What Do We Mean by e-Assessment?
e-Assessment is often seen as providing a partial solution to providing assessment for increasing numbers of students and declining staff to student ratios (Sim et al., 2004). In addition, students may experience cognitive conflict because they are generally expected to word process essays and engage in online tasks but use pens in examination halls (Brown et al, 1997) such that we are training them in one system and testing them in another. Gipps (2003: 26) reasons that:
'If teaching and its associated resources become electronic, then assessment too will need to take that route, to ensure alignment between the modes of teaching and assessment.'
When e-assessment is mentioned people often have certain assumptions, that:
The assessment will be objective.
The assessment will be limited in the type of question to multiple choice questions (MCQs).
The assessment can only test recall or low level learning outcomes.
We will address each of these in turn.
Assumption 1: Objective assessment
There is no such thing. Subjective judgment is always involved - when an educator creates a test they do so with their internal biases on the type and nature of material. When the limit of the assessment and the type and nature of the 'correct' answers are preset, the educator introduces their own judgement and bias into the system from the start.
However the extent of bias can be reduced because in e-assessment, the judgments made are only made based upon the original criteria and not on 'human introduced error' (e.g. marking at 2 a.m.) so that a second level of error is not introduced. In addition, levels of correctness can be programmed into the system that can score partially correct marks in a more consistent manner.
Assumption 2: Limited Types of assessment tools
Communication and information technologies have been incorporated into assessment in further and higher education in a number of ways. For example:
the analysis of student postings to discussion boards (Rovai, 2000; MacDonald and Twining, 2002)as discussed in the 'Computer Mediated Conferencing' section of this infoKit
automated essay marking (Christie 1998,1999) and two commercial packages
http://www.intelligentassessment.com/ and http://www.ets.org/research/erater.html)the delivery of exam papers (Pass-IT project and Technologies for Online Interoperable Assessment (TOIA)
use of modified essay questions (MEQs) particularly in Medicine
in helping to provide feedback to students on written work (e.g. Electronic Feedback System developed by Phil Denton from Liverpool John Moores University)
use of Personal Response Systems to assess students' undertanding during a class(e.g. JISC funded PRS project at Strathclyde University by D.Nichol and J.Boyle.
Students can be asked to interact with simulations and submit the results and analysis of these results for assessment. Students can create webpages, these can be assessed by their peers or be validated electronically. Computers can create sounds which students have to replicate, for example in music, or in languages, and the computer can gauge the amount of similarity.
Some of these assessment activities can be offline assessment of online delivery and could include the assessment of a presentation on the web or powerpoint for example, as well as the assessment of online skills, including how students use particular software packages such as databases and spreadsheets.
However, most would accept that e-assessment comes into its own with objective testing (examples can be found from the CAA conferences in 2004 and 2005 available from http://www.caaconference.com), where computers do the marking.
While many of the available systems, particularly those which come as 'quiz' facilities with VLEs or are available free of charge are indeed limited, others include a large variety of question types. [see types of questions and choosing a software package]
Assumption 3: Computerised assessment can only test low level learning outcomes
Even straightforward multiple-choice questions can, if carefully constructed, test higher order skills. The issue here isn't so much that of technology as creativity. For instance if you are looking at the application of knowledge (a higher order skill according to Blooms taxonomy
http://www.le.ac.uk/cc/rjm1/etutor/resources/learningtheories/bloom.html) it is possible to create an ordering question, which might appear easy to a student who has comprehended the topic whereas one who has not will struggle.
