Reading competences in the context of social and cultural diversity: Differential meanings of aspects of tasks, students and instruction in large-scale-assessments

关于能力的获得和教育的参与，德国教育系统中不同学生群体的学业成功差异很大，这取决于社会背景（已经Baumert & sch<e:1> mer, 2001; Maaz, Neumann & Baumert, 2014）以及家庭的语言或移民背景（Stanat & Edele, 2011; international Atger, 2009）。这些教育差异在小学（四年级）结束时就已经可以观察到。在这种背景下，大规模评估（LSA）完成了系统地记录不同学生群体习得能力的功能，并表明了学校条件的各个方面。阅读素养是教育过程的核心目标，同时也是前提条件（Pfost, Dörfler & Artelt, 2012）。对于阅读能力这一重要领域，计划中的项目将进行基于理论的分析，以确定所关注的学生群体，即处于社会不利地位的儿童和具有移民背景的儿童的具体优势和劣势在多大程度上受到(a)学习者的动机特征和(b)学校教学水平的数量和质量特征的影响。这是一种创新的方法，可以在学生和教学层面上分析学生分组在阅读测试项目方面表现出的优势和劣势的先决条件。分析的重点是不同的项目功能效应，这已经证实了项目与开放式回答格式。研究焦点1强调学生的动机特征(a)内在阅读动机和(b)阅读自我概念。研究重点2将教学特征作为先决条件，从而定量方面（即分配给阅读的时间）和定性方面（即课堂上练习阅读策略）的重要性进行调查。特别相关的是研究重点3，它通过同时检查学生和教学特征来巩固其他两个研究重点的发现。本文采用交叉分类多层次模型对2006年的PIRLS样本数据进行分析，以检验手头的问题。这些数据允许对三个兴趣水平（项目/学习者/教学）的影响进行建模。预期的结果与重要的发现有关，这些发现与关于主要学生分组中特定项目优势和劣势的决定因素（在学生和教学水平上）的理论的牢固建立有关。这些发现对于系统地推进和解释LSA和其他研究中使用的测试项目，以及在阅读教学和后续研究优先事项中确定特定个人支持的可能选择具有重要意义。

With respect to the acquisition of competencies and the participation in education the academic success of different student subgroups in the German educational system varies greatly, depending on social background (already Baumert & Schümer, 2001; Maaz, Neumann & Baumert, 2014) and language spoken or migration background of the family (Stanat & Edele, 2011; international Atger, 2009).These educational disparities can already be observed towards the end of elementary school (4th grade). In this context, Large-Scale-Assessments (LSA) fulfill the function of systematically documenting acquired competencies of different student subgroups as well as indicating aspects of school conditions. Core aim and, at the same time, prerequisite of educational processes is reading literacy (Pfost, Dörfler & Artelt, 2012). For this important area of reading literacy the in-tended project will conduct theory-based analyses in order to determine to what extent spe-cific strengths and weaknesses of the student subgroups in focus, children who are socially disadvantaged and those with migration backgrounds, are being influenced by (a) motivational characteristics on the level of the learners and (b) quantitative and qualitative characteristics on the level of school instruction. This is an innovative way to also analyze, on student and instruction levels, the prerequisites of strengths and weaknesses subgroups of students display with respect to reading test items. Analyses focus on Differential Item Functioning-effects, which have been already substantiated for items with an open-answer format. Research focus 1 emphasizes the motivational student characteristics (a) intrinsic reading motivation and (b) self-concept with respect to reading. Research focus 2 examines teaching characteristics as preconditions whereby quantitative aspects (i.e. time allocated to reading) as well as qualitative aspects (i.e. practicing reading strategies in class) are investigated with respect to their significance. Of particular relevance is research focus 3 which consolidates the findings of the other two research focusses by simultaneously examining student and instructional characteristics. Cross-classified Multilevel-Models are used to analyze data of PIRLS, sample from 2006, in order to examine the questions at hand. These data allow for the modelling of effects on the three levels of interest (items/learners/instruction). The anticipated results pertain to important findings relating to the firm establishment of theories regarding the determinants (on student and instruction levels) of item-specific strengths and weaknesses in major student subgroups. These findings are of fundamental importance for the systematic advancement and interpretation of test items used with LSA and other studies, as well as for the identification of possible options for specific individual support in reading instruction and subsequent research priorities.