Get best of the week

About one solution to the problem of thieves and prostitutes in online educational services systems

One of the obvious drawbacks of online education systems is the inability to control students' independence in solving problems. This leads to massive fraud and, as a result, to the depreciation of issued certificates. The article proposes some technical tool that allows to partially reduce the damage from this phenomenon in online education.

This article (under the heading "On one solution to the problem of" cheaters "and" consultants "in the provision of educational services online") was sent to the official, scientific pedagogical journal " Computer Instruments " and is under review. But something I doubt she will ever come out.

Description of the problem:


Ideally, there are three actors in the online education system: Teachers, students and employers. In the ideal case, the process of providing educational services online occurs as follows:

  1. Teachers give assignments to students.
  2. Students send assignments to teachers.
  3. Teachers check solutions.

This is repeated several times, and as a result, teachers pass on to students' employers along with a certain certificate, the implied meaning of which is: “This student can do something on his own”. Lecturing, handing out books, answering questions and other things related to learning, and not to providing an educational service online, is not considered in this article.

In reality, there are two other actors in the online education system: cheaters and consultants.

In the reality:

  1. Cheaters pretend to be students, receive assignments and pass them on to consultants.
  2. The consultants solve the tasks and pass the decision to the cheaters. 
  3. Cheaters pass the products of consultants to teachers.

Thus, some of the efforts of teachers are spent on meaningless filtering of the products of the labor of consultants and surreal conversations with cheaters.
In case there are a lot of cheaters with consultants, the provision of educational services is distorted to the point of complete caricature:

  1. Teachers are already spending most of their efforts on meaningless activities, and after a while they simply stop checking and filtering something in accordance with the principle of “let life itself punish them strictly”.
  2. The certificate transmitted to the employer together with the student (or cheater) ceases to mean anything, and the employers have to deal with filtering and conversations themselves.

As an illustration of the complete caricature of the situation, the famous story about drawing a diode bridge [1] and the phrase that was discovered on one of the resources with the consultants perfectly fit : “... your order will be completed as soon as possible. Most of the time it takes to pay for the order. Also for this reason, the cost of examples from these problem books is minimal (from 8 rubles per example / task). For other task books, the prices are slightly higher (from 20 rubles per example / task) ... ”. The reader can independently assess the extent of the problem by “google” the phrase “help with online learning”.

Traditional methods of dealing with this phenomenon, associated with calls for ethics and increased control, are useless when learning online. An attempt to make the problem invisible with the help of “objective tests” with “answer in the form of numbers” leads to even faster degradation of education, because makes it impossible to set meaningful tasks and meaninglessly complicates the life of those few students who are going to learn something.
As one of the solutions to the problems caused by the presence of cheaters and consultants in online education systems, two ideas are proposed:

  1. Partially shift the verification of work to consultants (and students).
  2. As a certificate, use the works of students (and cheaters) themselves.

Idea 1


The idea of ​​checking the work by students themselves is not new, it has been studied in sufficient detail, has already been implemented in some online systems and looks something like this:

  1. The student (or cheater) receives the task, solves it with a pen on paper, photographs by phone and sends it to the teacher (more precisely, to the training system).
  2. The solution (i.e., a photo of the solution) along with the correct answer or verification instruction is sent to several random students (or cheaters).
  3. These random students (or cheater-related consultants) check the decision and give a grade.
  4. Thus, the work of a student (or cheater) has several grades that were put by several random students (or consultants), and these grades are somehow taken into account when issuing the certificate.

It is proposed to bring this system of mutual checks to the logical end:

  1. - , ( ) .
  2. ( ) , , .
  3. ( ) ( ).
  4. , , , .
  5. , , ( ).
  6. , , , , «» .
  7. , , « », .. , .
  8. . . .

Thus, teachers are engaged in students, students become “not only writers, but also readers”, which is useful, cheaters with consultants take care of themselves and, as a nice bonus, also help to check students' work. And, even if there will be significantly more cheaters with consultants than students, the system will remain operational, since the work of the student will be checked by several highly qualified consultants, and vice versa, the student will come to check mostly exemplary works written again by highly qualified consultants.

Idea 2


As a certificate, the employer receives not just a piece of paper with a seal, but a complete set of all tasks with solutions written by the student (or consultant).
If the employer is interested, he can filter and talk, but now he no longer needs to come up with questions about the diode bridge, he now has information for more meaningful questions such as:

  1. Tell us what is written here, because you yourself wrote it, didn’t you?
  2. Here you have exactly the same problem, you have already solved it, can you solve it again?
  3. What does it mean we didn’t go through? Here is the task that you were asked, and which you have already successfully solved.

If the employer is not interested, then everything remains the same as before: the cheater, consultant and employer will absolutely honestly get what they wanted.

Possible implementation problems and solutions:

Problem 1: Possibility of prosecution in connection with the law on the protection of personal data or in connection with something else.

This problem is solved by anonymity, because A “certificate” (in the form of all work done) may not contain any personal data. With online training, it is still impossible to establish exactly who performed these works. This system can be called the beautiful word "portfolio", it is not tied to any university and is not even tied to the Ministry of Education with its educational institutions and educational institutions.

Problem 2:Trolling. Anonymous participants in the system can sabotage the process by sending unreadable control, inappropriate images, or performing deliberately incorrect checks.

To combat this, the system should calculate the “ratings” and take them into account when assigning a check fee:

  1. For a student’s incorrectly checked work, the “test rating” will decrease and his “salary” will be less for the next checks.
  2. Sending inappropriate pictures will be blacklisted and moderated for a separate “fee” by points.
  3. For poorly executed work, the student will have a lower “design rating”, and they will pay a large fee for checking the next control (and, accordingly, pay more for those who will check such work).

Thus, asocial behavior will lead to fines, therefore, to the need to check more and more other people's work, and will become unprofitable.
The task of calculating a fair check fee and the task of adequately assessing the results of verification are nothing more than a standard task from applied statistics, and it has been studied quite well.

Fair payment for checking complex or poorly executed works can be assigned by the “ second price auction ” method : Students (or cheaters) first look at the work, offer a fee for checking and thus participate in the auction.

Problem 3:When you start the system and / or at the beginning of a new topic, there will be many people who want to send work for verification, but who do not have enough points. And vice versa, there will be many who want to earn points, but are not ready to check the work on this topic.

Usually assignments have some natural hierarchy (for example, division takes place after multiplication, and multiplication after addition), so conditional undergraduates can check the tasks of conditional undergraduates.

After a while, the reference task bank will accumulate with the standard correct and incorrect decisions, which can be used to determine students' test ratings and “accumulation of initial capital”.

Problem 4: Ordinary tasks from ordinary problem books will soon become known along with the solutions and become unsuitable.

As a source of tasks for this system, any task generators are ideal. A lot of them have already been developed [2-14]. There are also open source open source generators [ 12 , 13 ]. In [14] there is a detailed review. The unpopularity of generators in traditional teaching was cleverly explained in [3] (p. 7) , but they will be very useful for the proposed system.

Problem 5: Technical difficulties with the implementation and support of the system in working condition.

The system does not need to be created from scratch. The popular Moodle training system already contains a student self-test mechanism (in the Russian-language localization of Moodle this is called a “resource seminar”), making the proposed changes should not be difficult.
Also, this training system can be implemented as a bot in one of the messengers, since many messenger-related technical problems, such as registering, sending, receiving and storing images, the messenger takes care of.

The intended interface of the proposed system might look like this:

  1. A teacher in the role of a teacher forms a chain of tasks in several versions (using any task generator or manually) and uploads it to the system. These assignments are sent to students according to a schedule.
  2. ( ), ( ), .

    :

    a) « » (, )
    b) ,  « ».
  3. , ( -), ( ) .  , , , .

  4. a) , , (/);
    b) ( ) . « », .

The practical implementation of such a system on the basis of the messenger can be a good thesis of a student - a programmer.

Financing


This educational service has a natural, interested client - the employer, who is interested in specialists who are able to work in their specialty. In the current reality, these employers have to conduct an exam on hiring themselves, spending their resources on it. A certificate for a student (or cheater) in the form of a set of work performed is valuable information for the employer, it is natural if employers pay for it, thereby paying for the operation of the system.

Organizations paying for the so-called “target recruitment” probably need control over their students, and the proposed system will be quite appropriate for them.


  1. , (, , ) , , – .
  2. « – – », -   .
  3. [2].
  4. . 

, , , . «» [15]. , , , « », «» - . .

, , .
, , « » , ,   .

( ) :

  1. ,   «, », , , « », , .
  2. ( ) , , , . , , , .
  3. . «» - , «» , « », . , , ( )/( ) , .

:

  1. , « », .
  2. ( « ») .
  3. , . , , , . . , .


[1]  . . , ? 
[2] .. // № 2/41, 2016. — . 65–67.
[3] .. // - « », 2018 №3.
[4] .., .. // . 2014. № 1. . 14–25.
[5] .., .., .. // . 2018. № 3 (71). . 5–11.
[6] .. // , 2013. № 6. . 54–59.
[7] .., .., .. // : . 2013. № 5(17). 14 . : engjournal.ru/catalog/pedagogika/hidden/740.html ( : 06.12. 2019). 
[8] . ., . ., . .  // . 2013. № 2. . 181–185.
[9] . ., . ., . ., . ., . . // . 2015. № 2 (2). URL: www.science-education.ru/ru/article/view?id=22636 ( : 21.01.2019).
[10] . . (-) // - «». – 2016. – № 8 (). – . 9–14. – URL: http://e-koncept.ru/2016/16158.htm.
[11] . ., . . // . 2019. № 2. . 65–71; doi: 10.32603/2071-2340-2019-2-65–71
[12]   .. .
[13 ] .,  .
[14] .. // . 2014. . 17. № 4. . 593–609.
[15] . // ( ) [ ].

More articles:


All Articles