Category Archives: Assessment

Don’t be Half-Assessed

Assessment is used to measure learning outcomes, but if you see it only as a testing tool you’re missing half its value. Assessment should not be punitive; it should uncover the learner’s strengths and help identify areas for improvement. Well planned and integrated assessment creates opportunities for learners to reflect on their learning, apply new skills and knowledge, and also enables the institution to recognize the value on the learning intervention. Adapting existing models can take you much further than re-creating the wheel. In this post, we examine some guiding principles for effective assessment.

An excellent example of assessment as a way to support learning goals is Adobe’s Certified Associate practice tests. Learners can take a certification exam without any structured practice, but there is also an unlimited certification prep test package. The practice test closely mimics the structure of the final exam. Both combine multiple choice questions and simulations of the Adobe environment. The multiple choice questions require the learner to demonstrate knowledge of design concepts and the production process. The simulations require the learner to complete a task within the simulated software application interface. Keyboard shortcuts are disabled, but otherwise, learners can use any correct method they choose to complete the task, using menus or panels as appropriate.

Another good example is the Sun (now Oracle) Java certification paths. Each path contains a test prep “kit” that includes preparation recommendations, additional resources, a practice test, and a re-take policy. Each path is designed to prepare the learner to achieve a specific level of certification, and used as a benchmark against industry standards. These certifications are recognized by employers and can advance a person in their career.

Accreditation or certification can be used to validate mastery of a topic, but this is not the only way that assessment can be useful. By seeing assessment an integral part of instruction, we can support the learner’s career development while measuring true performance for the business.

Factors that make an ASSESSMENT tool also a useful INSTRUCTIONAL tool are:

  • Authenticity
    • The learner performs the task in context, not recalling theory, but actually demonstrating competency.
  • Open ended
    • Because the end product is assessed, not the method used to get there, learners are able to use whatever menus or panels they choose.
  • Learning while doing
    • Learners use contextual clues and critical thinking to complete tasks. They may not know how to adjust alpha levels in Photoshop, but they may know to investigate the color panel to find them.
  • Self-reporting
    • Learners can mark questions that they’d like to return to if they have time or opportunity.
  • Cumulative time
    • The test is timed with one master clock, not with individual times for certain sections or items.
  • Feedback
    • Learners receive feedback on each item, with notes about the correct answer.
  • Tracking
    • Performance from one practice test to another is tracked.

Assessment is not something that should only occur in a testing situation, indicating pass/fail rates, but it should be integrated throughout instruction to allow the learner to know how they are doing, so they can learn more effectively. You don’t want your learner to leave half-assessed!

A version of this post originally appeared in The Total Learner Experience in August 2011.


Creative Effective Training Evaluation

Drj Dolly: Brandon ranted about ineffective evaluations while I was holed up under the 2 feet of snow here. The snow is still here, but my Internet is finally back. His comments about evaluation made me wonder what I consider effective evaluation to look like. Cumulative evaluation should be as authentic as possible. If you want someone to be an effective salesperson, you’d better identify what competencies are required and then have them replicate and practice as best they can in the classroom environment.

Bwc Brandon: Good point. I concur. However, more and more organizations are hit with the high-cost of travel and the inefficiency of removing people away from their jobs for instructor-led training. We need to provide effective strategies for authentic evaluation in an online format.

Drj So I can think of two ways to recreate these situations in the digital world: virtual simulation and/or role-playing. Learners will have different profiles. Does someone who is technically minded need the same practice as someone who is naturally a people person? Both these individuals need to come out with the same skill sets at the end, but perhaps they need different practice. Formative self assessments combined with flexible course sequencing can allow individuals to focus on their areas of real need.

Bwc That just describes a sales training course I completed using Thiagi’s 4-Door Model. It allowed the learner to “self-adjust” the content based on their own self-leveling of knowledge. It does beg the question though: how do you “branch” an evaluation based on different learner profiles? Can you have an effective evaluation instrument via a dynamic system that presents a contextual series of assessment items based on the learner’s individual profiles?

Drj I think that you can have a self-assessment that gives the learner guidance in their own strengths and weaknesses.

Bwc OK, so you’re talking about self-assessment. What about an actual skills certification? One that can affect a learner’s job status, salary, or in the case of compliance or regulatory situations, a learner’s knowledge that could have life or death consequences? Can a dynamic evaluation instrument provide the appropriate assessment of knowledge?

Drj I don’t know what you mean by “dynamic evaluation instrument”.


Bwc “Dynamic pooling” is when the system displays content based on a learner’s input at the time of input.

Drj What you are talking about now is less about the assessment/evaluation used and more the consequences related to that assessment or evaluation. The SATs are pretty weighty and they’ve been using a dynamic response for years. A student gets progressively harder questions until incorrect responses are entered. Then a cycle commences where easier and harder questions are given to the student until the logarithm determines what the student’s level of mastery is. I’d say that stakes are high, but the College Board feels comfortable using this dynamic response.

Bwc Right. So I’m saying we need a similar system in corporate learning where appropriate assessment techniques seem to be a missing factor. In eLearning, there is a cycle of ineffective self-check systems instead of situational problem-based assessments. For example, a colleague was discussing with me the usage of “misconception” problems in assessment. They can present more authentic situations such as posing the problem in a scenario and requiring the learner to identify the parts of the scenario that are wrong or inaccurate. Even using open-ended question types where the system evaluates based on a keyword or series of keywords can be quite effective. How are we truly able to measure whether knowledge or skills are transferred if we’re not willing to properly evaluate?

Drj I like that because some people are just natural or trained to be good test takers. That kind of exercise actually accurately tests knowledge that they have, rather than their ability to suss out a poorly written stem or item. I know you are all about performance. Shouldn’t people’s training be based upon actual deficits in their performance and knowledge? And if they are high performers or can demonstrate mastery of course materials, shouldn’t they be rewarded by being allowed to choose their own training regimen, anyhow?


This post first appeared in January 2010 at The Total Learner Experience