ABET Course Assessment
ENGRI/MAE 1170: Intro to Mechanical Engineering
Fall 2009
Assessment by Professor Andy Ruina with help from Teaching Assistant Anoop Grewal
Assessment date January 18, 2010

Documentation for assessment

All documents for this assessment are accessible from this page, from the course home page  or both.

Scans of example student HW, Quiz and Project report(anonymized)

1) HW 11:  good, average & not so good
2) Quiz 2:  good, average & not so good  
3) Final project report:  good, average & not so good

Syllabus (with video links), Quiz 1, Quiz 2,  Homework assignments, lab directions, Project description, Student grades (no names), Student Course evaluation

Course overview

The course is described in the Cornell official Courses of  Study as:

Introduction to fundamentals of mechanical and aerospace engineering. Students learn and understand materials characteristics, the behavior of materials, and material selection for performing engineering function. They also learn fundamentals of fluid mechanics, heat transfer, automotive engineering, engineering design and product development, patents and intellectual property, and engineering ethics. In the final project, students use the information learned to design and manufacture a product.

The course is further defined, for ABET and internal Mechanical Engineering (MAE) purposes by this Syllabet:

Topics covered: The topics, listed below, may be changed at the discretion of the instructor, as long as they are representative of the mechanical engineering curriculum.
   • Forces and Moments
   • Materials and Stresses
   • Design
   • Motion of Machinery
   • Fluids
   • Thermal Sciences/Energy

Finally, the ABET Syllabet describes these

outcomes and their relation to ABET outcomes a-k:

1. Gain experience with unit conversion, estimation, approximations, and critical thinking. (a, e)
2. Gain a basic understanding and ability to solve problems in major areas of the mechanical engineering    curriculum (a, e)
3. Have experience designing and building a device (e.g. a small battery-powered car), and performing and documenting laboratory experiments (b, c, d, g)
4. Become comfortable identifying a system and its interactions with surroundings and using this ap-proach to solve problems (a,e)
.

Philosophical/pedagogical issues

All engineering freshman are expected to take one Introduction to Engineering (ENGRI) course, but the topic need not be corellated with the student's eventual major in any way. ENGRI courses have no pre-requisites and are not required for any other course. Less than half of all Mechanical Engineering majors will have ever taken the introduction to mechanical engineering course (ENGRI 1170).

Rather, according to official policy (page 223 of the complete Courses of Study), an ENGRI course

introduces students to the engineering process and provides a substantive experience in an open-ended problem-solving context.

That is, there is some conflict between the Syllabus (see first paragraph above) and the intended purpose of the course.  It would be inappopriate for an ENGRI course to cover, in depth, any topic which will be covered thoroughly in later courses.  Doing so would actually cause trouble in later classes because some students in those later classes would have been in, say, ENGRI 1170 and some would not, leaving the teacher with an uneven classroom.  The resolution of this conflict, I feel, is to only superfically engage in most of the standard technical engineering content and to emphasize less standard technical content that will not be covered in any systematic way in later courses and also more general problem solving issues.

Design of the course for fall 2009, suggestion for future changes

This was my first semester teaching this course.  Before setting up the course I talked with the previous teachers, Betta Fisher (past 2 years) and Petru Petrina (several years previous) as well as to the undergraduate supervisor of the labs Matt Ulinski.   The overall advice,  my response, and suggested future changes are below.

1) Labs. All agreed that the five labs were basically good.  They functioned well and the students were happily engaged by them. Betta Fisher suggested that the labs be better tied to the lectures.

Thus, the labs were little changed.  Each was streamlined and clarified in minor ways. The lab organization and lab handouts are here. An attempt was made to discuss various aspects of the lab content in the lectures.

A fifth lab, an introduction to hand-tool use, was planned but cancelled due to shortage of time: Matt Ulinski was unusually busy with a) being sick (H1N1) and b) the Solar Decathalon.

Suggested changes: The lab is fine. Reference to the labs needs to be kept in the lectures. The tool-use lab should be introduced, as was planned for this semester.

2) Final Car Design Project. All agreed that this was also a successful part of the course in past years.

Thus the final car design project was only slightly modified.  The design-ahead-of-time aspects were de-emphasized as, at this level, these are a form of make-work with little real content.  Rather, students designed by imagination, trial and error.   The overall organization of the final design project are described in items 7-9 on the lab page.

Suggested changes: The project is good as it is.  Minor improvements could be:

a) Even more stringent requirements that the cars function, at least somewhat, well before the final competition.  Thus, the project should have an extra week, and working cars, in some form, should be demonstrated 2 weeks before the competition.

b) The motor needs to be weaker or the gear boxes stronger so fewer gears are stripped.

3) LECTURES and SYLLABUS.  These were not especially popular in recent years.   People felt the lectures were not well correlated to the lab or project and that the various in-semester reports were somewhat onerous.

Thus I dropped the old syllabus. The new syllabus is shown here (where you can also see videos of the lectures). I decided to emphasize things that I felt were most important: basic computer skills and basic problem solving skills.  I chose to weave these around the labs and especially around the final project. 

Breaking the Course Description (at the top of this page) into small pieces the coverage looks like this.

Introduction to fundamentals of mechanical and aerospace engineering.

The first lecture was exactly a general introduction. The rest of the course was, I feel, a reasonable introduction to a range of Mechanical Engineering issues.

Students learn and understand materials characteristics, the behavior of materials,

This was basically limited to  linear elasticity, hydrostatics, and linearly viscous fluids.

and material selection for performing engineering function.

This was not covered.

They also learn fundamentals of fluid mechanics,

Especially hydrostatics.  Next most emphasized was the quadratic nature of fluid drag in many engineering applications.  There was one lecture on linear viscous fluids.

heat transfer,

Very brief mention in lecture. Somewhat covered in the lab.

automotive engineering,

Various automotive engineering issues were covered at length. Especially the relation between power and acceleration, taking account mass, friction, air drag and gear ratios.

engineering design and product development,

Only in the sense that the final project involved design and building. As practiced in this course, this is an analogue for more professional design and development, not direct training in that experience.

patents and intellectual property,

Not covered, but for part of one lecture.

and engineering ethics.

One full guest lecture.

In the final project, students use the information learned to design and manufacture a product.

The better students did use the lecture/homework material to design and improve their final car project.

New material added this semester, and not mentioned in the syllabus, above included:

Introduction to the physical meaning of differential equaitons.

Numerical solution of differential equations.

Introduction to Matlab hacking (not formal Computer Science).

Challenging problems, not of the here's-the-formula-you-plug-it-in type.

Suggested changes: Simple improvements that should greatly improve the student experience are discussed below, after the discussion of the student surveys.

 

Outcomes

The ABET Syllabet describes various outcomes, listed below with some discussion about their place in this course.

1. Gain experience with unit conversion, estimation, approximations, and critical thinking. (a, e)

All of these areas were covered extensively in the lectures and in the homework challenges.


2. Gain a basic understanding and ability to solve problems in major areas of the mechanical engineering curriculum (a, e).

I do not trust any simple quantitative measure of improvement in these areas.  But students were certainly given ample opportunity to engage towards the end of gaining usch abilities.


3. Have experience designing and building a device (e.g. a small battery-powered car), and performing and documenting laboratory experiments (b, c, d, g)

Done.

4. Become comfortable identifying a system and its interactions with surroundings and using this ap-proach to solve problems (a,e)

Again, measuring progress towards these ends would be a major project in itself.  Students certainly had many opportunities to engage in such issues.

 

Evaluation and Student Course Surveys

Because this was not a skills based class I do not think quantitative evaluation of student performance on exams etc would be a good measure of the course success.  Rather, given that we offered competent and useful information, the useful indicators are

1) Student Engagement and

2) Student satisfaction.

Engagement. By superficial measures, students in this course were engaged.  The average student missed one lecture or less (see the course grade sheet), handed in all the homeworks, did all the labs, and was actively engaged in the final design project. Further, only 5 of 52 students reported spending 4 or fewer hours per week on this course outside of class.

Student satisfaction. Course surveys are a reasonable gauge of this. So I take them seriously. Students were told by email:

Please complete the Online Course survey. We get a list of all the people who do the survey. You will get one point bonus on your grade for doing the survey.

After grades are in we get the surveys too (but not the names that go with them). We will read every one. There is a reasonable chance I (Andy) will teach the class again next year so the "good things" and "bad things" are most useful.

Although the syllabus was qute different in fall 2009 than fall 2008, the teachers were different, and the organization was different, the numerical evaluations are nearly identical but for the higher evaluations of the great teaching assistants in Fall 2009.

Fall 2009 survey results (completed by 52 of 60 students)
(compare to Fall 2008 survey results  completed by 49 of 60 students)

In short, as explained below, the course needs to change to improve student satisfaction.

Key findings (numerical questions), as numbered on survey:

2. Utility of Homework:  More than 30% (16/52) of the students found the homework to be less useful than homework in their other courses.  Although the evaluations have not been tabulated in a way that allows looking at such correlations, I suspect that the 16 students who found the homework of less-than-average educational valure are also mostly those who were struggling too much with the homework.

3. Lab (Presumably this also includes the final car project.): Only 3 of 52 students did not find these educational (score of 1/5 or 2/5).  And as a guess, I would suspect that these 3 are likely people who were bitter about the course.   75%  (39/52) thought that the lab/project was of greater than average educational value. The mode of the distribution is 5/5. From the student point of view, which is of greatest importance, the lab and project are quite good.

5 & 8 Lecturer: Both Betta Fisher and I got similar evaluations. On the positive side, over half the class thought I was at least of average stimulation and almost half thought I was almost of average clarity.  On the negative side is this:  more than 25% of the class thought I was bad (14/52 students giving a score of 1/5 for overall effectiveness).   The comments reveal some of the sources of this dissatisfaction.

9-12 TAs: The TAs got high evaluations from the students with an overall rating of about 4/5. The course did have dedicated and friendly TAs. I also liked working with them.

13 Course overall:  40% of students (18/45) found this course to be worse than average (scoring 1/5 or 2/5).

14 Time investment: The same number of students spent 5-8 hours as spent 9-15 hours.  So the median student spent about 8-9 hours per week on this course outside of class.  This is about right for a 3 credit class, I think.  Unfortunately I cannot tell the correlation between student satisfation and the time they spent on the course. 

15 Overall level:  70%  of the students   (36/52) thought the course was pitched too high (scoring 4/ or 5/5 on their being under-prepared). Similarly on question 4, 68% of the students thought the 9es were unfairly difficult.

Key findings from student comments on course evaluations:

Although the numerical ratings for Betta Fisher's version of the course (fall 2008) and mine are nearly identical, the things I need to improve are different from the things she would need to improve were she to teach the course again. Without changing the content or organization much, the following changes would, I think, address most of the present student dissatisfaction with the course.

1) Syllabus. The course should have a syllabus posted at the start of the semester. This semester the course did not have a pre-planned or pre-posted syllabus. I do not think it would have helped to have one, given that making such a syllabus would have involved too much guess work. Next semester that can be fixed.

2) Matlab. Despite student comments to the contrary, I did not assume students had any programming background. However, based on student comments, the Matlab was obviously pitched too high.  I over estimated the weaker student's abilities to jump into the world of programming quickly. Matlab should be introduced more slowly, with direct and simple homeworks developing bit by bit. Topics such as loops need to be introduced carefully and  deliberately.   The teaching of, say, ODE23 should be dropped and students should stick with their own Euler integration (the Matlab syntax for such things is just too strange for beginners).

3) Exams and grading. The exam and grading policies need to be laid out explicitly and clearly at the start of the course.  The student comments show clear and direct misunderstanding and miscommunication about various grading  issues.  Probably best to simply have evening prelims and a standard final exam. And the grading formula needs to be given as such.

4) Homework difficulty. I wanted this to be a course about solving hard problems.  But the majority of students are not comfortable with that.  The course needs to have simple skill-based problems to build the student's confidence.  The hard problems need to be extra after the students build skills based on simple algorithms.

5) Lecture level. The students need some simple direct facts and algorithms that they can follow. The more abstract, problem-solving side needs to be introduced only after the more basic material.   Similarly, despite what some students thought, the course did not have "advanced physics concepts". Nonetheless, the basic physics needs to be taught a bit more deliberately. For example, the long section on gears will be much more clear to the students if it starts with a more clear and slo introduction to the ideas and use of moment balance.

6) Book. The course did not have one.  I don't think it needs one. There are few books available for this course and I did not find one I was comfortable with.  However some students would be helped by being pointed to directly relevant reference material when relevant.