Win a copy of The Little Book of Impediments (e-book only) this week in the Agile and Other Processes forum!
  • Post Reply
  • Bookmark Topic Watch Topic
  • New Topic

Bad Experience with OCA 7 exams

 
Willem De Bruyn
Greenhorn
Posts: 15
2
  • Likes 2
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Just some feedback from my side, i just failed the test by 3%... FOR THE 2nd time, FML!

I first wrote the exam in 2013, back then the passing score was 77% if I remember correctly, I went into the exam not really knowing much, and kind of "winged" it and managed to walk out with 73%, still failing none the less... Fast forward to the present and I managed to land a junior developer job thanks to an IT diploma I have, and I've been coding full time in JavaScript for about 18 months(Obviously the exam doesn't cover scripting language, but it helps to work in a coding environment on a daily basis).. I thought I'd give this test a go again to spice up my CV, read around on the forums and heard that the difficulty of the test is still the same, and the pass rate is now lowered... I Thought "great" should be "easier" this time round, I put aside a week and a half of full time studying (about 6-8hours a day), worked through the new OCAOCP 7 textbook by Kathy Sierra and Bert Bates, and also through all the mock exams in Enthuware, my scores were as follow:

Test 1: 48% ( I was rusty here, did this test prior to consulting any study material)
Test 2: 58%
Test 3: 61%
Test 4: 68%
Test 5: 71%
Test 6: 77%

As you can see I really improved in as little as week, and after reading that the actual exam is easier than these tests I thought I was ready.

So the first time around in 2013 I finished with like 30mins left, I knew the time of this exam was also reduced from 2-1/2 hours to only 2 hours, but seeing that I didn't struggle the first time, concern didn't even enter my mind.

Well I would say the actual exam was on par with most of the Enthuware as far as difficulty goes, but compared to the first time around this felt much harder, there were A LOT of questions with bloated code snippets that is just very hard to work through given that you only have about 90 seconds per question, I would say the first time I took this test for every 3 straight forward questions there was 1 question the needed some thinking and working through... This time it felt the other way around... And this totally caught me off guard, about 60mins into the exam, I realized I was only 33 questions in, with a few marked for review already, and that's when I started to panic, and just pushed to get through the questions, I'm fairly certain I would've easily made it if I had more time, but you know what they say about "IF's, but's and maybe's".

Anyway thats just feedback from my side, so that future exam candidates reading this can use this knowledge so they're prepared for the time limits

I'm rescheduling this for 2 week from now, this time I wont be caught off guard again.

Any idea where I can get different mock exams, don't want to work through the same ones again.
 
Paul Anilprem
Enthuware Software Support
Ranch Hand
Posts: 3819
10
  • Likes 2
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Hi Willem,
I am very sorry to hear that you failed the exam. We have taken note of your feedback and are working to improve our question bank based on the points that you've mentioned. If you could tell a bit more about the topics that you felt were not covered well in our question bank, we would really appreciate that.
Also, please do let us know your order id (send me a PM or send us an email) and we will issue you are refund.

thank you,
Paul.
 
Guillermo Ishi
Ranch Hand
Posts: 789
C++ Linux Python
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Ha ha! You experience is a whole lot like mine! I first took it maybe a year and a half ago. I had written some in java but didn't know how comprehensive the test was. Ok, got on the boards, found out about the test, found out about Enthuware, was scoring in the 80s and made good progress in Leitner mode too. Took the test in Jly 2014 and failed by THREE POINTS. But the thing is the previous June the test had undergone some changes in style, few direct questions and more about unraveling unrealistic code, which I had little experience with. Not only that, but the test room was noisy, cold, I had ridden a motorcycle 50 miles to the test site, so was a bit tired from that. I wrote Pearson Vue complaining about the noise at the test site but the response was no retest, since I had been able to finish the test! I guessed at the last 12 questions because I ran out of time.

In the meantime Enthuware updated its question bank to be more current with the June 2014 change. I really, really, really hope they haven't changed the test again.


 
Willem De Bruyn
Greenhorn
Posts: 15
2
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Paul Anilprem wrote:Hi Willem,
I am very sorry to hear that you failed the exam. We have taken note of your feedback and are working to improve our question bank based on the points that you've mentioned. If you could tell a bit more about the topics that you felt were not covered well in our question bank, we would really appreciate that.
Also, please do let us know your order id (send me a PM or send us an email) and we will issue you are refund.

thank you,
Paul.


I used the e-book/kindle version found here http://www.amazon.com/gp/product/B007SA1GNU/ref=as_li_tf_tl?ie=UTF8&camp=1789&creative=9325&creativeASIN=B007SA1GNU&linkCode=as2&tag=code0ac-20, but its okay, I don't want a refund, I feel that the topics are well covered, its just that I ran out of time in the actual exam.. Tell me, are the question banks you have on your website http://enthuware.com/index.php/mock-exams/oracle-certified-associate/java-programmer-certification-i the same as the one on amazon?
 
Roel De Nijs
Sheriff
Posts: 10662
144
AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Willem De Bruyn wrote:I used the e-book/kindle version found here http://www.amazon.com/gp/product/B007SA1GNU/ref=as_li_tf_tl?ie=UTF8&camp=1789&creative=9325&creativeASIN=B007SA1GNU&linkCode=as2&tag=code0ac-20, but its okay, I don't want a refund, I feel that the topics are well covered, its just that I ran out of time in the actual exam.. Tell me, are the question banks you have on your website http://enthuware.com/index.php/mock-exams/oracle-certified-associate/java-programmer-certification-i the same as the one on amazon?

Based on this important note from OCAJP Oracle Certified Associate Java SE 7 Programmer Practice Exams on Amazon, I would say yes.
OCAJP Oracle Certified Associate Java SE 7 Programmer Practice Exams on Amazon.com wrote:IMPORTANT: This is an eBook version of Enthuware's simulator JA+ V7 for OCA-JP 7 Certification. It contains the same questions as the simulator. If you are not comfortable with the formatting of this eBook, you may want to try the software version from Enthuware.com, which is available for the same price and is updated automatically on a regular basis.
 
Willem De Bruyn
Greenhorn
Posts: 15
2
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Roel De Nijs wrote:
Willem De Bruyn wrote:I used the e-book/kindle version found here http://www.amazon.com/gp/product/B007SA1GNU/ref=as_li_tf_tl?ie=UTF8&camp=1789&creative=9325&creativeASIN=B007SA1GNU&linkCode=as2&tag=code0ac-20, but its okay, I don't want a refund, I feel that the topics are well covered, its just that I ran out of time in the actual exam.. Tell me, are the question banks you have on your website http://enthuware.com/index.php/mock-exams/oracle-certified-associate/java-programmer-certification-i the same as the one on amazon?

Based on this important note from OCAJP Oracle Certified Associate Java SE 7 Programmer Practice Exams on Amazon, I would say yes.
OCAJP Oracle Certified Associate Java SE 7 Programmer Practice Exams on Amazon.com wrote:IMPORTANT: This is an eBook version of Enthuware's simulator JA+ V7 for OCA-JP 7 Certification. It contains the same questions as the simulator. If you are not comfortable with the formatting of this eBook, you may want to try the software version from Enthuware.com, which is available for the same price and is updated automatically on a regular basis.


Mmmmm but it also mentions "updated automatically on a regular basis." Waiting for Paul's reply on this, maybe worth getting this is there has been some changes
 
Guillermo Ishi
Ranch Hand
Posts: 789
C++ Linux Python
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Roel De Nijs wrote:...you may want to try the software version from Enthuware.com, which is available for the same price and is updated automatically on a regular basis.

Does Kindle notify you when new versions are available? There are automatic updates on the PC version, pretty frequently.

P.S. Beat me to the question.
 
Paul Anilprem
Enthuware Software Support
Ranch Hand
Posts: 3819
10
  • Likes 1
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Willem De Bruyn wrote:
Roel De Nijs wrote:
Willem De Bruyn wrote:I used the e-book/kindle version found here http://www.amazon.com/gp/product/B007SA1GNU/ref=as_li_tf_tl?ie=UTF8&camp=1789&creative=9325&creativeASIN=B007SA1GNU&linkCode=as2&tag=code0ac-20, but its okay, I don't want a refund, I feel that the topics are well covered, its just that I ran out of time in the actual exam.. Tell me, are the question banks you have on your website http://enthuware.com/index.php/mock-exams/oracle-certified-associate/java-programmer-certification-i the same as the one on amazon?

Based on this important note from OCAJP Oracle Certified Associate Java SE 7 Programmer Practice Exams on Amazon, I would say yes.
OCAJP Oracle Certified Associate Java SE 7 Programmer Practice Exams on Amazon.com wrote:IMPORTANT: This is an eBook version of Enthuware's simulator JA+ V7 for OCA-JP 7 Certification. It contains the same questions as the simulator. If you are not comfortable with the formatting of this eBook, you may want to try the software version from Enthuware.com, which is available for the same price and is updated automatically on a regular basis.


Mmmmm but it also mentions "updated automatically on a regular basis." Waiting for Paul's reply on this, maybe worth getting this is there has been some changes

Willem,
In case of the simulator, we control the update process and so a user gets the updated version as soon as we update it. We do update the eBook on Amazon also at the same time. However, we do not know when the customer of the eBook gets the update because it is done by Amazon. Therefore, an ebook user may see errors that have already been fixed. For this reason, we have included the statement that you've quoted.
If you have purchased the ebook, you will not get anything new, in terms of content, by purchasing the simulator.
HTH,
Paul.
 
Willem De Bruyn
Greenhorn
Posts: 15
2
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Paul Anilprem wrote:
Willem De Bruyn wrote:
Roel De Nijs wrote:
Willem De Bruyn wrote:I used the e-book/kindle version found here http://www.amazon.com/gp/product/B007SA1GNU/ref=as_li_tf_tl?ie=UTF8&camp=1789&creative=9325&creativeASIN=B007SA1GNU&linkCode=as2&tag=code0ac-20, but its okay, I don't want a refund, I feel that the topics are well covered, its just that I ran out of time in the actual exam.. Tell me, are the question banks you have on your website http://enthuware.com/index.php/mock-exams/oracle-certified-associate/java-programmer-certification-i the same as the one on amazon?

Based on this important note from OCAJP Oracle Certified Associate Java SE 7 Programmer Practice Exams on Amazon, I would say yes.
OCAJP Oracle Certified Associate Java SE 7 Programmer Practice Exams on Amazon.com wrote:IMPORTANT: This is an eBook version of Enthuware's simulator JA+ V7 for OCA-JP 7 Certification. It contains the same questions as the simulator. If you are not comfortable with the formatting of this eBook, you may want to try the software version from Enthuware.com, which is available for the same price and is updated automatically on a regular basis.


Mmmmm but it also mentions "updated automatically on a regular basis." Waiting for Paul's reply on this, maybe worth getting this is there has been some changes

Willem,
In case of the simulator, we control the update process and so a user gets the updated version as soon as we update it. We do update the eBook on Amazon also at the same time. However, we do not know when the customer of the eBook gets the update because it is done by Amazon. Therefore, an ebook user may see errors that have already been fixed. For this reason, we have included the statement that you've quoted.
If you have purchased the ebook, you will not get anything new, in terms of content, by purchasing the simulator.
HTH,
Paul.


I bought the ebook in 2013.... I see on the website the simulator has 7 tests, the ebook only has 6
 
Paul Anilprem
Enthuware Software Support
Ranch Hand
Posts: 3819
10
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Willem De Bruyn wrote:
Paul Anilprem wrote:
Willem De Bruyn wrote:
Roel De Nijs wrote:
Willem De Bruyn wrote:I used the e-book/kindle version found here http://www.amazon.com/gp/product/B007SA1GNU/ref=as_li_tf_tl?ie=UTF8&camp=1789&creative=9325&creativeASIN=B007SA1GNU&linkCode=as2&tag=code0ac-20, but its okay, I don't want a refund, I feel that the topics are well covered, its just that I ran out of time in the actual exam.. Tell me, are the question banks you have on your website http://enthuware.com/index.php/mock-exams/oracle-certified-associate/java-programmer-certification-i the same as the one on amazon?

Based on this important note from OCAJP Oracle Certified Associate Java SE 7 Programmer Practice Exams on Amazon, I would say yes.
OCAJP Oracle Certified Associate Java SE 7 Programmer Practice Exams on Amazon.com wrote:IMPORTANT: This is an eBook version of Enthuware's simulator JA+ V7 for OCA-JP 7 Certification. It contains the same questions as the simulator. If you are not comfortable with the formatting of this eBook, you may want to try the software version from Enthuware.com, which is available for the same price and is updated automatically on a regular basis.


Mmmmm but it also mentions "updated automatically on a regular basis." Waiting for Paul's reply on this, maybe worth getting this is there has been some changes

Willem,
In case of the simulator, we control the update process and so a user gets the updated version as soon as we update it. We do update the eBook on Amazon also at the same time. However, we do not know when the customer of the eBook gets the update because it is done by Amazon. Therefore, an ebook user may see errors that have already been fixed. For this reason, we have included the statement that you've quoted.
If you have purchased the ebook, you will not get anything new, in terms of content, by purchasing the simulator.
HTH,
Paul.


I bought the ebook in 2013.... I see on the website the simulator has 7 tests, the ebook only has 6

We did quite a bit of reorganization as well as add new questions a few months ago when Oracle made changes to the exam (they changed the number of questions, duration, and pass percent). The current version of ebook on Amazon reflects the same.
But it looks like Amazon has stopped updating yours. We are not really sure about their policy on for how long and how frequently they update the readers. In this case, you will see about 20% change between the old and the new version.
 
Willem De Bruyn
Greenhorn
Posts: 15
2
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Paul Anilprem wrote:
Willem De Bruyn wrote:
Paul Anilprem wrote:
Willem De Bruyn wrote:
Roel De Nijs wrote:
Willem De Bruyn wrote:I used the e-book/kindle version found here http://www.amazon.com/gp/product/B007SA1GNU/ref=as_li_tf_tl?ie=UTF8&camp=1789&creative=9325&creativeASIN=B007SA1GNU&linkCode=as2&tag=code0ac-20, but its okay, I don't want a refund, I feel that the topics are well covered, its just that I ran out of time in the actual exam.. Tell me, are the question banks you have on your website http://enthuware.com/index.php/mock-exams/oracle-certified-associate/java-programmer-certification-i the same as the one on amazon?

Based on this important note from OCAJP Oracle Certified Associate Java SE 7 Programmer Practice Exams on Amazon, I would say yes.
OCAJP Oracle Certified Associate Java SE 7 Programmer Practice Exams on Amazon.com wrote:IMPORTANT: This is an eBook version of Enthuware's simulator JA+ V7 for OCA-JP 7 Certification. It contains the same questions as the simulator. If you are not comfortable with the formatting of this eBook, you may want to try the software version from Enthuware.com, which is available for the same price and is updated automatically on a regular basis.


Mmmmm but it also mentions "updated automatically on a regular basis." Waiting for Paul's reply on this, maybe worth getting this is there has been some changes

Willem,
In case of the simulator, we control the update process and so a user gets the updated version as soon as we update it. We do update the eBook on Amazon also at the same time. However, we do not know when the customer of the eBook gets the update because it is done by Amazon. Therefore, an ebook user may see errors that have already been fixed. For this reason, we have included the statement that you've quoted.
If you have purchased the ebook, you will not get anything new, in terms of content, by purchasing the simulator.
HTH,
Paul.


I bought the ebook in 2013.... I see on the website the simulator has 7 tests, the ebook only has 6

We did quite a bit of reorganization as well as add new questions a few months ago when Oracle made changes to the exam (they changed the number of questions, duration, and pass percent). The current version of ebook on Amazon reflects the same.
But it looks like Amazon has stopped updating yours. We are not really sure about their policy on for how long and how frequently they update the readers. In this case, you will see about 20% change between the old and the new version.


In that case I'd say its worth buying, tested the trial version of the simulator now, a lot more organized, and keeps better track of what you need to improve on etc... Thanks for your input
 
Roel De Nijs
Sheriff
Posts: 10662
144
AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
  • Likes 1
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Willem De Bruyn wrote:Just some feedback from my side, i just failed the test by 3%... FOR THE 2nd time, FML!

Sorry to hear you failed the exam!

Willem De Bruyn wrote:Fast forward to the present and I managed to land a junior developer job thanks to an IT diploma I have, and I've been coding full time in JavaScript for about 18 months(Obviously the exam doesn't cover scripting language, but it helps to work in a coding environment on a daily basis).

Great to hear you got a job as a junior JavaScript developer. But although the 1st 4 letters are exactly the same, Java and JavaScript don't have anything in common So it's definitely a good thing you can code on a daily basis, but this coding doesn't really count for Java practice. And learning Java is just like driving a car, you don't learn it by just reading a text book. You need to get your hands dirty and code as if your life depended on it In that respect, a junior java developer has a bigger advantage.

Willem De Bruyn wrote:I Thought "great" should be "easier" this time round, I put aside a week and a half of full time studying (about 6-8hours a day), worked through the new OCAOCP 7 textbook by Kathy Sierra and Bert Bates, and also through all the mock exams in Enthuware

For other aspirants it might be helpful if you could elaborate on how you have studied/prepared yourself during these 10 days. Did you just read/study the K&B7 study guide? Did you create lots of code snippets? When coding did you use an IDE or a text editor? ...

Willem De Bruyn wrote:Well I would say the actual exam was on par with most of the Enthuware as far as difficulty goes, but compared to the first time around this felt much harder, there were A LOT of questions with bloated code snippets that is just very hard to work through given that you only have about 90 seconds per question, I would say the first time I took this test for every 3 straight forward questions there was 1 question the needed some thinking and working through... This time it felt the other way around...

That's not what I experienced at all when I took the (new format of the) OCAJP7 exam. Maybe I was lucky? Maybe I have a different standard of bloated code snippets? I believe at most 5 questions were not able to fit on the screen and you had to scroll down. I remember one of them very good, because the code of that question looked very complex and I was hesitant to start working through this code. But then I glanced at the possible answers and noticed they were just looking for the valid syntax to address a 2D array element, so there was completely no need to evaluate the whole code snippet.
And time management is indeed a very important part of the exam as well. And when you are under (time) pressure you don't want to go through some complex code snippet only then to discover there is a compiler error in the code snippet. That's a great loss of precious time. So before you start a questions, quickly glance at the possible answers. If "Compiler error" is listed, first verify the code compiles without errors before proceeding. Otherwise you can directly dive into the code. This is just one tip which you could benefit from. You'll find other ones (for your 3rd attempt) here, here and here.

Willem De Bruyn wrote:Anyway thats just feedback from my side, so that future exam candidates reading this can use this knowledge so they're prepared for the time limits

Thanks for sharing your experiences! Will be very helpful to other OCA candidates. Have a cow!

Willem De Bruyn wrote:I'm rescheduling this for 2 week from now, this time I wont be caught off guard again.

Good luck! And if you have some questions/doubts, just start a new thread. We will do our utmost to make you pass the exam! Third time's a charm

Hope it helps!
Kind regards,
Roel
 
Willem De Bruyn
Greenhorn
Posts: 15
2
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Roel De Nijs wrote:
For other aspirants it might be helpful if you could elaborate on how you have studied/prepared yourself during these 10 days. Did you just read/study the K&B7 study guide? Did you create lots of code snippets? When coding did you use an IDE or a text editor? ...



Read through K&B7 book, worked through 6 Enthuware mock exams (but I only found out in this thread the ebook i bought back in 2013 might actually contain mistakes)... I did write lots of code, looking back now there's at least 50-60 classes in my "Java 7 Prep" project in Netbeans... There's none of the topics I don't understand to be honest, quite frustrating, not to sure how to go about my preparation for the test again next month

The only one thing I didn't do in my preparation was to time myself, I took my sweet time with the mock exams, and this got me into the habit of taking things slowly, which is a very bad thing I suppose
 
Guillermo Ishi
Ranch Hand
Posts: 789
C++ Linux Python
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Willem De Bruyn wrote:There's none of the topics I don't understand to be honest, quite frustrating, not to sure how to go about my preparation for the test again next month

The latest rev PC version of Enthuware you have is probably different enough from what you have to be fresh. It is also more "figuring out bloated code" oriented. I too understand Java at the ocajp level but haven't passed the test. I score higher on Enthuware tests when I just pay extreme attention to every question. Some of the answers are simple but you have to read carefully.

Roel is convinced if you get in the 80s on Enthuware you will pass the test. So that is how to go about preparing for it. I doubt Enthuware is harder than the real exam; probably in the same ballpark. The stats on the Enthuware site, I think the Enthu test scores are entered automatically. The real test scores are entered manually. My gut feeling is the higher your score the more likely you are to submit it, so...

The right way to do it would be to when a person enters their real score, at that point their Enthu scores are included too. If you can't get both for some reason then you act to the user like the real score was submitted but really don't. That might be what they do but I doubt it.

 
Roel De Nijs
Sheriff
Posts: 10662
144
AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Willem De Bruyn wrote:I did write lots of code, looking back now there's at least 50-60 classes in my "Java 7 Prep" project in Netbeans...

Excellent! Creating lots of code snippets is really good but you used an IDE. I strongly advice against using an IDE while preparing for a certification exam, certainly if you are a junior Java programmer. The reason is simple: the IDE takes care of certain tasks for you (detecting compiler errors, code completion,...), but on the actual exam you don't have an IDE and you are completely on your own. I'm convinced you'll be able to spot errors more quickly in a given code snippet when you don't use an IDE. But this is just my opinion, here you'll find different opinions about using an IDE while preparing for a certification exam. Here and here you'll find experiences from other ranchers who stopped using an IDE while preparing. So that would be my strongest advice for your exam preparation!

Willem De Bruyn wrote:The only one thing I didn't do in my preparation was to time myself, I took my sweet time with the mock exams, and this got me into the habit of taking things slowly, which is a very bad thing I suppose

I would assume from mock exam software you can take the mock exams in "normal" mode and "exam" mode. In the first mode there's no time limit, you can get a hint (if any) and so on; but in exam mode it's like the actual exam with the limited time, no hints/explanations, just you against the questions I would strongly suggest to take the Enthuware mock exams in "exam" mode. This will give you a very reliable indication if you are ready for the exam or not (and will prevent you from spending money again while you are not ready for the task). Other mock exams you can take in "normal" mode (just to exercise).

Hope it helps!
Kind regards,
Roel
 
Paul Anilprem
Enthuware Software Support
Ranch Hand
Posts: 3819
10
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Guillermo Ishi wrote:
The right way to do it would be to when a person enters their real score, at that point their Enthu scores are included too.

This is a very good idea. We will implement it asap.


If you can't get both for some reason then you act to the user like the real score was submitted but really don't. That might be what they do but I doubt it.

I am not sure if I understand what you mean. Some users submit their real exam scores, some don't. We can't do anything about that. We certainly don't cook the numbers. We just show the data that we receive.
 
Roel De Nijs
Sheriff
Posts: 10662
144
AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Paul Anilprem wrote:
If you can't get both for some reason then you act to the user like the real score was submitted but really don't. That might be what they do but I doubt it.

I am not sure if I understand what you mean. Some users submit their real exam scores, some don't. We can't do anything about that. We certainly don't cook the numbers. We just show the data that we receive.

I think I can look into Guillermo's head If user submits his/her real exam score, the software collects the mock exam scores and submits both to Enthuware data centre. If (for some reason) collecting the mock exam scores fails, the software doesn't submit the real exam score. In both scenarios the user receives a confirmation about successfully submitting his/her real exam score.
 
Guillermo Ishi
Ranch Hand
Posts: 789
C++ Linux Python
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Exactly. If you can't get both for some reason you probably (certainly) don't want to trouble the user with it. So you you act like everything was fine, don't use any of his data, and just let it slide.

I didn't mean to say anything that could be interpreted as numbers being fudged. Just making a suggestion that will give definite correlation between mock scores and real. If I got 69% on the real test I might submit it. If I got 100% I would definitely submit it and send all my third cousins a letter about it too. I wouldn't but some would. I'm taking it out of those irresponsible hands
 
Paul Anilprem
Enthuware Software Support
Ranch Hand
Posts: 3819
10
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
OK, no, if the submission fails, the user is notified that it has failed. They can try again.
We put together this feature originally without much thought. Now, that it is being used so much, we will definitely work on enhancing it.

thank you for your suggestion. We sincerely appreciate it.

-Paul.
 
Guillermo Ishi
Ranch Hand
Posts: 789
C++ Linux Python
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
You're sincerely welcome. I was thinking like they might enter their real score from a phone or different computer, but you can control all that.
It will be interesting to see and it will help you evaluate your product more closely. It will also tell you immediately if they change the test
 
Roel De Nijs
Sheriff
Posts: 10662
144
AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Guillermo Ishi wrote:Roel is convinced if you get in the 80s on Enthuware you will pass the test. So that is how to go about preparing for it.

Yes, I'll still believe if you score consistently 80% or more on Enthuware mock exams, you'll pass the actual exam. There is just 1 requirement: you have to take the mock exam as it was an actual exam (time limit, no google, no CodeRanch, no IDE,... in short no external resources besides a few sheets of paper). And another thingy (depending on how good/bad your memory is): only your 1st attempt of a given mock exam counts (scoring 94% on your 4th attempt of that mock exam doesn't count).

And I'm hanging around this forum for quite some time now and I have not seen an exception to this "rule". So I'm pretty confident about this one And if you really want some actual facts: look at the experiences in the OcajpWallOfFame, some ranchers shared their Enthuware mock exam scores together with their actual score.
 
Roel De Nijs
Sheriff
Posts: 10662
144
AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Paul Anilprem wrote:OK, no, if the submission fails, the user is notified that it has failed. They can try again.

The one and only correct procedure!

Paul Anilprem wrote:thank you for your suggestion. We sincerely appreciate it.

This sounds like you can expect a free license key to use the Enthuware mock exams, Guillermo
 
Guillermo Ishi
Ranch Hand
Posts: 789
C++ Linux Python
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Roel De Nijs wrote:
Paul Anilprem wrote:OK, no, if the submission fails, the user is notified that it has failed. They can try again.

The one and only correct procedure!

Paul Anilprem wrote:thank you for your suggestion. We sincerely appreciate it.

This sounds like you can expect a free license key to use the Enthuware mock exams, Guillermo

There's no reason the submission should fail except for some reason the user can't do anything about. Unless he tries to spell the number or enters it in hex Or it's over 100. Or less than zero. Otherwise leave him in peace. I would add a confirmation in case he enters 9 instead of 99.

Don't need nuthin free. I work for a livin. Could use a new tire, though.

I would find it strange to find 40% on the wall of fame with some high Enthuware scores, too. But I would find it even stranger to find a 40% in the wall of fame. Your prediction algorithm is using overfitted data. My gut says you're right; my brain says you could be wrong. But I did recommend your theory.







 
Roel De Nijs
Sheriff
Posts: 10662
144
AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Guillermo Ishi wrote:I would find it strange to find 40% on the wall of fame with some high Enthuware scores, too. But I would find it even stranger to find a 40% in the wall of fame. Your prediction algorithm is using overfitted data. My gut says you're right; my brain says you could be wrong. But I did recommend your theory.

You won't find a 40% on the OcajpWallOfFame, because with that score you didn't pass the test which is the one and only requirement to add your name to the wall
 
Paul Anilprem
Enthuware Software Support
Ranch Hand
Posts: 3819
10
  • Likes 1
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Based on the above discussion, the Average Scores page now works as follows -

1. The viewer already let the user submit their average mock exam score as well as the real exam score. But the score page displayed only the average real exam score. Average of average mock exam scores was not displayed. The page now displays both of these numbers i.e. average of average mock exam score and average of real exam score as reported by the users. These two numbers show the correlation between the mock exams and the real exam.

2. Earlier, the user could change the average mock exam score value while submitting the scores. Now, the viewer automatically picks only the first attempt scores of Standard Tests, computes their average, and displays that average on the submit dialog. But it doesn't allow the user to change this value. The user needs to enter only the real exam score. Further, only those users who have taken at least one Standard Test and have scored more than 35% in that test can submit the scores. This restriction is put in place to eliminate the cases where a user attempts the mock tests casually without any preparation.

3. Averages of each mock exam score is still displayed as before. Although these averages may not correlate too well to the real exam score, they still give you an idea about the relative toughness of each mock exam. They also tell you how well other people have scored on a particular mock exam.

Here is the updated page: http://enthuware.com/index.php/ocajp-7-average-scores

Comments welcome
 
Guillermo Ishi
Ranch Hand
Posts: 789
C++ Linux Python
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Roel De Nijs wrote:You won't find a 40% on the OcajpWallOfFame, because with that score you didn't pass the test which is the one and only requirement to add your name to the wall

I know. The point of my rambling was to say that from the Wall we don't know anything about Enthu scores from the set of people who didn't pass the real, so we can't scientifically say if you make X on Enthu you make X on real. We assume Enthu scores of the people who didn't pass the real were significantly lower than the Enthu scores of people who passed the real, but we can't prove it.

I think including mock scores only from people who entered real scores tightens it up a lot, but the biggest gain would come from somehow insuring that the entered real scores reflect what happens out in the field. The entered real scores may be random enough, but I suspect there would be a bias toward high scores being entered. Even if you paid everybody to enter their real score, there's no assurance it is a genuine real score! I don't know how to fix this for sure, even if you included Certview somehow. If whatever you require users to do isn't simple, nobody will bother to do it.
 
Roel De Nijs
Sheriff
Posts: 10662
144
AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Paul Anilprem wrote:Comments welcome

Fasten your seatbelt, here I go!

Paul Anilprem wrote:Now, the viewer automatically picks only the first attempt scores of Standard Tests, computes their average, and displays that average on the submit dialog. But it doesn't allow the user to change this value.

Of all possible implementation, that's the best one!

Paul Anilprem wrote:Further, only those users who have taken at least one Standard Test and have scored more than 35% in that test can submit the scores.

Assume I take in total 4 standard tests (mock exams):
  • Test 1: 25% (I was unprepared)
  • Test 2: 55% (I studied really hard)
  • Test 3: 30% (I clearly don't understand a few key objectives)
  • Test 4: 65% (I focused on those few key objectives)
  • What's my average mock score? Based on your explanation I guess 60% (only test 2 & 4 are taken into account). Or is it 50% (test 2, 3 & 4)? Because once you passed the 35% treshold you are considered to be prepared and all tests from that point count.

    Finally I have 2 other suggestions. Just remember I have never used Enthuware mock exams (I know, shame on me ), so if I describe a feature which is already in the software just say so.

    1/ I know from your website a user gets a refund if he/she fails. But what's the procedure? Do I have to submit my actual exam score as well? Or send Enthuware a mail with a copy of my certification status? Maybe it could be interesting if it was linked with submitting the scores. Then you have statistics for exam failures. Or maybe these scores are already incorporated in the current average scores. If it is, it might be useful to have two seperate average scores tables.

    2/ Assume I want some practice, but I don't have time to do a complete mock exam, I only have 75 minutes available. Can I start a customized test in "real exam mode"? So I enter the number of minutes and the software calculates how many questions I must solve in this duration (based on the actual exam details (70 questions in 7200 seconds). These questions could be randomly chosen based on exam objective or from mock exams I already completed or ... I would not allow to pick questions from mock exams which are not completed yet (it might mess up the average score )

    Kind regards,
    Roel
     
    Roel De Nijs
    Sheriff
    Posts: 10662
    144
    AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
    • Mark post as helpful
    • send pies
    • Quote
    • Report post to moderator
    Guillermo Ishi wrote:The point of my rambling was to say that from the Wall we don't know anything about Enthu scores from the set of people who didn't pass the real, so we can't scientifically say if you make X on Enthu you make X on real. We assume Enthu scores of the people who didn't pass the real were significantly lower than the Enthu scores of people who passed the real, but we can't prove it.

    That's simply impossible! Even if everybody would share their exact mock exam and actual exam scores, you can't predict the outcome on the exam based on an average score. Although both of us have an average of 85% on the Enthuware mock exams, it's very plausible you score 90% and I only 80% on the actual exam. And I can think of a bunch of reasons why: you better handle a deadline situation than me, my memory is older and thus more volatile , the shape of the day, distracting/confounding/annoying factors at the exam center, it's very likely we both get a different set of questions from the question pool,...

    Guillermo Ishi wrote:Even if you paid everybody to enter their real score, there's no assurance it is a genuine real score! I don't know how to fix this for sure, even if you included Certview somehow. If whatever you require users to do isn't simple, nobody will bother to do it.

    Who cares if the submitted score is genuine or not? Or if people who score 85+% are more likely to submit their real score? This average real/mock exam scores are great to have an idea about your level (and possible score), but nothing more than that. It's (like you already mentioned) by no means a scientific approach. And you should only take the actual exam if you feel confident about your Java knowledge and level. If you score 85% on the last day test, but you have doubts about your readiness, just study another week (or maybe two) and if it's needed, simply reschedule/postpone your exam (it's free)
     
    Guillermo Ishi
    Ranch Hand
    Posts: 789
    C++ Linux Python
    • Mark post as helpful
    • send pies
    • Quote
    • Report post to moderator
    If you were to graph Enthu scores against real ones we expect the results would show clustering around a straight line. That is what Enthuware is intended to do. At that point you can begin to make predictions. It is obvious why you care if submitted scores are genuine or not. I'm not claiming this is practical. You would ultimately like to control a lot of things I don't see any practical way to control. It was just an idea.
     
    Roel De Nijs
    Sheriff
    Posts: 10662
    144
    AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
    • Mark post as helpful
    • send pies
    • Quote
    • Report post to moderator
    Guillermo Ishi wrote:It is obvious why you care if submitted scores are genuine or not.

    Enlighten me please.
     
    Paul Anilprem
    Enthuware Software Support
    Ranch Hand
    Posts: 3819
    10
    • Mark post as helpful
    • send pies
    • Quote
    • Report post to moderator
    Roel De Nijs wrote:

    Paul Anilprem wrote:Further, only those users who have taken at least one Standard Test and have scored more than 35% in that test can submit the scores.

    Assume I take in total 4 standard tests (mock exams):
  • Test 1: 25% (I was unprepared)
  • Test 2: 55% (I studied really hard)
  • Test 3: 30% (I clearly don't understand a few key objectives)
  • Test 4: 65% (I focused on those few key objectives)
  • What's my average mock score? Based on your explanation I guess 60% (only test 2 & 4 are taken into account). Or is it 50% (test 2, 3 & 4)? Because once you passed the 35% treshold you are considered to be prepared and all tests from that point count.

    Yes, 60%. You would have to not understand a lot more key objectives to score 30% Such low score implies that either you are zipping through a test just to see the type of questions and answering without giving much thought, or you are grossly underprepared. In both the cases, your score is of not much use to other candidates.
    Also, I think if you go from scoring 55% to 30%, there must have been some other unrelated reason such as disturbance which caused you to either just end the test or zip through it. Not a good data point to keep.


    1/ I know from your website a user gets a refund if he/she fails. But what's the procedure? Do I have to submit my actual exam score as well? Or send Enthuware a mail with a copy of my certification status? Maybe it could be interesting if it was linked with submitting the scores. Then you have statistics for exam failures. Or maybe these scores are already incorporated in the current average scores. If it is, it might be useful to have two seperate average scores tables.

    Procedure is quite straight forward. We do verify (through multiple means, depends on various factors) that a user failed the exam after purchasing our product. Just like any other user, one can submit their scores through the viewer itself. We don't do that from our end. But such cases are really very very few. Among them, most of the time we see requests from candidates who have attempted one or two mock tests just before the real exam and failed in them with very low scores. A few candidates keep taking the same test multiple times until they reach 90% and then take the real exam only to be disappointed. I don't think adding their scores would add much value.


    2/ Assume I want some practice, but I don't have time to do a complete mock exam, I only have 75 minutes available. Can I start a customized test in "real exam mode"? So I enter the number of minutes and the software calculates how many questions I must solve in this duration (based on the actual exam details (70 questions in 7200 seconds). These questions could be randomly chosen based on exam objective or from mock exams I already completed or ... I would not allow to pick questions from mock exams which are not completed yet (it might mess up the average score )
    Kind regards,
    Roel

    1. You can create a custom test containing any number of randomly selected questions on any exam objective (s). The time is automatically adjusted as per the time limitation per question of the real exam. It is technically possible to avoid questions from any standard test(s) but avoiding them is a little more work. You can see the list of question ids for any standard test, you can copy it and paste it while creating a customized test (it asks you if you want to exclude certain questions while creating a customized test).
    2. You can enter the amount of time you have and it will keep throwing you as many questions as you can answer in that time limit. If you do not want to see questions from other unused tests, you can focus only on questions that you've missed earlier or attempted earlier. This will not mess up the average score.
    3. It is currently not possible to just specify the time and expect a fixed number of questions depending on the time limit per question.

    In general though, I would say there is not much need to customize all these things. If you just take standard tests one by one and keep reading up on topics one which you score less before attempting the next test, within a couple of tests, your score will improve substantially. You can take more time than prescribed in the initial tests and later you can focus on answering them within the time limit. After 6 tests, I am sure you will be ready.

    HTH,
    Paul.
     
    Guillermo Ishi
    Ranch Hand
    Posts: 789
    C++ Linux Python
    • Mark post as helpful
    • send pies
    • Quote
    • Report post to moderator
    Roel De Nijs wrote:
    Guillermo Ishi wrote:It is obvious why you care if submitted scores are genuine or not.

    Enlighten me please.

    If it doesn't matter if they're phony, then don't bother collecting them - just make them up?

     
    Roel De Nijs
    Sheriff
    Posts: 10662
    144
    AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
    • Mark post as helpful
    • send pies
    • Quote
    • Report post to moderator
    Paul Anilprem wrote:Also, I think if you go from scoring 55% to 30%, there must have been some other unrelated reason such as disturbance which caused you to either just end the test or zip through it. Not a good data point to keep.

    True! I just was wondering if every test beneath the treshold was ignored or once you managed to score higher all mock exam scores onwards are included in the average. And you gave me a satisfactory answer, so I'm happy

    Paul Anilprem wrote:1. You can create a custom test containing any number of randomly selected questions on any exam objective (s). The time is automatically adjusted as per the time limitation per question of the real exam. It is technically possible to avoid questions from any standard test(s) but avoiding them is a little more work. You can see the list of question ids for any standard test, you can copy it and paste it while creating a customized test (it asks you if you want to exclude certain questions while creating a customized test).
    2. You can enter the amount of time you have and it will keep throwing you as many questions as you can answer in that time limit. If you do not want to see questions from other unused tests, you can focus only on questions that you've missed earlier or attempted earlier. This will not mess up the average score.
    3. It is currently not possible to just specify the time and expect a fixed number of questions depending on the time limit per question.

    With 1 and 2 available, I don't need 3, I can simulate 3 using 1 and a little bit calculation on my behalf Sounds like a great product! And this new feature (about submitting 2st attempt mock exam scores with the actual exam score) was implemented/added very quickly, so that betrays a good design and excellent code quality as well

    Paul Anilprem wrote:After 6 tests, I am sure you will be ready.

    No doubt! You'll definitely be ready after a gazillion mock/practice questions. But I've seen a few people already mentioning that they encountered on the actual exam some questions which were almost identical to Enthuware practice questions. This worries me a bit, because that could probably be a violation of Oracle's exam policy.
     
    Roel De Nijs
    Sheriff
    Posts: 10662
    144
    AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
    • Mark post as helpful
    • send pies
    • Quote
    • Report post to moderator
    Guillermo Ishi wrote:
    Roel De Nijs wrote:
    Guillermo Ishi wrote:It is obvious why you care if submitted scores are genuine or not.

    Enlighten me please.

    If it doesn't matter if they're phony, then don't bother collecting them - just make them up?

    But what benefit as a user would you have to submit a phony (higher) actual exam score as it's submitted anonymously. I don't get why you would submit 85% instead of 70%. It's anonymous, nobody will ever know your score. As opposed to the OcajpWallOfFame where your score is visible for the whole wide world. So if you are not completely satisfied with your score, you can always embellish your score a little with 5, 10 or 15%.
     
    Paul Anilprem
    Enthuware Software Support
    Ranch Hand
    Posts: 3819
    10
    • Mark post as helpful
    • send pies
    • Quote
    • Report post to moderator
    Roel De Nijs wrote:
    Paul Anilprem wrote:After 6 tests, I am sure you will be ready.

    No doubt! You'll definitely be ready after a gazillion mock/practice questions. But I've seen a few people already mentioning that they encountered on the actual exam some questions which were almost identical to Enthuware practice questions. This worries me a bit, because that could probably be a violation of Oracle's exam policy.

    We do not purposefully put them there. Out of hundreds of questions, it is possible that a few may look very close the real exam questions but only because there are only a few ways you can frame a question on a concept. For example, if you want to test a candidate on what happens to an arraylist if you try to add an element twice. If two entirely different people formulate 4-5 unique questions on this point, a couple of them will be bound to be almost same in structure, except may be variable names (and who doesn't name a variable of type ArrayList as arrayList or list )

    Even on our website we have clearly mentioned that we do not sell brain dumps ( http://enthuware.com/index.php/ocp-brain-dumps )

    We have been in the market for more than 10 years now and we haven't heard from Sun or Oracle's lawyers yet. I am sure if we violated their policy, they would come after us in no time.

     
    Guillermo Ishi
    Ranch Hand
    Posts: 789
    C++ Linux Python
    • Mark post as helpful
    • send pies
    • Quote
    • Report post to moderator
    Roel De Nijs wrote:But what benefit as a user would you have to submit a phony (higher) actual exam score as it's submitted anonymously.

    If you were working on a machine that would cause you bodily harm if its settings had been changed, you would check the settings before working on it even though no one would benefit from changing the settings. I can imagine an experiment that shows people are as likely to enter a wrong score in private as in public. If one in four people is a narcissist, maybe one in four of your entered real scores are wrong. We just don't know... But really, to guarantee some conclusion you obviously have to guarantee your data.
     
    Roel De Nijs
    Sheriff
    Posts: 10662
    144
    AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
    • Mark post as helpful
    • send pies
    • Quote
    • Report post to moderator
    Guillermo Ishi wrote:But really, to guarantee some conclusion you obviously have to guarantee your data.

    Luckily we don't draw any conclusions from this data
     
    Roel De Nijs
    Sheriff
    Posts: 10662
    144
    AngularJS Chrome Eclipse IDE Hibernate Java jQuery MySQL Database Spring Tomcat Server
    • Mark post as helpful
    • send pies
    • Quote
    • Report post to moderator
    Paul Anilprem wrote:We do not purposefully put them there.

    I certainly didn't want to accuse Enthuware from doing this practice on purpose.

    Paul Anilprem wrote:Out of hundreds of questions, it is possible that a few may look very close the real exam questions but only because there are only a few ways you can frame a question on a concept. For example, if you want to test a candidate on what happens to an arraylist if you try to add an element twice. If two entirely different people formulate 4-5 unique questions on this point, a couple of them will be bound to be almost same in structure, except may be variable names (and who doesn't name a variable of type ArrayList as arrayList or list )

    True! For some concepts/topics there's very little variation possible. It would only become suspicious if the mock questions add exactly the same elements to the ArrayList as on the actual exam.

    Paul Anilprem wrote:We have been in the market for more than 10 years now and we haven't heard from Sun or Oracle's lawyers yet. I am sure if we violated their policy, they would come after us in no time.

    As Oracle is very lawyer-savvy, that's probably the best designation everything's just fine
     
    Jeanne Boyarsky
    author & internet detective
    Marshal
    Posts: 35279
    384
    Eclipse IDE Java VI Editor
    • Mark post as helpful
    • send pies
    • Quote
    • Report post to moderator
    Paul Anilprem wrote:Out of hundreds of questions, it is possible that a few may look very close the real exam questions but only because there are only a few ways you can frame a question on a concept. For example, if you want to test a candidate on what happens to an arraylist if you try to add an element twice. If two entirely different people formulate 4-5 unique questions on this point, a couple of them will be bound to be almost same in structure, except may be variable names (and who doesn't name a variable of type ArrayList as arrayList or list )

    Yeah. I made a point of writing ALL my OCA mock questions before taking the beta. There were a couple that turned out similar for the above reasons.
     
    Guillermo Ishi
    Ranch Hand
    Posts: 789
    C++ Linux Python
    • Mark post as helpful
    • send pies
    • Quote
    • Report post to moderator
    Roel De Nijs wrote:
    Guillermo Ishi wrote:But really, to guarantee some conclusion you obviously have to guarantee your data.

    Luckily we don't draw any conclusions from this data

    Since when?

     
    • Post Reply
    • Bookmark Topic Watch Topic
    • New Topic