Block-based programming languages like Scratch,Alice and Snap are increasingly popular in novice programming education.They allow novices compose program to implement interesting artifacts without struggling with complicated syntax.However,few researches provide insight into how novices do and where they get stuck during programming process.To this end,students programming characteristics and their block category transition from an assignment process are explored in this work.A statistic distribution analysis was first given to investigate which and how many block categories are used in this assignment.Furthermore,we made cluster analysis on the frequency of block category transitions to find the transition pattern.Based on this,the lag sequential analysis were used to contrast and analyze the block category transition pattern among clusters.Our results show that 1)although the system embeds the hint function,not all students are willing to use it,2)the difference of students programming style and challenge can be found from their block category transitions.