This study investigates morphosyntactic complexity in children’s utterances from a multidimensional perspective. Results from Mean Length of Utterance and Index of Productive Syntax were compared with the results of Utterance Analysis, a method developed for the purposes of this study, to discover what aspects of absolute complexity are reflected by each of the scales. A second purpose of the study was to compare children with high genetic risk of dyslexia (N= 20) with their controls (N= 20) in order to determine whether morphosyntactic complexity in their utterances differed on a group level. The data were gathered in a play situation when the children were 30 months old. MLU and IPSyn analyses were carried out on the 80 longest utterances from each child. For the Utterance Analyses three subgroups showing contradictory results in MLU and IPSyn were selected. The study revealed three major findings. The first, concerning morphosyntactic development in general, suggests that children usually start with morphological elaboration first in one and then in two components in an utterance. Only after that does the elaboration of components start to deepen vertically. The more components there are in an utterance the more similar is children’s performance in elaboration. The second finding, concerning methodology, suggests that different scales of measurement reveal different aspects of complexity. MLU focuses on linear length and IPSyn on the inventory of resources, but the relationship between these two methods uncovers developmental trends in morphosyntactic elaboration. Only Utterance Analysis can provide a complete picture of multidimensional structural complexity. The third finding concerns the comparison between risk and control children. It seems that children with a high genetic risk of dyslexia do have the same morphosyntactic resources as their controls, but they use them differently. Risk children concentrate more on elaborating only a single component per utterance whereas control children spread elaboration to more components.