Description of creating a program for testing. We create interactive tests and lectures for financial professionals

Test (from the English test - “test”, “test”) - standardized, short, time-limited tests designed to establish quantitative and qualitative individual differences.

One of the considerations underlying the creation of tests is to have a tool for quick and relatively accurate assessment of large contingents of subjects. The requirement to save time becomes natural in mass processes, which is what education has become.

Testing in pedagogy performs three main interrelated functions: diagnostic, teaching and educational:

• The diagnostic function is to identify the level of knowledge, skills and abilities of the student. This is the main and most obvious testing function. In terms of objectivity, breadth and speed of diagnosis, testing surpasses all other forms. pedagogical control.
• The educational function of testing is to motivate the student to intensify work on the assimilation of educational material. To enhance the learning function of testing, additional measures to stimulate students can be used, such as the distribution by the teacher of an approximate list of questions for self-study, the presence of leading questions and hints in the test itself, a joint analysis of the test results.
• The educational function is manifested in the frequency and inevitability of test control. It disciplines, organizes and directs student activities, helps to identify and eliminate gaps in knowledge, forms the desire to develop one's abilities.

Testing is a fairer method, it puts all students on an equal footing, both in the control process and in the evaluation process, practically eliminating the subjectivity of the teacher.

Actual direction modern organization test control is an individualization of control, leading to significant savings in testing time. The main costs during testing are for the development of high-quality tools, that is, they are of a one-time nature. The cost of conducting the test is much lower than with written or oral control.

MyTest is a software system - a student testing program, a test editor and a results log - for creating and conducting computer testing, collecting and analyzing results, and grading according to the scale specified in the test.

The program is easy and convenient to use. All students quickly and easily master it.

To create tests, there is a very convenient test editor with a friendly interface. Any subject teacher, even a beginner computer savvy, can easily create their own tests for the MyTest program and use them in the classroom.

If you have a computer network, you can organize a centralized collection and processing of test results using the log module. The results of the assignments are displayed to the student and sent to the teacher. The teacher can evaluate or analyze them at any time convenient for him.

The program works with seven types of tasks: single choice, multiple choice, ordering, matching, manual number entry, manual text entry, selection of a place in the image.

Each test has an optimal testing time, the decrease or excess of which reduces the quality of the test. Therefore, in the test settings, there is a restriction on the execution time of both the entire test and any answer to the task (for different tasks you can set different times).

Test parameters, tasks, images for tasks - everything is stored in one test file. No databases, no extra files - one test - one file. The test file is encrypted and compressed.

And finally, with the right choice control material the content of the test can be used not only for control, but also for learning. The use of test tasks in automated control and training programs allows the subject to independently detect gaps in the structure of his knowledge and take measures to eliminate them. In such cases, we can talk about the significant training potential of test items, the use of which will become one of the effective directions for the practical implementation of the principle of unity and the relationship between training and control. When you turn on the learning mode, the student receives information about his mistakes and correct answers.

With the help of programs, you can organize both local and network testing. Do what is more convenient for you.

More information about the program and latest version You can find the programs at http://mytest.klyaksa.net - a section of the [email protected] information and educational portal dedicated to this program. Questions, suggestions, error messages, send your tests by e-mail: [email protected] or ask on the forum in the "computer testing" section.

All rights to the MyTest program belong to its author. The author of the program: Bashlakov Alexander Sergeevich, Unecha, Bryansk region.

The MyTest program is distributed for free (Freeware). Any educational institution, teacher and student can use the program for free on the basis of a license agreement without any monetary deductions. For more detailed information For the rights to use and distribute the program, see the license agreement.

Program website -

See the full text of the material MyTest - the best free Russian test creation program in the downloaded file.
The page contains a snippet.

A significant part of the production process relies on software testing. What it is and how such activities are carried out will be discussed in this article.

What is called testing?

This is understood as the process during which the software is executed in order to detect places of incorrect functioning of the code. For achievement best result difficult input data sets are deliberately constructed. the main objective The verifier is to create optimal opportunities for failure Although sometimes testing of the developed program can be simplified to the usual verification of the functionality and performance of the functions. This saves time, but is often accompanied by unreliability. software, user dissatisfaction and so on.

Efficiency

How well and quickly bugs are found has a significant impact on the cost and duration of developing software of the required quality. So, despite the fact that the testers receive wages several times less than programmers, the cost of their services usually reaches 30 - 40% of the cost of the entire project. This is due to the number of personnel, since looking for a mistake is an unusual and rather difficult process. But even if the software has passed a solid number of tests, there is no 100% guarantee that there will be no errors. We just don't know when they will show up. To incentivize testers to choose types of testing that are more likely to find an error, various means of motivation are used: both moral and material.

Approach to work

The optimal situation is when various mechanisms are implemented to ensure that there are no errors in the software from the very beginning. For this, it is necessary to take care of competent architecture design, clear terms of reference, and it is also important not to make adjustments to communications when work on the project has already begun. In this case, the tester is faced with the task of finding and identifying a small number of errors that remain in the final result. This will save both time and money.

What is a test?

This is an important aspect of the verifier's activity, which is necessary for the successful identification of flaws in the program code. They are necessary in order to control the correctness of the application. What is included in the test? It consists of their initial data and values, which should be obtained as result (or intermediate). In order to more successfully identify problems and inconsistencies, tests should be written after the algorithm has been developed, but programming has not begun. Moreover, it is desirable to use several approaches when calculating the necessary data. In this case, the probability of finding an error increases due to the fact that you can examine the code from a different point of view. Comprehensive tests should provide verification of the external effects of the finished software product, as well as its operation algorithms. Of particular interest are the limiting and degenerate cases. Thus, in the practice of activity with errors, it is often possible to find out that the cycle works one time less or more than it was planned. It is also important to test the computer, thanks to which you can check the compliance desired result on various machines. This is necessary in order to make sure that the software can run on all computers. In addition, testing the computer on which the development will be performed is important when creating multi-platform developments.

The art of finding bugs

Programs are often aimed at working with a huge amount of data. Does it really need to be created completely? No. Wide use acquired the practice of "miniaturization" of the program. AT this case there is a reasonable reduction in the amount of data compared to what should be used. Let's look at an example: there is a program that creates a 50x50 matrix. In other words, you must manually enter 2500 thousand values. This is certainly possible, but it will take a very long time. But to check the performance, the software product receives a matrix whose dimension is 5x5. To do this, you will need to enter 25 values ​​already. If in this case normal, error-free operation is observed, then this means that everything is in order. Although here, too, there are pitfalls, which lie in the fact that during miniaturization, a situation occurs, as a result of which changes become implicit and temporarily disappear. It is also very rare, but it still happens that new errors appear.

Pursued goals

Software testing is not an easy task due to the fact that this process cannot be fully formalized. Large programs almost never have the necessary precise reference. Therefore, a number of indirect data are used as a guide, which, however, cannot fully reflect the characteristics and functions of software developments that are being debugged. Moreover, they must be selected in such a way that the correct result is calculated even before the software product is tested. If this is not done in advance, then there is a temptation to consider everything approximately, and if the machine result falls into the expected range, then it will be accepted wrong decision that everything is correct.

Verification in various conditions

As a rule, testing of programs takes place in the volumes that are necessary for a minimum verification of functionality within a limited range. The activity is carried out with a change in the parameters, as well as the conditions of their work. The testing process can be divided into three stages:

• Check in normal conditions. In this case, the main functionality of the developed software is tested. The result obtained should be as expected.
• Checking in emergency conditions. In these cases, it implies the receipt of boundary data, which can adversely affect the performance of the created software. An example is the work with extremely large or small numbers, or in general, the complete absence of information received.
• Exception check. It involves the use of data that lies beyond the processing. In such situations, it is very bad when the software perceives them as suitable for calculation and produces a plausible result. Care must be taken to reject any data that cannot be correctly processed in such cases. It is also necessary to provide information to the user about this.

Software testing: types

It is very difficult to create software without errors. This takes a significant amount of time. To get a good product, two types of testing are often used: "Alpha" and "Beta". What are they? When talking about alpha testing, then it means a check that the development staff itself conducts in “laboratory” conditions. This is the last stage of verification before the program is distributed to end users. Therefore, developers are trying to turn around to the maximum. For ease of operation, data can be logged to create a history of problems and fixes. Beta testing refers to the delivery of software to a limited number of users so that they can operate the program and identify missed bugs. A feature in this case is that often the software is used for purposes other than its intended purpose. Due to this, malfunctions will be detected where nothing was previously noticed. This is quite normal and you don't need to worry about it.

Completion of testing

If the previous stages were successfully completed, then it remains to conduct an acceptance test. In this case, it becomes a mere formality. During this check, it is confirmed that no additional problems are found and the software can be released to the market. The more important the final result will be, the more carefully the check should be carried out. It is necessary to ensure that all stages are completed successfully. This is what the whole testing process looks like. Now let's get into the technical details and talk about useful tools like test programs. What are they and in what cases are they used?

Automated testing

Previously, it was believed that the dynamic analysis of the developed software is too heavy an approach that is inefficient to use to detect defects. But due to the increase in the complexity and volume of programs, the opposite view has emerged. Automated testing is used where availability and security are the top priorities. And they should be for any input. Examples of programs for which such testing is appropriate include the following: network protocols, web server, sandboxing. We will next look at a few samples that can be used for such activities. If interested free programs testing, it is rather difficult to find quality ones among them. But there are hacked "pirated" versions of well-established projects, so you can turn to their services.

Avalanche

This tool helps to detect defects by testing programs in dynamic analysis mode. It collects data and analyzes the execution path of the developed object. The tester is provided with a set of input data that causes an error or bypasses a set of existing restrictions. Due to the presence of a good verification algorithm, a large number of possible situations are developed. The program receives various sets of input data that allow you to simulate a significant number of situations and create conditions where a failure is most likely to occur. An important advantage of the program is the use of a heuristic metric. If there is a problem, then an application error is found with a high probability. But this program has limitations like checking only one labeled input socket or file. When performing an operation such as testing programs, it will contain detailed information about the presence of problems with null pointers, infinite loops, incorrect addresses, or malfunctions due to the use of libraries. Of course it's not full list detected errors, but only their common examples. Unfortunately, the developers will have to fix the shortcomings - automatic tools are not suitable for these purposes.

KLEE

This is good program for memory testing. It can intercept about 50 system calls and a large number of virtual processes, thus executing in parallel and separately. But in general, the program does not look for individual suspicious places, but processes as much as possible possible number code and analyzes the data transfer paths used. Because of this, the testing time of the program depends on the size of the object. When checking, the bet is made on symbolic processes. They are one of possible ways performing tasks in the program being tested. Thanks to parallel work, you can analyze a large number of options for the work of the application under study. For each path, after the end of its testing, the sets of input data from which the test began are saved. It should be noted that testing programs using KLEE helps to identify a large number of deviations that should not be. It can find problems even in applications that have been in development for decades.

Today, not a single teacher can do without tests, whether it be a school teacher or a professor of a multi-thousand online course. Choosing from several options is not the only possibility: you can insert pictures and videos into surveys, change task formats almost indefinitely. Edutainme has chosen seven services that will facilitate the preparation and conduct of testing.

Google Forms

Google Forms is part of the Google Drive office toolkit. Perhaps this is one of the fastest and simple ways create your own survey or test: write a task, select the type of answer (choose from several options, write your own) - done! The resulting test can be sent to students by e-mail or embed on your site using a special code. To speed up the work, we recommend adding the Flubaroo plugin - it automatically checks student answers and grades according to specified criteria. Forms are completely free - you only need to have a Google account to use the resource.

With Quizlet, you can create quizzes where students choose the correct answers, match images and information, or write in their own answers. It is very easy to use - you can understand the interface in just a couple of minutes, in addition, Quizlet supports Russian and works on Android and iOS. Quizlet is free, but there's also a $10 Plus subscription that lets you upload your own pictures and create an unlimited number of study groups.

Proprofs

Proprofs prepares tests for every taste - you can offer one or more options to choose from, ask you to fill in a missing word or write a detailed answer. The service allows you to insert text documents and presentations, PDF files, as well as images, audio and video files into tasks. Once you've completed your test, you can share it on the Proprofs website or embed it on your page.

Although the service is free, Proprofs features are expanded in paid plans. Teachers should pay attention to the Basic and Professional tariffs. The first opens all the main features of the service and allows you to create an unlimited number of tests for $20 per month; the second allows you to combine students in closed groups and will cost $40. New users can use all Proprofs features for free for 15 days after registration.

Kahoot! allows you to submit in the format of surveys and tests almost the entire educational material. To forge feedback with students, you can beat new topics in the form simple questions and answers, and consolidate knowledge through more detailed testing. Kahoot! designed for use in the classroom - the teacher shows the material on the main screen, while students answer questions and discuss information using a special client for computers or a browser on smartphones (Android, iOS, Windows Phone). In order to enter the virtual classroom, students must enter a special code sent by the teacher. The service allows you to find out how each student answered the questions, or build progress charts for the entire class. The students themselves can monitor their results in special tables. Kahoot! free and fully accessible after registration.

ClassMarker

In ClassMarker, you can do polls with different formats answers - in addition to the usual options, there is even an essay. To get started, the teacher needs to create a virtual classroom and send invitation codes to students. ClassMarker stores the results of all tests taken, keeping performance statistics. If the teacher has his own web page, he can embed test tasks on it.

The free version of ClassMarker allows you to create a maximum of 100 tests. 400 tests per month will cost $16.50, and 1000 tests will cost $33. The service has annual packages for those who rarely test online. Minimal amount tests (50 per year) will cost $25 per year, and the maximum (5000 per year) will cost $1000.

A mobile application for teachers that helps to organize surveys right in the classroom. Students are given special forms with answer options (A, B, C and D) - when they hear a question, they raise the necessary cards, which the teacher scans with a smartphone camera. Plickers allows you to analyze the results of an individual student or study statistics for the entire class. The app works on Android and iOS and is free to download.

Easy Test Maker

The most interesting thing in Easy Test Maker is the ability to create tasks where you need to choose true and false statements. Texts can be formatted for ease of reading on a tablet, or downloaded in .pdf or .doc formats for testing in a more traditional format. The service can mix questions and answer options to make it harder for students to cheat.

The free version of Easy Test Maker allows you to create 25 tests without the ability to export to "paper" formats. In the Plus plan ($44.95 per year), you can check English spelling, create an unlimited number of tests, and export to offline formats. With a Premium subscription ($74.95 per year), the service will automatically check the results, and will also allow you to set a time limit for completing tasks and attach graphic files to surveys.

Egor Antonenkov

SunRav TestOfficePro is a knowledge control and testing software. Allows you to create tests different levels complexity and direction; conduct testing efficiently and comfortably for users and organizers.

TestOfficePro works in a local network and on computers that are not connected to the network (it is possible to test from flash drives, CDs without installing the program).

SunRav TestOfficePro

SunRav TestOfficePro is included in the Unified Register Russian programs for electronic computers and databases.

Computer knowledge testing

Pedagogical testing is a widely used method for monitoring knowledge and evaluating learning outcomes. It is used to determine the level of knowledge, for learning, for motivation for learning. With mass checks and self-preparation, the computerized form of testing is the most effective. It is with computer testing that it is easy to correct knowledge simultaneously with verification. After each incorrect answer, the program sends the student to the block of information he needs. Useful for pedagogical control Additional Information, for example, a detailed report for each test-taker - what questions the student received, how much time he spent on a separate answer, how he answered ...
Both high-quality test tasks and educational material, as well as professional software products, are very important in the testing process.

SunRav TestOfficePro makes it easy to create and apply tests in any academic discipline, subject school curriculum, tests to determine professional competence, staff certification, psychological tests.

The program includes

• tMaker- a program for creating tests. It is possible to accompany questions and answers with images, animations, audio and video clips. Compose and edit tests in this program can be a computer user with any level of training. It is possible to import tests created in a text editor (for example, in MS Word) or an editor spreadsheets(for example, in MS Excel).
• tTester- a program for testing. Has the most simple interface. Extensive settings and command line options allow you to adapt its work to any requirements.
• tAdmin- a program for managing users and processing the results of computer testing. Provides the ability to view/print results, as well as create, edit, export, print test reports for groups and/or individual users. It is possible to create a matrix of answers.

Key Features of SunRav TestOfficePro

Creation of tests. Work with test tasks

Compose tests for any purpose, a variety of types and types will help following features programs:

• Variety of question types. The program allows you to use 5 types of questions in tests:
  1. single choice. One answer out of several options.
  2. Multiple Choice. One or more answers from several options.
  3. Open question. The user must enter the answer from the keyboard. The quiz creator can use a powerful templating language to correctly evaluate the user's response.
  4. Conformity. The user needs to arrange the statements in the two lists so that they correspond to each other.
  5. ordered list. Arrange the answers in the list in a certain order.


• adaptive tests. The order of the questions can be not only linear, but also depend on the user's answers.
• Using Themes. The program can break the test into several topics. Knowledge can be assessed both for each topic separately and for the test as a whole.
• Comments on the question. Each question can be accompanied by a comment with testing instructions, a hint, information about the correct answer, etc.
• Options for reactions to the user's response:
  1. No response - the user is prompted to respond to next question.
  2. A message that the user answered correctly or incorrectly.
  3. Show any document related to the question. In particular, it can explain in detail the reason for the incorrect answer and provide additional material, which will allow you to study the issue in more depth.


• The weight of the question and answer options. Each question and answer can have its own "weight". This allows the user to be awarded more points for correct answers to difficult questions and less points for answers to easy questions.
• Visualization. The program allows you to insert images, formulas, charts, tables, audio and video files, HTML documents and any OLE documents. The built-in text editor in tMaker for writing tests by function resembles MS Word familiar to many.

Ensuring the objectivity of testing

• Setting up the testing process. Administrator can:
  1. Prohibit exiting the program until the end of testing.
  2. Disable desktop and taskbar access.
  3. Close the program after passing one test.


• random questions. Questions in the test can be mixed. In addition, the quiz creator can determine how many questions from each topic the user will receive. Let's say the topic consists of 300 questions. If only 30 questions are randomly selected, test takers will receive a completely different sets questions from the same test. The answer options for each question can also be mixed. Thus, with enough in large numbers questions, the test-taker cannot thoughtlessly use prior knowledge of the test, with a predetermined sequence of questions and answers. Note: the described function is set by selecting "Examination mode" in the settings, see the documentation for the program.
• Introduction of time frames. Testing can be limited in time - both for the test and for each question. The amount of time allocated for each question may vary.
• Setting the information visible during testing. In the program, you can determine whether to show the user:
  1. The number of correct answers.
  2. Time until the end of testing.


• Limit on the number of testing attempts. In the program, you can set the maximum number of passing one test.
• Tracking test attempts in the program

Software testing is the evaluation of the software/product being developed to check its capabilities, capabilities and conformity with the expected results. Exist Various types methods used in testing and quality assurance about them and will be discussed in this article.

Software testing is an integral part of the software development cycle.

What is software testing?

Software testing is nothing more than testing a piece of code to controlled and uncontrolled operating conditions, observing the output, and then examining whether it meets predefined conditions.

Different sets of test cases and testing strategies aim to achieve the same common purpose- elimination of bugs and errors in the code, and ensuring accurate and optimal software performance.

Test Methodology

The commonly used testing methods are unit testing, integration testing, acceptance testing, and system testing. The software is subjected to these tests in a specific order.

3) System testing

4) Acceptance tests

The first step is a unit test. As the name suggests, this is an object-level test method. Individual software components are tested for errors. This test requires precise knowledge of the program and each installed module. Thus, this verification is done by programmers, not testers. To do this, test codes are created that check whether the software behaves as intended.

Individual modules that have already been unit tested are integrated with each other and checked for faults. This type of testing primarily identifies interface errors. Integration testing can be done using a top-down approach, following architectural structure systems. Another approach is the bottom-up approach, which is done from the bottom of the control flow.

System testing

In this testing, the entire system is checked for errors and bugs. This test is carried out by interfacing the hardware and software components of the entire system, and then it is tested. This testing is listed under the "black box" testing method, where the user's expected operating conditions of the software are verified.

Acceptance tests

This is the last test that is carried out before handing over the software to the client. It is carried out to ensure that the software that has been developed meets all customer requirements. There are two types of acceptance testing - one that is carried out by members of the development team is known as internal acceptance testing (Alpha testing), and the other that is conducted by the customer is known as external acceptance testing.

When testing is done with the help of prospective customers, it is called customer acceptance testing. When testing is performed by the end user of the software, it is known as acceptance testing (beta testing).

There are several basic testing methods that form part of the software testing regimen. These tests are usually considered self-sufficient in finding errors and bugs throughout the system.

Black box testing

Black box testing is done without any knowledge inner work systems. The tester will incentivize the user environment software by providing various inputs and testing the generated outputs. This test is also known as black-box, closed-box testing or functional testing.

White box testing

White box testing, unlike black box testing, takes into account the internal functioning and logic of the code. To perform this test, the tester must have code knowledge to know the exact part of the code that has errors. This test is also known as White-box, Open-Box or Glass box testing.

Gray box testing

Method Testing gray box or Gray box testing, this is something between White Box and Black Box testing, where the tester has only general knowledge of this product necessary to perform the test. This verification is done through documentation and information flow diagram. Testing is conducted by the end user, or users who appear to be end users.

Non-functional tests

Application security is one of the main tasks of the developer. Security testing checks software for confidentiality, integrity, authentication, availability, and non-repudiation. Individual tests are carried out in order to prevent unauthorized access to the program code.

Stress testing is a technique in which software is exposed to conditions that are outside the scope of normal conditions software work. After reaching critical point, the results are recorded. This test determines the stability of the entire system.

The software is tested for compatibility with external interfaces such as Operating Systems, hardware platforms, web browsers, etc. The compatibility test checks if the product is compatible with any software platform.

As the name suggests, this testing technique tests the amount of code or resources used by a program to perform a single operation.

This testing tests the aspect of user friendliness and usability of the software. The ease with which a user can access a device forms the main point of testing. Usability testing covers five aspects of testing - learnability, performance, satisfaction, memorability, and errors.

Tests in the software development process

The waterfall model uses a top-down approach, whether it is used for software development or testing.

The main steps involved in this software testing methodology are:

• Needs Analysis
• Design Test
• Implementation test
• Testing, debugging and verifying code or product
• Implementation and maintenance

In this technique, you only move to the next step after you have completed the previous one. The model uses a non-iterative approach. The main advantage of this methodology is its simplified, systematic and orthodox approach. However, it has many drawbacks, as bugs and bugs in the code will not be discovered until the testing phase. This can often result in wasted time, money, and other valuable resources.

Agile Model

This methodology is based on a selective combination of a sequential and iterative approach, in addition to a fairly large variety of new development methods. fast and progressive development is one of the key principles of this methodology. The emphasis is on getting quick, practical, and visible outputs. Continuous customer interaction and participation is an integral part of the entire development process.

Rapid Application Development (RAD). Rapid Application Development Methodology

The name speaks for itself. In this case, the methodology takes a rapid evolutionary approach, using the principle of component construction. After understanding the various requirements this project, a rapid prototype is prepared, and then compared to an expected set of output conditions and standards. Necessary changes and modifications are made after a joint discussion with the customer or development team (in the context of software testing).

While this approach has its share of advantages, it may not be appropriate if the project is large, complex, or extremely dynamic in nature, in which requirements are constantly changing.

spiral model

As the name suggests, the spiral model is based on an approach that has whole line cycles (or spirals) from all successive steps in the cascade model. After the initial cycle is completed, a thorough analysis and review of the achieved product or output is carried out. If the output does not meet specified requirements or expected standards, a second cycle is performed, and so on.

Rational Unified Process (RUP). Rational unified process

The RUP methodology is also similar to the spiral model, in the sense that the entire testing procedure is divided into several cycles. Each cycle consists of four stages - creation, development, construction, and transition. At the end of each cycle, the product/yield is reviewed and a further cycle (consisting of the same four phases) follows as needed.

The use of information technology is growing every day, and the importance of proper software testing has grown significantly. Many firms maintain staff for this special teams, whose capabilities are at the developer level.