This study explores how children from two to nine years of age develop Computational Thinking (CT) skills. The term "CT" denotes a set of cognitive processes that are useful for framing and solving problems using computers and other information processing agents. Acquisition of CT skills is traditionally associated with learning to code but other factors may influence accrual in the course of normal development. To explore this issue, I examined two- to nine-year-old children's performance on four grade-specific versions of the 15-item "TechCheck" "unplugged" assessment probing six domains of CT. I collected data from coding-naive children receiving the version of "TechCheck" designed for their grade and compared their performance to that of coding-naive students in one grade higher who were administered the same "TechCheck" version. "TechCheck" scores in all grades were normally distributed. Mean scores were significantly greater in students in the higher of each grade pair. This finding suggests that average performance in solving unplugged problems that probe CT skills may improve with advancing grade in the absence of coding instruction. Linear mixed modeling identified grade and the interaction of grade with age as predictors of "TechCheck" performance in coding-naive students. Next, I examined "TechCheck" data obtained from first graders before and after they received one of three coding educational interventions. "TechCheck" scores improved significantly after children learned to code compared to a non-coding control group, providing evidence that the acquisition of CT can be accelerated by coding education. Finally, I explored whether grade and coding interventions differentially affected performance across CT domains. With each advancing grade, coding-naive students scored higher in all six CT domains evaluated. Children who were taught to code showed more selective improvements in specific CT domains depending on which coding educational intervention they received. Limitations to this study include the use of heterogeneous cohorts drawn from multiple studies and the exploratory nature of domain analysis using "TechCheck." I conclude that CT skills can improve in the course of early childhood without CS instruction, perhaps as a result of learning from everyday experiences, non-CS education, and/or brain maturation. Learning to code at a young age, which is known to foster improved communication and creative self-expression, can accelerate the acquisition of CT but may have more selective effects on specific CT domains depending on how coding is taught. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]