A 3-year longitudinal study of children's comprehension of counting: Do they recognize the optional nature of nonessential counting features?

• Children's justifications increased the reliability of the coding criterion.
• Erroneous counts were easily detected than pseudoerrors.
• The presence of the cardinal value for the set improved performance on pseudoerrors.
• Longitudinal data showed different developmental trajectories in children's counting knowledge.

This 3-year longitudinal study examines developmental changes in children's ability to differentiate essential from nonessential counting features. Kindergarteners watched a computer-presented detection task which included three kinds of counts: correct conventional, erroneous and pseudoerrors (with and without statements of cardinal values for the sets). Children had to judge the correctness of those counts and justify their responses. Our data showed that children's explanations provided additional information and thus increased reliability of the assessment. Children were better at detecting erroneous counts than pseudoerrors and at detecting pseudoerrors with cardinal value than pseudoerrors without it. Group analysis showed that children's performance improved with age but analysis of individual differences qualified this result by identifying individual differences in developmental patterns. This study thus provides a more detailed picture of the developmental trajectories of children's comprehension of essential and nonessential counting aspects.