Mnemonic Variable Names in Parsons Puzzles

Abstract

In Parsons Puzzles, students are asked to arrange the lines of a program in their correct order. We investigated the effect of using mnemonic variable names in the program on the ease with which students solved the puzzles -whether students were able to solve puzzles containing mnemonic variable names with fewer actions or in less time than single-character variable names. We conducted a controlled study with cross-over design over four semesters. Much to our surprise, we found no statistically significant difference between students solving puzzles with mnemonic variable names versus single-character variable names -either in terms of the number of actions taken, the grade earned or the time spent per puzzle. In this paper, we will describe the experimental setup and data analysis and present the results of the study. We will discuss some hypotheses as to why the readability of the variable names did not impact students' ability to solve Parsons puzzles.