## Affecting Students’ Behavior with Plagiarism Detection in Evaluation Process
/ [**DisCo 2018**](https://disconference.eu/)
### It has never been easier to cheat than today.
## Don't Forget to Interact with the Audience!
(Manager)
## [Programming](http://it4kt.cnl.sk/c/pvjc) Course
* ~~not very~~ introductory programming course
* known as _CS0_ or _CS1_
* _C_ language, so not so easy
* 2nd semester of Bc study
* more than _600_ students
* _5_ projects during semester
## The Goal is Clear
> learn to code
(including thinking, problem solving, having fun, but **NO CHEATING!**)
## Data Science Process
## During Last 5 Years
* process of heavy **data acrquiring**
* development of test platform _Arena_
* application of **TDD** and **CI** with [Git](https://git-scm.com/) and [Gitlab](https://about.gitlab.com/)
* over than _200.000_ entries
## Let the Data Talk
* (real time) data **analysis** and **interpretation**
* when the students work
* what is the [score distribution](https://arena.kpi.fei.tuke.sk/arena/courses/prog.2018) during the time
* what is the [pattern of they behavior](https://arena.kpi.fei.tuke.sk/arena/courses/prog.2018)
* setting **predictions** and **hypothesis**
* according to the work of student during the semester, what will be his grade on the final exam?
* is there corelation between the qualithy of teacher and his students?
## The Missing Part
* automatic detection of plagiarism in students source codes :-(
* we have the tools ([JPlag](https://github.com/jplag))
* we have the [honor code](http://it4kt.cnl.sk/c/pvjc/honor.code.html)
* simple rule to follow: _Don't copy/use foreign code and don't allow anyone to copy your code._
* it's easy to interpret/work with the data
* we are too ~~lazy~~ busy to implement the service
## The Results
| Assignment | Min. Level | Students | % |
| ------------------|:----------:| ----------:|:--:|
| K | 80% | 548/47/77 | 14 |
## The Results
| Assignment | Min. Level | Students | % |
| ------------------|:----------:| ----------:|:--:|
| K | 80% | 548/47/77 | 14 |
| Top Secret | 80% | 504/102/86 | 17 |
## The Results
| Assignment | Min. Level | Students | % |
| ------------------|:----------:| ----------:|:--:|
| K | 80% | 548/47/77 | 14 |
| Top Secret | 80% | 504/102/86 | 17 |
| Problems to Solve | 70% | 449/150/51 | 11 |
## The Results
| Assignment | Min. Level | Students | % |
| ------------------|:----------:| ----------:|:--:|
| K | 80% | 548/47/77 | 14 |
| Top Secret | 80% | 504/102/86 | 17 |
| Problems to Solve | 70% | 449/150/51 | 11 |
| Adventure | 60% | 166/434/14 | 8 |
## The Results
| Assignment | Min. Level | Students | % |
| ------------------|:----------:| ----------:|:--:|
| K | 80% | 548/47/77 | 14 |
| Top Secret | 80% | 504/102/86 | 17 |
| Problems to Solve | 70% | 449/150/51 | 11 |
| Adventure | 60% | 166/434/14 | 8 |
| Mastermind | -- | -/-/- | - |
## Distribution of Plagiarism Pairs in K
```
90% - 100% 141 #
80% - 90% 37 #
70% - 80% 87 #
60% - 70% 437 #
50% - 60% 1088 #
40% - 50% 3811 #####
30% - 40% 11657 ################
20% - 30% 27935 #######################################
10% - 20% 52464 ###########################################################################
0% - 10% 52221 ##########################################################################
```
## Results Interpretation
* 5 years ago...
* ~~confessions~~ essays about the topic _"Why I cheated?"_
* funny stories behind
* surveys
* most of the respondents strongly agree...
* ...except the situation, if it is about them (personally)
* habbits from high school or from previous semester?
(**https://bit.ly/2KkkwaK**)