Project-based engineering competition in upper-level engineering laboratory¶

Ryan C. Cooper¶

UConn engineering logo

About me - Ryan C. Cooper¶

GitHub: octocat image @cooperrc

email: ryan.c.cooper@uconn.edu

Twitter: @cooperrc84

  • Assistant Professor-in-Residence
  • University of Connecticut Mechanical Engineering Department
  • Father of two boys
  • Runner, bicyclist, and skateboarder
Out[13]:

Accessing this presentation and *more*¶

  • My website has Open Educational Resources: cooperrc.github.io
    • Join me, redistribute work, add suggestions
    • Textbooks should be open and educational
  • These slides are hosted at github.com/cooperrc/2021-ASEE-NE_PjBL-labs
    • code to evaluate grades included
    • student data is not online (some info is PII)

What is Project/Problem-Based Learning¶

Problem-based Learning (PBL)¶

  • Identify a problem
  • Give clear instructions or steps to solve the problem

Project-Based Learning (PjBL)¶

  • Identify a clear goal
  • Students iterate problem-solving approaches
  • Design their own approach

Motivation - Why am I using Projects and Problems?¶

  • Students expect a hands-on engineering experience
  • We, professors, give them an analytical background
  • Computational work (i.e. Jupyter) s used here to connect empirical+analytical
    kant-vs-hume and analytical-vs-empirical venn diagram

Motivation - Why am I using Projects and Problems?¶

Most important ABET outcomes ranked by practicing engineers, employers, and recent graduates¶

1. problem solving¶

Students need practice solving problems

2. communication¶

Students need to share results e.g. improve technical writing

Grading PBL + PjBL work: Specifications grading¶

  • Lab reports are given scores
  • Scores that meet overall specifications (>70) pass
  • Detailed rubric focuses TA feedback
  • Token system:
    • Groups have 2 tokens
    • Token is used for late assignment or revision
  • Also called competency-based or mastery-based [I don't like this term]

Methods: Fall 2018 + Fall 2019 - Fall 2020¶

  • 5 PBL labs with content in Jupyter (Labs 0-4)
  • 1 PjBL lab open-ended with definite goal Lab 5
    • $150 cash prize most accurate
    • party for most precise section
  • Assess technical writing improvements with rubric
  • survey students for most important skills
  • survey students for capstone preparedness
  • measure total error in student Project predictions
  • I assumed improvements in technical writing scores are improvements in communication
  • I feared students with >70 on lab 0 would decrease scores

Methods: Fall 2018 + Fall 2019 - Fall 2020¶

Results: Labs 0-4 saw continuous improvement¶

  • The technical writing scores continued improving throughout semester
  • Red-dash line shows minimal passing grade
  • Average continually improved

Individual technical writing showed continuous improvement¶

  • Average slope of grades from 0-4 calculated
  • Plotted here vs grade on Lab 0
  • Over 56% showed improvement
  • 30-40% maintained high quality work
  • less than 4% of class could not meet standards, but showed improvemnt

Problem-solving evaluated with project results¶

  • The students were
    • given access to experimental equipment from labs 0-4
    • told to measure its mass with a vibrating cantilever beam
  • Error = (reported value - the actual value) in mass reported in
    • Fall 2018: 18 ± 33 g
    • Fall 2019: 11.4 ± 27 g
    • actual objects: 32 ± 2 g

Student preparedness survey¶

  • polled the 2019-2020 senior capstone teams
  • Students’ comments:
    • “Was a great and helpful class”
    • “Great class! Very helpful for senior design”,
    • “ME3263 was a great course for technical writing.”
  • Students asked how useful each skill
  • Over 50% of the class of 270, agreed that all eight skills were useful

Student preparedness survey¶

  • 50% of the class considered technical writing to be a crucial skill
  • Fall 2018 and Fall 2019, over 45% felt prepared
  • students hadn’t taken the course less than 30% felt prepared
  • statistically significant difference between preparedness that took the PjBL course and those that did not. This measurement

Conclusions¶

  • Students need opportunities for PBL and PjBL in every class
    • it improves tech writing
    • it connects empirical and analytical methods
  • Jupyter is an awesome way to build open educational resources
    • connects empirical to analytical methods
    • interactive materials encourage experimentation and exploration
    • free and open: you can download/modify/redistribute great for building student portfolios
  • Critera-based grading is first step to my ungrading journey
    • >80% of students liked the system
    • the failing cases were a mix of apathetic and enraged

Lessons learned¶

  • Community is crucial:
    • Establish safe and respectful communication
    • Encourage collaboration
  • Students thrive with goals
  • Test-taking students scared of open-ended assignments
    • Rubrics help (detailed, but not prescriptive)
    • Grace in deadlines and revisions helps

Future work¶

  • Criteria-based grading is my first step towards ungrading
  • My classes based upon this work have moved to unlimited revisions no tokens or P/F
  • Ideally, students create portfolio of engineering work
  • My role would be:
    • identify resources
    • troubleshoot modeling/testing
    • teaching and learning

Question + Discussions?¶

In [ ]: