This paper attempts to reconcile different portrayals of early literacy resulting from different theoretical analyses. The responses of 293 four to six-year-old children to some early literacy assessment tasks were analysed using quantitative and qualitative methods at group and individual levels. A comparison of the findings suggests that different ways of processing and analysing the data reflect different assumptions about the nature of early literacy, and could lead to distinctive portrayals of early literacy. Results from the item response modelling suggested that early literacy development appears to be structured and hierarchical, while thematic analysis found that the children’s early literacy understanding seemed to be diverse and meaning-situated. At the individual level, some children with the same ability estimates appeared to approach the early literacy assessment tasks in different ways, supporting the idea of multiple developmental pathways. The combined findings support the premise that research is an intrinsically interpretive process. The theories that researchers use to generate knowledge about the world (epistemology) influence the kinds of information that are considered as evidence and the inferences that are made about the evidence. This paper argues for the need to critically examine the theoretical lenses currently used in early literacy research and assessment.