如果思科将其所有的认证考试改为基于性能的考试，会产生多大的影响呢?答:很多。几个星期前，温德尔·奥多姆在他的NWW博客中提出了这个问题，关于将所有助理和专业级别的思科认证测试从目前的格式更改为现有格式的可能性基于性能的测试。在2007年12月我为温德尔写的博客中，我表达了一些关于认证的价值为了给报告带来一些观点，即技术认证的价值(以美元计算)正在下滑在这个行业。在今天的文章中，我将把这些概念结合在一起，重新审视认证供应商使其证书更有价值的方法。首先，让我们看看在供应商的认证项目中什么对他们来说是重要的。一般来说，他们关心三个概念:
- 他们的认证在行业中被认为是积极的和有价值的。这其中既有声望的因素，也有实际的因素。认为“品牌”。
- 他们的认证对候选人，以及这些候选人的雇主和潜在雇主来说，实际上意味着一些东西——是有关联的。如果行业媒体喜欢某个供应商的认证，这也不仅仅是一件好事，因为这能产生大量的收入和宣传机会。比如，思科系统和思科出版社之间就有很大的协同作用。
- 对于任何认证供应商来说，最大的好处是它推动了个人层面上对该供应商产品和解决方案的偏好。几乎从定义上来说，一个通过获得专业认证而投入大量精力的人就表明了他对供应商的巨大奉献。毕竟，供应商已经证明了那个人的技能，这是对个人努力的明确确认。那些获得认证的人几乎肯定会通过偏爱那个供应商来表达他们的感激之心，并为游戏带来强大的技能和知识。实际上，认证人员成为了供应商劳动力的延伸，为供应商带来了声望和承诺——而不需要像工资和福利这样的直接成本。
想想那些最知名的供应商的技术认证;名单很短。现在考虑一下这些供应商提供的认证数量，以及他们“更好”的认证在市场上的价值。然后乘以成千上万的个体。基于此，很容易看出为什么每个认证供应商都希望将其认证的声誉保持在尽可能高的水平。由于这些原因，每个认证供应商都需要担心其认证的技术严密性、考试安全性、内容的新鲜度、考试交付、培训内容，以及认证轮中的其他关键问题。让我们关注一下考试的表现，关注一下考试中的问题类型。每次考试都必须测试应试者的技术知识，不了解材料的人会得到较低的分数——好的问题会惩罚猜测。然而，糟糕的考试问题，那些只是模糊了正确的答案或花了太长的时间来回答，也会惩罚好的考生。诀窍在于，在考试时，既要奖励那些理解材料的人，又不要让同样的应试者受到伤害，即使他们理解内容，也要提出简单的困难或回答缓慢的问题。 For this reason, most certification exams use several types of exam questions to balance test-taking skill with technical knowledge. Performance-based testing conceptually requires candidates to show real-world skills for a successful outcome. The trick, once again, is balance. Cisco achieves this balance for expert-level certifications by using two exams—one multiple-choice and multiple-answer (the qualification exam), and the other strictly performance-based (the lab exam). This is a potent one-two punch that results in a highly respected certification that also requires a lot of commitment from candidates. Most certifications, however, use a single exam (or several exams on different topics), with blended question types, to arrive at the same result. What, then, is the right balance for these exams? Focusing on Cisco’s Associate- and Professional-level certification exams, I think they have the balance about right. Cisco mixes several different types of questions, including performance-based questions, on these exams. Coming back to Wendell’s point in his earlier post, it may be true that solely performance-based exams—those made up of only simulation or “simlet” scenario questions—might do the best job of identifying the most qualified candidates. The difficulty with that approach is that the test-taking skill aspect of an exam with solely performance-based questions become perhaps as important as knowing the material. For example, someone who is a very strong critical thinker, with good reading and good organization skills, and whose native language is the same as that of the exam, will do comparatively much better at performance-based questions than someone who is weak in any of those areas. It’s not necessarily bad that this is the case, but you need to zoom out and consider what the exam is testing for—is it the technical skill or is it a reading and speed test? The answer, of course, is “yes,” but there’s a balance point that exam developers must find. I think that for the entry-level certifications, the balance should be more toward the technical skill evaluation and less along the lines of complementary test-taking skills. As you climb the certification ladder, the balance should lead toward a more well-rounded, professional skill set—higher level technical competency along with a solid complement of analytical skills and the organizational ability to handle complex scenarios in a high-pressure environment. In other words, what you’d see at work every day. Therefore, higher-level exams should be driven by more performance-based questions. A certification program profile that meets this description should hit all three goals I mentioned early in this post. Assuming, that is, that you agree with my points on the topic. What do you think is the right balance of performance-based questions for each level of certification testing? Which vendors get it right and which ones miss the mark?