Individualization can be defined as the adaptation of instructional parameters to relevant characteristics of a specific learner. This definition raises several questions, however: Which characteristics are actually relevant? Which parameters of instruction need to be adjusted, and in which way, to positively interact with those characteristics? In a classroom context, additional questions arise: how can information about the relevant learner characteristics be delivered to the teacher? How can individualized instruction be delivered to each learner in a context that has originally been designed for whole-class instruction? By focusing on the measurement and modelling of learner characteristics and instructional adaptations, this dissertation aims to provide an insight into each of these issues. This dissertation is divided into two parts. The first part is concerned with the theoretical (Paper 1) and statistical (Paper 2) modeling of learner characteristics in the context of individualized instruction. The second part is concerned with the measurement (Paper 3) and implementation (Paper 4) of individualized instruction in the classroom context. Paper 1 summarizes existing research on individualization from different research traditions. From this summary I derive the need for a dynamic conceptualization of learner characteristics (acknowledging that learners change during and in interaction with the learning process) and synthesize a dynamic framework that details the opportunities for individualization on three different timescales. Paper 2 reports results from an exploratory study that investigated the potential benefits of utilizing person-centered analysis for the assessment of multivariate learner prerequisites and their interaction with instruction. We found that latent profiles over several reading related abilities could explain differential effectiveness of self-reported teaching foci in German third grade reading lessons. These findings indicate not just a need for stronger individualization of teaching but also an advantage of multivariate conceptualizations of learner characteristics. Additionally, they show the utility of person-centered approaches for the investigation of such multivariate learner characteristics and their interaction with instruction. In the second part, I investigate possible approaches to the implementation and measurement of individualization in a classroom context. Paper 3 investigates whether teacher-, student- and observer perspectives converge when rating the amount of individualization present in regular classroom instruction. We found considerable agreement between the perspectives, indicating a common understanding of the construct at the classroom level as well as providing some evidence for the validity of the used measurement instruments. Paper 4 replicates findings concerning the effectiveness of formative assessment procedures for fostering reading education, supplemented by a moderator analysis showing that only children with low performance at the beginning of the school-year profited from its implementation. This indicates that the information provided by formative assessment procedures helps teachers to identify struggling readers but does not seem to be utilized for adapting instruction to specific deficits of average or high performing children. In sum, this dissertation contributes to research on individualized instruction by demonstrating necessary conditions for its effectiveness. It posits the need for a dynamic conceptualization of learner characteristics, demonstrates the advantage of multivariate learner profiles, and points out ways towards the successful implementation of individualized instruction in the classroom.