Research Design
Home > Research Design
This Research Programme Draws On
Multiple Research Designs.
Longitudinal:
Cross-sectional designs are useful for explaining the variability and relations amongst variables of interest at a particular time periods. But longitudinal designs provide more powerful mechanisms that allow measurements of change (using repeated measures) and direct inferences regarding causal pathways underlying inter and intra-level relationships (using longitudinal measures), and are therefore pivotal for the study of life course transitions and development trajectories.
Multi–level:
Multi-level designs are based on nested analytical frameworks and are well-suited for evaluating factors and processes at the individual (micro) and social context (macro) levels simultaneously. In education for example, the nesting or clustering of students within classrooms, classrooms within schools, and schools within communities and/or systems needs to be accounted for due to the lack of independence between variables that operate at different levels. Multilevel data also differs along the degree to which within-group units correlate with one another, depending on the degree of dependency. For example, patterns of relationships among students nested within classrooms or courses are likely to be more correlated than within schools and schools more than neighbourhoods and so on.
Longitudinal Multilevel:
As educational pathways and trajectories of individual development are shaped by life course transitions and social contexts that are inherently group-level constructs, longitudinal multilevel designs are needed to specify and evaluate their consequent effects on subsequent outcomes and for evaluating the nesting of time within students, students within schools, and schools within systems.
Cross classified:
This research design permits the identification and control for the impact of simultaneous multiple membership effects such as neighbourhoods and schools. Additionally, complex school effects can also be efficiently decomposed into main and crossed effects that allow addressing complex research questions. For example, the persistence of primary on higher education outcomes or choices after adjusting for secondary schools. Misleading inferences can result when such effects are present but ignored, thus leading researchers to overlook important features group-level variation.
Ecological:
The research can draw on an “ecological” human development approach to understand youth and adulthood development, taking into account the variety of micro and macro social or institutional processes and interactions.