Methods to prevent fraud attempts in online assessments
Especially since the Corona crisis, more and more companies are using Online assessments, to select the right talent for your company Online assessments make it possible to carry out a wide range of different things, regardless of time and location Test procedure over the Internet. The methods range from measurements of cognitive abilities to knowledge tests and job-related personality tests.
But the advantages of such digital solutions, independent of time and location, are not only evident in the Corona crisis. Compared to physical selection processes carried out on site, online assessments are particularly convincing due to their high objectivity and economic efficiency (see also: Test quality criteria in personnel selection): Companies receive an objective basis for evaluating the applicant's abilities and also benefit from high Time and cost saving potentials. Because personal selection days and Assessment Center are usually very time-consuming and expensive. And it's not just companies that save time and money — applicants also benefit from greater flexibility and save themselves long journeys to application days. That sounds like an ideal solution for everyone involved. But is this really true? Do online assessments have the potential to completely replace physical selection processes?
Even though aspects of economic efficiency clearly speak in favour of this, a central weakness of online testing methods has so far proved against it: The increased risk of fraud. In contrast to selection processes carried out on site, which usually take place under physical supervision, the applicant in a digital test situation is often unobservable. Many HR managers therefore fear that online selection processes could become a paradise for cheaters. Because applicants are, of course, highly motivated to get a good test result in order to get their dream job. So why not research the answers to a knowledge test on the Internet, ask a friendly mathematics student for help with a math problem, or consult a native speaker for a language test? The skepticism of many HR managers does not seem entirely unfounded.
To your dream job through clever cheating in the selection test? Is this really successful?
Typically, the answer is: No.
At least not if some important precautions are taken when designing the test process. Because with the increase in online testing procedures, so does the number of procedures that enable fraud behavior to be detected. We have selected some of the most interesting processes that could also be exciting for you.
A total of two different types of fraud can be differentiated, the probability of occurrence of which is professional personnel selection It is important to reduce. On the one hand, there is a risk that applicants will deliberately manipulate the test result (also as Faking ) by using unauthorized tools such as textbooks or sources on the Internet. On the other hand, there is a risk that applicants will seek help with the test or even ask another person to complete the entire test process (including as Impersonation referred to).
I. Methods to prevent faking
a) Communication
In order to reduce faking, companies can do a great deal when it comes to communication. For example, the test is not recommended as a selection test but as Self-Assessment to announce to test your own abilities. Because if applicants believe that the test is primarily about identifying their own strengths and weaknesses, the motivation to cheat is naturally lower. More and more companies are also demanding a affidavit from the applicant that the test was completed without unauthorized tools. Because similar to falsifying certificates, cheating during an online assessment can also be classified as an attempt at fraud in the sense of employment law. Warnings that cheating can be detected in the test have also been shown to be effective (e.g. Dwight & Donovan, 2003).
b) Test setup
Designing the test process can also help to reduce the risk of faking to a minimum. To prevent applicants from using solution patterns, the questions should be presented in random order, for example, and the test content should be slightly varied between applicants, for example by exchanging numbers in math problems (this is also known as “permutating test content”). Time restrictions can also help prevent the use of unauthorised tools. So-called Rapid Response Measures, which often require an answer from the applicant in just a few seconds (Mäde et al., 2020). Research information in textbooks or on the Internet? DThe applicant simply has no time for that.
c) Software
In addition to design options in terms of communication and test setup, companies are increasingly benefiting from modern software solutions, which can be used to detect fraud. For example, a change of browser (e.g. away from the test process to a website, see “PageFocus”) can be recognized and considered an indication of faking. For data protection reasons, only the change of browser, but not the page accessed, is recognized and saved.
II. Method for preventing impersonation
It becomes more difficult when it comes to preventing impersonation, i.e. unauthorised help from third parties. While many companies are already using methods to prevent faking profitably, they are still struggling to implement methods to prevent impersonation. Particularly in personnel selection, however, it is of course essential to ensure that it is actually the applicant who is completing the test and not someone else. In recent years, scientists have therefore developed various new methods to authenticate applicants in selection processes:
a) Physical properties
One way to ensure that the applicant is actually completing the test and not someone else is to record the applicant's physical characteristics at the beginning or during the testing process. Two important methods for personnel selection are fingerprint authentication (e.g. fingerprint scanner on smartphone) and facial recognition (e.g. webcam recordings). As part of facial recognition, applicants are either observed during the entire test situation (similar to a physical procedure) or pictures of the applicant are taken at various, randomly selected points in time. The pictures taken are then compared with the applicant profile. The advantage of these methods is, of course, that they enable almost 100% fraud detection — because physical properties are not transferable (Sabbah et al., 2011). However, the disadvantages include the necessary provision of technical equipment (e.g. webcam, fingerprint scanner) and possible data protection concerns. The storage of data on secure servers and the deletion of all data after the selection process has been completed must be ensured at all times.
b) Knowledge bases
Other methods aim for authentication based on the applicant's unique knowledge base. During the test process, he or she is asked questions that are highly unlikely to be answered by a third person. Suitable questions can be generated from information in application documents (e.g. information in the curriculum vitae) or in upstream learning or test environments (this is also referred to as Dynamic Question Profiling). Research shows that these methods are very suitable for recognizing processing by third parties (Ullah et al., 2019). The disadvantage of the process is that joint processing in working groups cannot be completely ruled out.
c) Behavioral characteristics
Another option is to measure the applicant's behavioral characteristics (such as keystrokes, mouse movements, behavior, and response patterns) and use them for authentication. For example, a user-specific authentication model can be calculated from the behavior of the applicant in a short pre-placed mini-game (Mohamed & Saxena, 2016). Thanks to continuous progress in the area of machine learning Such user profiles can usually be determined in just a few minutes and with high accuracy (e.g. using artificial neural networks or random forest models). The candidate's behavior during the testing process is then regularly compared with the created user profile. A significant deviation (determined using equidistance measures) can be regarded as an indication of impersonation. Previous research supports the success of this process: This is how a Detecting impersonation in 95-97% of cases, and this is based solely on the data collected in an entertaining mini-game (Mohamed & Saxena, 2016). It is also possible to carry out the test during a final on-site interview in order to finally confirm the identity of the applicant. The advantages of these processes are that they are carried out in a timely and cost-effective manner and that they comply with strict data protection guidelines. However, use requires expertise in the area of machine learning and therefore usually cooperation with a provider such as Aivy, who has the appropriate know-how to create appropriate user profiles.
Recognize attempts at counterfeiting and find “real” talent?
As the brief insight into some of the many existing fraud detection methods shows, this can also be achieved in a purely online environment. It is recommended that the procedures be tailored to the target group, the position to be filled and the test methods used. Would you like to make your selection process fraud-proof and leave “cheaters” no longer a chance? We are happy to help you!
