817*** 0.337 0.008 43.815***
    线性斜率1 0.960 0.004 257.225*** 0.017 0.002 10.318***
    线性斜率2 0.368 0.003 143.902*** 0.004 0.001 4.646***
    曲线斜率2 -0.022 <0.001 -94.611*** <0.001 <0.001 <0.001
    *** p<.001
    下载: 导出CSV

    表  4  不同发展组的增长曲线参数估计结果

    缓慢发展组 中等发展组 快速发展组
    估计值 SE t 估计值 SE t 估计值 SE t
    截距 -1.768 0.048 -36.914*** -2.416 0.054 -44.821*** -1.236 0.105 -11.783***
    线性斜率1 0.977 0.013 76.282*** 0.873 0.05 17.327*** 0.983 0.028 35.220***
    线性斜率2 0.378 0.014 26.955*** 0.530 0.02 25.883*** 0.256 0.024 10.575***
    曲线斜率2 -0.023 0.001 -20.330*** -0.034 0.002 -20.741*** -0.013 0.002 -5.540***
    *** p<.001
    下载: 导出CSV

    表  5  三个不同类别上预测变量的描述统计结果

    中等发展组 缓慢发展组 快速发展组
    频次 百分比 频次 百分比 频次 百分比
    性别(男) 2970 73.4% 613 15.2% 463 11.4%
    性别(女) 3040 81.0% 282 7.5% 432 11.5%
    家庭语言非英语 663 74.9% 167 18.9% 55 6.2%
    家庭语言英语 5173 77.4% 686 10.3% 825 12.3%
    M SD M SD M SD
    年龄(月) 68.446 4.136 67.700 4.649 69.378 4.076
    社会经济地位 0.270 1.661 -0.042 2.063 0.593 1.357
    父母评价 3.488 1.475 3.626 2.067 3.383 0.911
    教师评价 3.034 0.639 2.489 0.627 3.299 0.580
    下载: 导出CSV

    表  6  阅读能力发展类别预测的参数估计结果

    标准化系数 SE Wald检验值 p 发生比
    缓慢发展组
    截距 2.942 0.665 19.591 <.001
    社会经济地位 -0.704 0.057 150.117 <.001 0.495
    年龄 -0.037 0.01 14.024 <.001 0.964
    父母评价 0.020 0.028 0.535 .464 1.020
    教师评价 -1.080 0.066 269.708 <.001 0.340
    性别 0.532 0.086 38.741 <.001 1.703
    家庭语言 0.305 0.109 7.805 .005 1.357
    快速发展组
    截距 -6.83 0.647 111.434 <.001
    社会经济地位 0.633 0.049 167.602 <.001 1.884
    年龄 0.043 0.009 22.539 <.001 1.044
    父母评价 -0.059 0.043 1.89 .169 0.942
    教师评价 0.616 0.066 86.118 <.001 1.852
    性别 0.175 0.077 5.195 .023 1.192
    家庭语言 -0.251 0.154 2.666 .102 0.778
    注:对于阅读能力发展,参考组为中间组;对于分类预测变量,性别的参照组为女生,家庭交流所使用语言的参照组为使用英语组。
    下载: 导出CSV

    表  7  阅读能力发展水平和速度影响因素分析结果

    缓慢发展组 中等发展组 快速发展组
    标准化系数 t 标准化系数 t 标准化系数 t
    初始水平(截距)
    性别 -.138 -4.184 -.014 -1.138 -.017 -.467
    家庭语言 .052 1.599 .133 10.382 .012 .337
    初始年龄 .243 7.322 .142 11.239 .139 3.847
    社会经济地位 .340 10.424 .203 15.583 .192 5.156
    父母评价 .010 .302 .006 .513 .177 4.944
    教师评价 .133 4.025 .296 22.786 .146 4.096
    第一阶段斜率
    性别 .140 4.163 .044 3.170 -.080 -2.280
    家庭语言 -.017 -.520 -.039 -2.767 -.060 -1.690
    初始年龄 -.202 -5.980 -.139 -10.151 -.208 -5.973
    社会经济地位 -.341 -10.273 -.108 -7.610 .303 8.495
    父母评价 .019 .579 -.001 -.101 .068 1.977
    教师评价 -.139 -4.149 -.100 -7.064 .119 3.462
    第二阶段斜率
    性别 .068 1.839 -.013 -.925 -.045 -1.189
    家庭语言 -.104 -2.856 -.110 -7.793 -.037 -.987
    初始年龄 -.193 -5.171 -.116 -8.370 -.121 -3.247
    社会经济地位 -.016 -.447 .077 5.424 .003 .067
    父母阶段评价 .011 .311 .005 .333 -.170 -4.623
    教师阶段评价 -.008 -.219 -.102 -7.167 .009 .245
    下载: 导出CSV
  • [1] 刘红云. (2007).如何描述发展趋势的差异:潜变量混合增长模型.心理科学进展, 15(3), 539-544. http://www.docin.com/p-1408831259.html
    [2] 刘源, 骆方, 刘红云. (2014).多阶段混合增长模型的影响因素:距离与形态.心理学报, 46(9), 1400-1412. http://www.oalib.com/paper/4943869
    [3] 温忠麟, 刘红云, 侯杰泰. (2012).调节效应和中介效应分析.北京:教育科学出版社.
    [4] 温忠麟, 侯杰泰, 张雷. (2006).有中介的调节和有调节的中介变量.心理学报, 38(3), 448-452. http://www.docin.com/p-1301065582.html
    [5] 张洁婷, 焦璨, 张敏强. (2010).潜在类别分析技术在心理学研究中的应用.心理科学进展, 18(12), 1991-1998. http://www.wenkuxiazai.com/doc/e61323a855270722192ef7d3.html
    [6] Berninger, V., Abbott, R., Nagy, W., & Carlisle, J. (2010). Growth in phonological, orthographic, and morphological awareness in Grades 1 to 6. Journal of Psycholinguistic Research, 39(2), 141-163. doi:  10.1007/s10936-009-9130-6
    [7] Blozis, S. A., Feldman, B., & Conger, R. D. (2007). Adolescent alcohol use and adult alcohol disorders:A two-part random-effects model with diagnostic outcomes. Drug and Alcohol Dependence, 88, Supplement 1(0), S85-S96. http://www.sciencedirect.com/science/article/pii/S0376871606004790
    [8] Boscardin, C. K., Muthén, B., Francis, D. J., & Baker, E. L. (2008). Early identification of reading difficulties using heterogeneous developmental trajectories. Journal of Educational Psychology, 100(1), 192-208. http://psycnet.apa.org/record/2008-01796-014
    [9] Chou, C.-P., Yang, D., Pentz, M. A., & Hser, Y.-I. (2004). Piecewise growth curve modeling approach for longitudinal prevention study. Computational Statistics & Data Analysis, 46(2), 213-225. https://www.sciencedirect.com/science/article/pii/S016794730300149X
    [10] Choudhury, K. R., Kasman, I., & Plowman, G. D. (2010). Analysis of multi-arm tumor growth trials in xenograft animals using phase change adaptive piecewise quadratic models. Statistics in Medicine, 29(23), 2399-2409. http://www.ncbi.nlm.nih.gov/pubmed/20564736?dopt=Abstract
    [11] Collins, L. M., & Sayer, A. G. (2001). New methods for the analysis of change. Washington, DC: American Psychological Association.
    [12] Dawson, B. A., & Williams, S. (2008). The impact of language status as an acculturative stressor on internalizing and externalizing behaviors among Latino/a Children:A longitudinal analysis from school entry through third grade. Journal of Youth and Adolescence, 37(4), 399-411. doi:  10.1007/s10964-007-9233-z
    [13] Ferrer, E., Shaywitz, B. A., Holahan, J. M., Marchione, K. E., Michaels, R., & Shaywitz, S. E. (2015). Achievement gap in reading is present as early as first grade and persists through adolescence. The Journal of Pediatrics, Aug 2015.
    [14] Grimm, K. J., Ram, N., & Estabrook, R. (2010). Nonlinear structured growth mixture models in Mplus and OpenMx. Multivariate Behavioral Research, 45(6), 887-909. doi:  10.1080/00273171.2010.531230
    [15] Grimm, K. J., Ram, N., & Hamagami, F. (2011). Nonlinear growth curves in developmental research. Child Development, 82(5), 1357-1371. doi:  10.1111/j.1467-8624.2011.01630.x
    [16] Hipp, J. R., & Bauer, D. J. (2006). Local solutions in the estimation of growth mixture models. Psychological Methods, 11(1), 36-53. doi:  10.1037/1082-989X.11.1.36
    [17] Hunter, A. M., Muthén, B. O., Cook, I. A., & Leuchter, A. F. (2010). Antidepressant response trajectories and quantitative electroencephalography (QEEG) biomarkers in major depressive disorder. Journal of Psychiatric Research, 44(2), 90-98. doi:  10.1016/j.jpsychires.2009.06.006
    [18] Kaplan, D. (2000). Structural equation modeling: Foundations and Extensions. Thousand Oaks, CA: Sage.
    [19] Kieffer, M. (2011). Before and after third grade:Longitudinal evidence for the shifting role of socioeconomic status in reading growth. Reading and Writing, 1-22. doi:  10.1007/s11145-011-9339-2.pdf
    [20] Lei, L., Pan, J., Liu, H., McBride-Chang, C., Li, H., Zhang, Y., et al. (2011). Developmental trajectories of reading development and impairment from ages 3 to 8 years in Chinese children. Journal of Child Psychology and Psychiatry, 52(2), 212-220. doi:  10.1111/j.1469-7610.2010.02311.x
    [21] Liu, Y., & Hou, S. (2017). Potential reciprocal relationship between motivation and achievement: A longitudinal study. School Psychology International, 014303431771057. doi: 10.1177/0143034317710574.
    [22] Liu, Y., Liu, H., & Hau, K. T. (2016). Reading ability development from kindergarten to junior secondary:Latent transition analyses with growth mixture modeling. Frontiers in Psychology, 7:1659. doi:  10.3389/fpsyg.2016.01659/full
    [23] Logan, J. A. R., & Petscher, Y. (2010). School profiles of at-risk student concentration:Differential growth in oral reading fluency. Journal of School Psychology, 48(2), 163-186. doi:  10.1016/j.jsp.2009.12.002
    [24] Malone, P. S., Van Eck, K., Flory, K., & Lamis, D. A. (2010). A mixture-model approach to linking ADHD to adolescent onset of illicit drug use. Developmental Psychology, 46(6), 1543-1555. doi:  10.1037/a0020549
    [25] Masyn, K. E., Henderson, C. E., & Greenbaum, P. E. (2010). Exploring the latent structures of psychological constructs in social development using the dimensional-Categorical spectrum. Social Development, 19(3), 470-493. http://www.ncbi.nlm.nih.gov/pubmed/24489441
    [26] McAuley, E., Mailey, E. L., Mullen, S. P., Szabo, A. N., Wojcicki, T. R., White, S. M., et al. (2011). Growth trajectories of exercise self-efficacy in older adults:Influence of measures and initial status. Health Psychology, 30(1), 75-83. doi:  10.1037/a0021567
    [27] McCoach, D. B., O'Connell, A. A., Reis, S. M., & Levitt, H. A. (2006). Growing readers:A hierarchical linear model of children's reading growth during the first 2 years of school. Journal of Educational Psychology, 98(1), 14-28. doi:  10.1037/0022-0663.98.1.14
    [28] Muthén, B. (2004). Latent variable analysis: Growth mixture modeling and related techniques for longitudinal data. Thousand Oaks, CA: Sage.
    [29] Muthén, B., & Brown, H. C. (2009). Estimating drug effects in the presence of placebo response:Causal inference using growth mixture modeling. Statistics in Medicine, 28(27), 3363-3385. doi:  10.1002/sim.3721
    [30] Muthén, L., & Muthén, B. (2012). Mplus (Version 7. 0). Los Angeles, CA: Muthén & Muthén.
    [31] Nylund, K. L., Asparouhov, T., & Muthén, B. O. (2007). Deciding on the number of classes in latent class analysis and growth mixture modeling:A Monte Carlo simulation study. Structural Equation Modeling:A Multidisciplinary Journal, 14(4), 535-569. doi:  10.1080/10705510701575396
    [32] Palardy, G. J. (2010). The multilevel crossed random effects growth model for estimating teacher and school effects:Issues and extensions. Educational and Psychological Measurement, 70(3), 401-419. doi:  10.1177/0013164409355693
    [33] Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods(2nd ed). Thousand Oaks, California: Sage Publications.
    [34] Schaeffer, C. M., Petras, H., Ialongo, N., Poduska, J., & Kellam, S. (2003). Modeling growth in boys' aggressive behavior across elementary school:Links to later criminal involvement, conduct disorder, and antisocial personality disorder. Developmental Psychology, 39(6), 1020-1035. doi:  10.1037/0012-1649.39.6.1020
    [35] Singer, J. D., & Willett, J. B. (2003). Applied longitudinal data analysis:Modeling change and event occurrence. New York:Oxford University Press.
    [36] Sun, Y., & Li, Y. (2011). Effects of family structure type and stability on children's academic performance trajectories. Journal of Marriage and Family, 73(3), 541-556. doi:  10.1111/j.1741-3737.2011.00825.x
    [37] Tourangeau, K., Nord, C., Lê, T., Sorongon, A. G., and Najarian, M. (2009). Early Childhood Longitudinal Study, Kindergarten Class of 1998-99 (ECLS-K), Combined User's Manual for the ECLS-K Eighth-Grade and K-8 Full Sample Data Files and Electronic Codebooks (NCES 2009-004). National Center for Education Statistics, Institute of Education Sciences, U. S. Department of Education. Washington, DC.
    [38] Uher, R., Muthén, B., Souery, D., Mors, O., Jaracz, J., Placentino, A., et al. (2010). Trajectories of change in depression severity during treatment with antidepressants. Psychological Medicine, (40), 1267-1277. http://www.ncbi.nlm.nih.gov/pubmed/19863842/
    [39] Votruba-Drzal, E., Li-Grining, C. P., & Maldonado-Carreño, C. (2008). A developmental perspective on full-versus part-day kindergarten and children's academic trajectories through fifth grade. Child Development, 79(4), 957-978. http://www.ncbi.nlm.nih.gov/pubmed/18717901
  • 加载中
图(3) / 表(7)
计量
  • 文章访问数:  50
  • HTML全文浏览量:  26
  • PDF下载量:  300
  • 被引次数: 0
出版历程
  • 刊出日期:  2018-01-20

目录

    /

    返回文章
    返回