-  An unexpected finding was that students who averaged less days of working per week actually had higher exam scores. However, this finding was not explained by previous programming experience. This finding is contrary to past research findings.
     -  One potential explanation is that past research has typically focused on recall which might not be directly applicable to programming. 
-  Students in the second course spent half a day more on assignments than those in course 1, which the researchers attempt to explain in the paper. 
    -  One explanation is that the workload was marginally increases in Course 2.
    -  Another explanation was the grading scheme- course 2 was a pass/fail class while Course 1 was graded. 
    -  Lastly, course 1 had more CS majors and course 2 had a lot of CS minors.
-  In course 2, those students in the top 25% in exam points spent less days on assignments compared to the bottom 25%. 

Spacing Work Over Multiple Days 

-  Negative correlations were found between the relative amounts of short pauses and exam scores. However, it is unclear whether the short pauses were to look at other course material (on-task), or off-task behavior (such as going on social media). 
-  This negative correlation, however, aligns with previous research in that if students were off-task during the pauses, this would be detrimental and for the very short off-task times, it could be considered multitasking which would also be detrimental. Simultaneously, with a short enough pause, the advantages from the spacing effect would not have time to occur. 
-  In a similar effect, even if the students were on task during the pauses, the negative effects of task-switching would still occur. This also indicates a higher cognitive load of the task in which the student is unable to hold all information necessary for the problem and has to jump back and forth between material. 
-  The “downslide” effect described by Collins & Genter can be applied to this research and programming in general. Downsliding is when an individuals “slides down” from higher to lower level processing, ultimately losing the big picture. They use this to suggest that both off-task and on-task pauses can harm the studying and learning process. 

Short Pauses and Course Outcomes 

- This research does not completely answer the question of how pause lengths are correlated with lower exam scores. 
- The data shows that extremely short pauses of 1-2 seconds are positively correlated with a high exam score, while pauses between 1-4 minutes are negatively correlated with exam scores. 
    - One possible explanation for this is that these extremely short pauses do not have an effect by itself but really just show that longer pauses were not taken. 
    - Another explanation is that these very short pauses were times where the student spends a very short time thinking about his/her work, which is an explanation that aligns with past research. 

Correlations of Pauses Lengths And Exam Scores 

- Out of the observed background variables, the only background factor correlated with the spacing and relative amounts of pauses of lengths in relation to lower exam scores was programming background. Although intuitive that this variable would play a role, it was very surprising that this was the only statistically significant variable. 
- The researchers thought that the MSLQ variables would be correlated with the days worked and the amount of pauses correlated with lower exam scores. However, these correlations were not observed. One possible explanation is that the MSLQ only measures what the students think they are doing as opposed to what they are actually doing and students are not aware of the harmful effects of multitasking and the positive effects of spacing. 

Correlations Between Background Variables and Pausing Behavior 

-  Spacing Work Over Multiple Days
-  Short Pauses and Course Outcomes
- Correlations of Pauses Lengths And Exam Scores
-  Correlations Between Background Variables and Pausing Behavior

University of Michigan - Ann Arbor

Students in two subsequent introductory programming courses completed programming tasks in an integrated development environment that saved detailed logs of their actions, including time stamps of all the participants keypresses. 

Student Pauses and Spacing in Programming

Leo Leppänen, Juho Leinonen, and Arto Hellas. 2016. Pauses and spacing in learning to program. In Proceedings of the 16th Koli Calling International Conference on Computing Education Research (Koli Calling '16). Association for Computing Machinery, New York, NY, USA, 41–50. DOI:https://doi.org/10.1145/2999541.2999549

Pauses and spacing in learning to program

-  Past research has shown that spacing produces better learning results than massing (or cramming). Research also shows that pausing is beneficial within the context of a single problem, due to the dual phenomena of fixation or incubation. Fixation is when someone is stuck on a problem but takes time off to get unstuck, while incubation is the act of spending time off-task, essentially letting the subconsciousness solve the problem. 
-  Studies also show that multitasking can have a beneficial effect on learning through decreasing boredom. 
-  Research has also shown disadvantages of pausing. For example, it has been seen that the act of changing your focus causes slower response and more errors. Research also shows that multitasking using an electronic device is detrimental to learning. 
-  This is important, because programming is an activity that includes working on multiple files at a time while multitasking through referencing manuals and documentation— meaning that programming might inherently require both multitasking and task-switching. 

Student Pauses and Spacing in Learning to Program: Background

-  This study looks at how students in the course take pauses and space their work, and further look at how this behavior affects their exam scores. 
-  Looking to identify pausing and spacing patterns and their relations with good or bad course outcomes. 
-  Also want to discuss the transferability of past work on pausing and spacing to the context of learning a task like programming. 

Student Pauses and Spacing in Learning to Program: Introduction & Aim

- 1) How do students space out their work? 
    - 1.1 On how many days on average per week do students work on course assignments?
    - 1.2 When working on course assignments, do students take short pauses and if so, what are the typical pause lengths?
- 2) How does the students’ tendency to pause influence the course outcomes?
- 3) Do pausing or spacing behaviors correlate with student background variables? 
    - 3.1 To what extent do the students’ background variables and answers to the Self-Regulation-specific questions from the MSLQ correlate with students spacing out their work over days?
    - 3.2 If there are pause lengths that are negatively correlated with exam scores, how are the relative amounts of such pauses correlated with students’ background variables and answers to the Self-Regulation-specific questions from the MSLQ?

Learn Before

Related

Learn After