With the increasing popularity of computers, more college instructors have converted from paper-and-pencil to computer-based tests for benefits such as reduced grading effort and the ability to test more frequently (Etrurk, et al. 2004). Computer based assessment has increased in the recent times. Different organizations are using it for different purposes.
According to Etrurk and others (2004), several organizations are currently using computer based assessment for drivers’ license exams, job interviews, certification of exams and entrance exams for post-secondary education. It has been argued that the mode of testing administration affect scores (Clariana & Wallace, 2002). So far, there is evidence that identical paper-based and computer based tests cannot obtain same results. However, study results have been inconsistent (Kingston, 2009,).
Even when Computer-based and paper-based testing with have identical items, they may not produce equivalent measures of student learning. Some research also found that the format of the exam and the subject being tested on also matters. In this essay, I will analyze the current body of research and attempt to determine whether administration via computer yields equivalent scores as paper-and-pencil administration, and further explore the validity of computer-based testing.
Most of the studies so far conducted shows that computer-based tests are more superior to paper-based test. Some instructional quota insists that the two tests methods should give same scores. Practically these methods do not give same scores (Mason et al. 2001). Other studies have concluded that computer-based and paper-based tests are nearly identical (Erturk et al. 2004).
There are several variations of Computer-based and paper-based testing. Some of these differences highlight advantages of one over the other method. In a synthesis of 81 recent studies, Kingston (2009) concludes that the mode of administration does not affect student achievement across a grade level.
However, recent studies by Butters and Walstad (2011) in which multiple-choices answers were used for testing shows that students who completed the test on computers outperformed a paper-based test group and that computer-based testing has the potential to reduce item guessing.
Variations of score have been caused by some discrepancies. Researchers who support the test mode effect argue that previous experiments which showed no differences in achievement and arrived at a different result did so due to differences in distribution of assessment and sample size (Maguire et al. (2010).
Wide range of students gives consistence results than skewed population. They further found out that most students were more comfortable interfacing with computers than using a pencil and pen during their tests; this mitigate students’ test anxiety. Clariana and Wallace (2002) found out that scores variations were caused by settings of the system in computer-based and level of strictness of examiners in paper-based.
The difference in computer-based and paper-based tests can be affected by learner’s characteristics. Learners with poor knowledge of computers are likely to fail an online exam. On the other hand, computer literate student with higher academic attainment taking an online exam is likely to do well (Clariana and Wallace, 2002).
Clariana and Wallace (2002) further examined previous studies which showed that computer based tests on verbal quantitative and analytic sections of all tests were all greater than complementary paper-based test scores. They also found out that computer familiarity is the most fundamental factor in test mode effect for unfamiliar content and low attaining examinees.
They also found out that computer familiarity is the most fundamental factor in test mode effect for unfamiliar content or for low attaining examinees. This is mainly for students with limited access to computer. Higher attaining students will easily adapt to any assessment method and develop test-taking strategies for any new approach. Computer familiarity should become a less factor when all students have become computer literate (Butters et al. 2004).
Although some differences are common, testing methods sometimes have multiple choice questions. Together with multiple choices testing, subjects should be put into consideration before a method is used. Kingston (2009) concludes that English, arts and social sciences had an advantage on computer-based test.
On the other hand, Mathematics and science test favored paper-based test (Kingston, 2009). I have been exposed to computers all my life and been using internet for over ten years. This makes me feel more comfortable in computerized testing environment. I have learned that writing through a computer give better results.
There are several befits associated with computer based testing. This test method saves time for students under it. Choices are mostly made by a simple click of a mouse rather than writing. This makes it easy and fast for students who are familiar with computers and high attaining. Furthermore, examinees do not have to go to examination centers, instead they just can have an internet connection (Noyes and Garlandb 2008).
Another benefit of computer-based test is immediate scoring and reporting of results. The test can be easily monitored since the system records start and end time, amount of time spent on a particular item and integrated survey responses. Tests are recorded and scored electronically with minimum challenges. Results of tests can be accessed by third parties easily in spreadsheets, word and other application packages. The massive data generated makes analysis, access, retrieval and reporting of results easy (Meissner, 2007).
Flexible test scheduling is one benefit of computer-based tests (Meissner, 2007). There is availability of single-day tests administration window to continuous ones. An approach chosen by a student is based on his or her access, publishing frequency and exposure.
Exam administrators have an opportunity to ensure test security, standard setting and forms assembly. Candidates are flexible to take exam at their convenience since it is offered all year round. After administrators create an exam, it becomes hard for content to be jeopardized or items to be compromised.
Item statistics available for tests sponsor are monitored as they are generated. The flexibility of technology provides the testing program with experience by virtue of unique testing experience from various candidates. Various forms of test navigations are used (Noyes and Garlandb 2008)
Instructors and administrators have an opportunity to include innovative item formats that are made possible by use of technology. Computer-based tests engage navigation and presentation of items in several ways. Test authors can pre-define sequence of sections or test-lets through linear by considering performance of previous tests. This testing system affords several unique content presentation styles. Various elements of tests may be presented on one page per item or clustered together according to the preferred method.
There are reduced costs of test production, administration and scoring. Tests are not published as hard copies before they are done by candidates. During tests, no invigilators are required to supervise candidates. Finally, no teachers are required to mark tests before results are released. This reduces the cost of preparing, administering and processing of results.
According to Meissner (2007), use of computer based tests enhances security of the tests and results. The system reduces cumulative impact of item exposure on subsequent administrators. Although the scheduling is flexible, items are exposed and security of tests jeopardized.
This situation can be mitigated by having by several forms at a time. It can also be dealt with by having a smaller number of forms but refreshed them frequently. Security can be enhanced by encrypting the processed data before transmission. To access the data, owners can be subjected to authentication procedure which may include figure prints and digital photograph.
Computer-based tests are consistent and reliable as compared to paper-based tests. Computer-based testing applies high degree of standardization. Policies and procedures are easily adhered to. These structures ensure consistency and reliability of the system. This can be seen in timing which is strictly and rigidly controlled. Furthermore, referencing materials can be incorporated as attachments (Meissner, 2007).
Apart from the several benefits highlighted above, computer-based test has a number of detrimental to students. When computer-based test is made mandatory for all students, those with poor computer skills are affected negatively. They end up scoring lower than what they would have passed had paper-based test been used (Russell et al. 2003).
Computer-based test does not provide examinees with the opportunity to revise their work after marking. This affects students’ future performance negatively. On the other hand, Vispoel et al. (1992) concluded that such tests do not always yield equivalent results. They added that the tests should be equated to ensure use of fair test scores. The amount of control in computer-based tests does not allow review of items by examinee.
Computer based tests do not allow examinee to change their answers as they would have done in a paper-based test (Russell et al. 2003). Previous research has shown that most answers change by student become correct. Therefore, failure by students to change answers when they think they have entered a wrong one, effects negatively on their performance during computer-based test (Russell et al. 2003).
According to Russell et al. (2003), item layout and presentation have some effects on examinee. Tests requiring multi-screens, graphical or complex display have negative effects on examinee performance. Sometimes graphical display issues may be detrimental to performance of examinee. Such issues include screen size, font size and resolutions of graphics of some items. Such situation affects performance of examinee negatively (Russell et al. 2003).
Malfunctioning of computer software or hardware during a test can be detrimental to examinee. When a computer part is changed or computer system is restarted, a student is forced to restart again leading to double submission (Noyes & Garlandb, 2008).
There are ethical issues concerning computer-based tests. Confidentiality of examinee’s work is not guaranteed in online examination. Respondents also tend to create a particular impression of their results (Noyes & Garlandb, 2008).
To conclude, benefits of computer based exams outweigh its detrimental. Benefits are on both sides of the exams. Students or examiners have their benefits, for instance, score and flexibility of exams. Sponsors spend less money to offer such tests because there is no supervision, printing and physical marking. Administrators too have their work simplified as the system requires little supervision.
Detrimental of this testing system are few. However, almost all are on the examinees’ side. As time goes by, more people become computer literate and access computers, most of the tests would be conducted on computer in most places. Negative effects will be less because majority of examinees would be computer literate.
To ensure that it does not disadvantage certain student, computer-based testing needs close cooperation between exam administrators and examinees. Preparing online questions needs a lot of efforts from technical team. Questions should be standardized with student’s level.
Some of the instructional designers are sometimes not equipped with skills to develop the best examination procedure. Sponsors need to ensure that all instructional designers and administrators are qualified to reduce detrimental. If all these issues are attended to, advantages of computer-based testing outweigh detrimental.
The method will be beneficial to instructors and students if well administered. More studies need to be conducted to ascertain some of the variation in scores of computer based and pencil-based tests. Methods and procedures which ensure variations are significantly low should be developed through extensive research.
As long as that the test developers ensure that the composition mode of these test is not putting certain students at a disadvantage and detrimental are reduced to minimum, Computer-based test will provide higher reliability, I believe it is beneficial for both students and teachers to adopt this testing method.
Works Cited
Butters, Roger B., and Walstad, William. “Computer VersusPaper Testing in Precollege Economics.”Journal of Economic Education 42.4 (2011): 366–374. Print.
Clariana, Roy and Patricia Wallace. “Paper-based Versus Computer-based Assessment: Key Factors Associated with the Test Mode Effect.” British Journal of Educational Technology 33.5 (2002): 593-602. Print.
Erturk, Ismail, Yasar Ozden, M. and Refik Sanli. “Students’ Perceptions of Online Assessment: A Case Study.”Journal of Distance Education 19 (2004): 77-93. Print.
Kingston, Neal M. “Comparability of Computer- and Paper-administered Multiple Choice Test for K–12 Populations: A synthesis.”Applied Measurement in Education 22.23 (2009): 31-32. Print.
Maguire, Karen A, Daniel Smith, Sara Brallier, and Linda J. Palm. “Computer-based Testing: A Comparison of Computer-based and Paper-and-pencil Assessment.” Academy of Educational Leadership Journal 14.4 (2010): 117-125. Print.
Mason, Jean B, Patry Marc, and Berstein J. Daniel. “An Examination of the Equivalence between Non-Adaptive Computer Based and Traditional Testing.” Journal of Educational Computing Research 24.1 (2001): 29-39. Print.
Meissner, Dennis. A Successful Conversion: The Benefits and Best Practices of Computer-Based Testing. 2007. Web.
Noyes, Jan Michael, and Kate Garlandb. Computer- vs. paper-based tasks: Are they equivalent? Ergonomics 51.9 (2008): 1352–1375. Print.
Russell, Michael, Goldberg Amie and O’Connor Kathleen. Computer-Based Testing and Validity: A Look Back and Into the Future. 2003. Web.