Organizer: Ellen Hamaker
Development is at its core an intraindividual phenomenon. However, the dominant approach in developmental research consists of taking a few repeated measurements from a large number of individuals, and determining the mean changes in this group. Even when research focuses on individual differences in developmental trajectories, the models are often static and describe deterministic developmental paths. This practice does not do justice to the complex nature of development, which tends to be characterized by large individual differences, and by dynamic relationships between developmental processes. As a result, interventions and policy changes based on such research may be missing the mark. In this symposium we discuss alternative methods to study development, with intraindividual variability, changes, and dynamics as its centerpiece.
Ellen Hamaker. Development as an intraindividual phenomenon: An introduction
Development can be defined as “the process in which someone grows or changes and becomes more advanced.” There are two key elements to this definition: Development takes place within an individual, and occurs over time. Hence, to study development at its core, we need to measure a person repeatedly, and we need a statistical technique that uncovers the key features of the intraindividual process unfolding over time. In this presentation, I will explain why we cannot use shortcuts such as cross-sectional data (based on a single measurement), and why many conventional longitudinal approaches (based on a few repeated measurements) are falling short. Moreover, I will discuss how recent methodological innovations have opened up a new horizon of research opportunities, which allow us to truly focus on intraindividual variability, changes, and dynamics. I will end by discussing some of the major challenges that the field now needs to face.
Lars-Erik Malmberg. Intraindividual variability – a construct in its own right
In order to enhance our understanding of developmental dynamics we can apply an intraindividual perspective in which we focus on fluctuations in the shorter-term, termed variability. Intensive longitudinal data are well suited for this purpose. Taking an example from educational research, we observed adaptive educational processes characterised by stable (less variability), and maladaptive processes characterised by instable (more variability) learning experiences (e.g., task difficulty, competence evaluation and intrinsic motivation) from one learning situation to the next. We specified multilevel structural equation model of state, trait and individual differences in intraindividual variability constructs in a sample of 285 English primary school students’ (Years 5 and 6) who completed the Learning Experience Questionnaire using handheld computers, on average 13.6 learning episodes during one week (SD = 4.6; Range = 5-29; nepisodes = 3,433). We defined mean squared successive differences (MSSD) for manifest indicators of the variability constructs. Overall, we find support for intraindividual variability as a construct in its own right, which has the potential to provide novel insight into students’ learning processes.
Han L. J. van der Maas. Intraindividual development: Theories and measurement
The human cognitive system, and especially its development in the first years, is inconceivably complex. The rich intraindividual variation in complex developmental processes requires the study of development at the level of the individual. Yet, the intraindividual approach is all but easy in practical research. One major challenge is the collection of short interval times series of reliable data on development. We present an measurement approach to this problem based on an innovative online learning system which is now used by thousands of Dutch primary school children on a daily or weekly basis, providing a new window on cognitive development. We will introduce the origin of this new instrument, called Math Garden, explain its setup, and discuss its potential in studying developmental processes.