Findings are presented from two separate year-long longitudinal studies of the development of planning skills among school aged children in relation to learning LOGO programming, and a theoretical context is provided for predictions of greater improvement by the programming groups. In the first year, experimental groups comprised students in each of two classrooms in a private school in Manhattan. One classroom included 25 8- and 9-year-old children; the other consisted of 25 11- and 12-year-old children. The control groups were made up of students in the same grade level classrooms in the same school. Both experimental groups were administered a classroom chore-scheduling planning task. Process and product measures of planning skill revealed no benefits for students doing LOGO programming. The second experiment took place one year later in the same school in the same two teachers' classrooms. The second study comprised 32 students in each of the age groups of the first study. This time a microcomputer version of the task was implemented in which students gave commands to a robot to carry out the chores, and similar assessments of planning performances were collected online. Again, learning to program did not differentiate experimental from control group performances. Further tests of the programming transfer hypothesis are proposed. Data tables and references are included. (Author/THC)