ENCMP100 complaint
Apr. 27th, 2005 12:07 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
kyrasantaeIf you are interested in signing in support of this please let me or ![[livejournal.com profile]](https://www.dreamwidth.org/img/external/lj-userinfo.gif)
forgottenlord know immediately. Thanks.
Tuesday, April 26, 2005
Dear Trevor Buckle:
In the first year program of studies, no course is more consistently disliked by a large number of students than ENCMP 100. The reasons behind this general dislike of the course are due to the policies and methods of teaching and evaluation of the course material. While these policies and methods are not well-liked by the students, they are valid and should not be significantly changed, as they are an immense improvement from those of the Winter 2004 term. However, the final exam for the Winter 2005 session did not accurately reflect the material that was covered in the course and was thus unfair to the students writing the exam.
Pointers are one of the most important, but also one of the most complex, concepts of C++ programming. An understanding of pointers is crucial for a full understanding of the language, since they explain how the language interacts with computer hardware, memory in particular. Such a complex concept requires time to understand, and although it is essential for advanced C++ programming – such as for students interested in entering the field of computer engineering – it is arguable whether it is necessary or even useful for someone who will only be writing basic programs.
Not only was the course material pertaining to pointers allocated to the end of the course and any mention of pointers to explain other aspects of C++ avoided altogether, a total of only one to one-and-a-half lecture’s worth of time was spent on pointers. Even though computer programming is offered as a Career and Technology Studies (CTS) course in many high schools, it is not a mandatory high school course. ENCMP 100 instructors acknowledge that students enrolled in this course come from a variety of backgrounds – from students who program on a regular basis as a hobby to those who have never touched a programming language in their lives. Therefore it is unreasonable to expect that 80 to 120 minutes of lecture is sufficient to develop an understanding of a concept that is a core concept of C++.
The number of questions about pointers on the final exam and their point values were extremely disproportionate to the amount of time spent on the subject and other indicators throughout the course – 44 points out of a total of 222 points, including six questions that were worth six points apiece. There were no assignments that involved the use of pointers and the handful of exercises on pointers that were posted on CodeLab was not for credit. Instructors also stressed that pointers would not be a significant topic on the final exam, and review instructors from the Math and Applied Sciences Centre (MASC) echoed this assertion.
Depending on whether the final exam emphasized only the latter part of the term or was cumulative, calculations indicate that a reasonable number of points allocated to pointers on the exam would be between nine to sixteen points (4 – 7%)1. These numbers fall significantly short of the actual 20% of the exam that was devoted to pointers.
Only in the CodeLab assignments did we discover any evidence that such a weighting would be appropriate. Supposing the final exam was based only on material from after the midterm, the amount of available CodeLab exercises on pointers was 46 points, or 20.4%. Since the final exam was cumulative, the percentage of CodeLab exercises relevant to pointers dropped to 28 points, or 12.8%, once all exercises available were taken into account2. Of course, these figures disregarded the fact that these exercises were not for credit, since it was reasonable to expect that students would look at recommended but not-for-credit material. Additionally, the CodeLab component of the course was entirely optional.
Students who have taken ENCMP 100 before us have remarked that this course often generates a double curve – which splits the class into those who have prior C++ knowledge and those who do not – rather than a single curve which would be an indicator of how much students learned during the course. We fear that the results of the final exam this term will split us by such a curve. Although we in no way demand that our grades be readjusted or the papers be regraded, we believe that the ENCMP 100 instructors did not take due care in ensuring that exam content reflect what was stated and taught in class, and it is our hope that more consideration be put into this in the future.
Sincerely,
Mildred Lau
First-year student
Scott Murray
Second-year – Computer Engineering (Software)
1 Assuming a non-cumulative exam,
(222 points on exam)×(1 lecture on pointers)÷(14 post-midterm lectures) = 15.9 points.
Assuming a cumulative exam,
(222 points on exam)×(1 lecture on pointers)÷(26 total lectures) = 8.5 points.
2Assuming a non-cumulative exam,
(222 points on exam)×(16 exercises on pointers)÷(78 post-midterm exercises) = 45.5 points.
Assuming a cumulative exam,
(222 points on exam)×(16 exercises on pointers)÷(125 total exercises) = 28.4 points.
forgottenlord know immediately. Thanks.Tuesday, April 26, 2005
Dear Trevor Buckle:
In the first year program of studies, no course is more consistently disliked by a large number of students than ENCMP 100. The reasons behind this general dislike of the course are due to the policies and methods of teaching and evaluation of the course material. While these policies and methods are not well-liked by the students, they are valid and should not be significantly changed, as they are an immense improvement from those of the Winter 2004 term. However, the final exam for the Winter 2005 session did not accurately reflect the material that was covered in the course and was thus unfair to the students writing the exam.
Pointers are one of the most important, but also one of the most complex, concepts of C++ programming. An understanding of pointers is crucial for a full understanding of the language, since they explain how the language interacts with computer hardware, memory in particular. Such a complex concept requires time to understand, and although it is essential for advanced C++ programming – such as for students interested in entering the field of computer engineering – it is arguable whether it is necessary or even useful for someone who will only be writing basic programs.
Not only was the course material pertaining to pointers allocated to the end of the course and any mention of pointers to explain other aspects of C++ avoided altogether, a total of only one to one-and-a-half lecture’s worth of time was spent on pointers. Even though computer programming is offered as a Career and Technology Studies (CTS) course in many high schools, it is not a mandatory high school course. ENCMP 100 instructors acknowledge that students enrolled in this course come from a variety of backgrounds – from students who program on a regular basis as a hobby to those who have never touched a programming language in their lives. Therefore it is unreasonable to expect that 80 to 120 minutes of lecture is sufficient to develop an understanding of a concept that is a core concept of C++.
The number of questions about pointers on the final exam and their point values were extremely disproportionate to the amount of time spent on the subject and other indicators throughout the course – 44 points out of a total of 222 points, including six questions that were worth six points apiece. There were no assignments that involved the use of pointers and the handful of exercises on pointers that were posted on CodeLab was not for credit. Instructors also stressed that pointers would not be a significant topic on the final exam, and review instructors from the Math and Applied Sciences Centre (MASC) echoed this assertion.
Depending on whether the final exam emphasized only the latter part of the term or was cumulative, calculations indicate that a reasonable number of points allocated to pointers on the exam would be between nine to sixteen points (4 – 7%)1. These numbers fall significantly short of the actual 20% of the exam that was devoted to pointers.
Only in the CodeLab assignments did we discover any evidence that such a weighting would be appropriate. Supposing the final exam was based only on material from after the midterm, the amount of available CodeLab exercises on pointers was 46 points, or 20.4%. Since the final exam was cumulative, the percentage of CodeLab exercises relevant to pointers dropped to 28 points, or 12.8%, once all exercises available were taken into account2. Of course, these figures disregarded the fact that these exercises were not for credit, since it was reasonable to expect that students would look at recommended but not-for-credit material. Additionally, the CodeLab component of the course was entirely optional.
Students who have taken ENCMP 100 before us have remarked that this course often generates a double curve – which splits the class into those who have prior C++ knowledge and those who do not – rather than a single curve which would be an indicator of how much students learned during the course. We fear that the results of the final exam this term will split us by such a curve. Although we in no way demand that our grades be readjusted or the papers be regraded, we believe that the ENCMP 100 instructors did not take due care in ensuring that exam content reflect what was stated and taught in class, and it is our hope that more consideration be put into this in the future.
Sincerely,
Mildred Lau
First-year student
Scott Murray
Second-year – Computer Engineering (Software)
1 Assuming a non-cumulative exam,
(222 points on exam)×(1 lecture on pointers)÷(14 post-midterm lectures) = 15.9 points.
Assuming a cumulative exam,
(222 points on exam)×(1 lecture on pointers)÷(26 total lectures) = 8.5 points.
2Assuming a non-cumulative exam,
(222 points on exam)×(16 exercises on pointers)÷(78 post-midterm exercises) = 45.5 points.
Assuming a cumulative exam,
(222 points on exam)×(16 exercises on pointers)÷(125 total exercises) = 28.4 points.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
Wow
Date: 2005-04-27 09:21 am (UTC)While I think that a double curve such as the one you refer to is inevitable to a point, as those with past experience will naturally be less prone to making errors or oversights in programs, it sounds like there is going to be much more of a separation between those curves than there should be.
Best of luck getting some resolution on this.
Re: Wow
Date: 2005-04-27 12:43 pm (UTC)