Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations Chris Miller on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

TEST QUESTION REVIEW 1

Status
Not open for further replies.

testingnewbie

IS-IT--Management
Apr 23, 2006
2
US
I love the learn-key CBT and have actually studied for the 290 using this method. I work for a company in which I feel failure is NOT an option. Some of my co-workers resent new people being brought in and so I feel an almost desperate need to pass these tests the first time around. That is the reason I am not going to try and cheat unlike 'most' of my other co-workers have done. However, having studied, I want to make darn sure I pass with flying colors. That is why I want to research this topic.

What set of practice tests is either exactly like the real test or most closely resembles the actual test?

TestKing, Transcender, CBT Nuggets, Learnkey, Fravo etc?
I understand a local community college has the EXACT TEST but have not verified this. Some say there is a pool of as many as 450 questions per test and that they simply ask randomly from that pool. Others say, test 1, 2, 3, 4, etc are chosen each of 50 questions. Different practice test methods suggest different ways of doing this.

Does anyone know how it REALLY works? I have passed my MCP and will be using a Pearson Vue testing center for each of the other 6 test. Any help would be greatly appreciated.

 
If you want to cheat use Test King, if you want questions that cover the relevant test areas and use the same style of questions as the real exam use Self Test or Transcender. Personally I used Self Test (Microsoft's recommendation) but from what I've seen Transcenders look pretty good to.

No clue how which questions you get is determined but it can't be purely random from one big set as the tests cover different areas (i.e. you would never get 50 DNS questions :p ). I'd hope they don't just have a few pre-set versions though.
 
When MS first started coming out with this style of tests they did a write up of how it was going to work. Basically, they have a database of an unpublished number of questions. The night before your test the testing facility will download X number of questions that match the test you're taking. The questions that are downloaded consist of "Easy", "Medium" and "Hard" questions that are broken down by categories that are to be tested. For example, if you take a test that has 5 categories that need to be tested, the engine may download 30 questions per category(10 each easy/medium/hard) and have 150 questions to pull from locally. Once it has the pull the engine follows an algorithm (again, unpublished) that will ask the questions in a specific order. The for instance that was given then was as follows:

You're asked a Medium Question:
If you get it right, you're asked a hard question in same category
If you get it wrong, you're asked an easy question in same category

If you get the hard question right, you may not have to answer any more questions in that category.

If you get the easy question right, you get a hard question...if you get it wrong, you're not asked any more questions in that category and you fail the category.

This method can cause the tests to be as long as 40 questions and as short as just a few. The shortest exam that I took was 14 questions and I made a 1000 on it(MCSE 4.0).
 
Macleod,
I appreciate your response. What you suggest is an "adaptive approach" which according to my understanding, MS doesn't use anymore. Each test is over 40 questions. Transcender actually has 4 versions to choose from about 45 questions each. I was wondering if they were random out of the 450 pool or if they were the actual test itself in four varieties. It seems more likely that they are just made to resemble the actual test and that the questions are random pulls by the testing center's MS test pulling engine. I have been through the learnkey CBTs and am ready to look at some questions. It is my understanding that test king is likely the best option. Feel free to comment
 
They certainly don't use adaptive testing anymore.

I failed the 70-291 exam the first time I took it and had to take it again; the second time I got some of the same questions, but only a few. This leads me to beleive that there is a pool of questions in which the testing engine pulls from.

 
Mircrosoft doesn't have adaptive exams for MCSE. AFAIK, CompTIA doesn't use adaptive anymore either, which is relevant for those looking to use CompTIA exam(s) for MCSE elective(s).

As far as questions being weighed as easy/medium/hard, that may explain some of the scores I've gotten better than all questions weighed equally.

Wishdiak
A+, Network+, Security+, MCSA: Security 2003
 
hehehe...guess I should have added to my first post that I took those tests in 1999 :)

I took test 70-306 on Saturday, and I actually read through the first parts of the test that give information about MS's testing philosophy. They do mention that questions will be weighed differently. They also mention that there may be "beta" questions in the test that won't count towards your final score.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top