Lunch and Learn: Fostering an Inclusive Classroom

Logo for Lunch and Learn program showing the words Lunch and Learn in orange with a fork above and a pen below the lettering. Faculty Conversations on Teaching at the bottom.On Thursday, February 15, the Center for Educational Resources (CER) hosted the third Lunch and Learn—Faculty Conversations on Teaching, for the 2017-2018 academic year.

Anne-Elizabeth Brodsky, Senior Lecturer, Expository Writing and Karen Fleming, Professor, Biophysics, presented on Fostering an Inclusive Classroom.

Anne-Elizabeth Brodsky led off by describing the Expository Writing program. [See presentation slides.] Students learn in small class settings (10-15 students) how to write and academic argument using evidence. The academic argument is framed as an ongoing conversation—they say (identifies the problem or viewpoint), I say (recognizes a flaw in the argument and establishes a thesis to correct the flaw), so what? (what’s at stake or what’s next?)—to which the student is contributing. The vast majority of her students are in the STEM fields. This is the background for Brodsky’s introduction of diversity and inclusion in the classroom.

Students bring their complex identities and a sense of whether or not they feel they belong on campus to the classroom. Brodsky seeks to make sure that her curriculum includes reading diverse authors. Students are given a number of choices for each major writing assignment so that they can find something that interests and speaks to them. Additionally, students do informal writing on topics fincluding What is College For? Learning (science of learning, recent research, advice), School (campus life, student realities, ideas for redesign), Stories—why they matter, Words (language change, texting, big data). For each topic area Brodsky identifies a number of articles that students can choose from to read and write a response.  She tries to make topics current and relevant, and to be flexible, for example, after the death of Freddie Gray in April 2015, “I offered my students supplemental reading. I hoped to show how the JHU community engaged in local, ‘real-life’ struggles with rigorous, historically responsible, and creative thinking.” She encourages interdisciplinary conversations so that students don’t shut themselves off from possibilities.

Brodsky works with students who are afraid of writing and finds that different approaches work for different students. She is an advocate of Universal Design for Learning (UDL). In the classroom she practices eight ways of teaching.

  1. Giving students permission to fail successfully. The writing process is all about Slide from Anne-Elizabeth Brodsky's presentation listing 8 ways of learning.drafts and revisions; students learn that their first efforts may not work well, but they can learn from mistakes and rewrite.
  2. Using color highlighting while reading, by Brodsky and the students, to understand how an academic argument works.
  3. Sorting through key words used in an academic argument to understand how language works. For example, when identifying a flaw in an argument, key words are repudiate, contest, reject. Next, Brodsky has students perform textual analysis, by looking at a text and deciding whether it is either a conversation over coffee or actual textual analysis. For example, the text being analyzed is a comment on this quote (taken from a John Oliver segment): “A former British prime minister once described refugees as a “swarm of migrants.” Answer A is: “That made me crazy.” [conversation] Answer B is: “The word ‘swarm’ suggests invasion, threat, as well as equating humans to insects.” [textual analysis]
  4. Using Lego building blocks with statement taped to the sides, students “build their argument.”
  5. Learning to analyze and interpret by practicing describing things in different ways.
  6. Employing different ways of talking and discussing in the classroom. Students prepare elevator talks, engage in think/pair/share activities, anonymously share worries and wishes on index cards, and write questions to establish knowledge.
  7. Crafting a succinct “elevator pitch” to streamline their own academic argument.
  8. “What I mean here is to help and expect students to engage fully in their work on campus—not just their science brains for science, or even just their intellectual skills for their courses. Rather, they come here with richness of experience far beyond the academic classroom.”

In engaging students, Brodsky helps them to see frustration as unrealized potential and uncertainty as curiosity. She drew on a 2013 interview with British psychoanalyst and essayist Adam Phillips: “A student who’d been frustrated for weeks by my suggestion that thesis statements were probably not good things to have about works of literature began to recognize that her frustration was partly a fear of freedom. ‘You’re really just increasing my allowance,’ she said.”

By engaging in inclusive teaching practices, Brodsky has given her students the freedom to learn.

Karen Fleming started off by stating her goals for an inclusive classroom: To create a learning environment where everyone feels safe to express ideas and questions, where discussion brings multiple diverse questions and all opinions are considered, and where student experiences of marginalization are minimized. Two of the many challenges that instructors face are unconscious bias and student stereotype threat. [See presentation slides.]

Fleming noted that 99% of our cognitive processing uses unconscious reasoning. Unconscious bias stems from our experiential expectations; bias being an error in decision making. Expectation bias is grounded in stereotypes such as women are not as good at math or spatial learning; Hispanics and African Americans are lower achieving than Whites; Hispanics, African Americans, poor, and obese people are lazy; male athlete are jocks (muscular but not smart); Asians have higher math abilities; and blondes are less intelligent than brunettes. Stereotypes are cultural and both faculty and students have biases.

Fleming described a research study on science faculty and unconscious bias [Moss-Racusin et al. (2012) “Science faculty’s subtle gender biases favor male studentsPNAS 109: 16474-79.] The study was headed by Jo Handlesman, Professor, Department of Molecular, Cellular & Developmental Biology, Yale University. Fleming walked through the study details, noting that the institutions selected—research intensive universities—were geographically diverse, the demographics on faculty participants agree with national averages, and it was a randomized double-blind study (n = 127). Science faculty in biology, physics, and chemistry were asked to rate the application of a student for a lab manager position in terms of competency and hireability. Faculty participants were randomly assigned an application with a male or female name—the applications were otherwise identical and designed to reflect high but not outstanding qualifications.

Both male and female faculty participants showed gender bias by rating the male applicant as significantly more competent and hireable than the female applicant. They were willing to offer a significantly higher starting salary (and more career mentoring) to the male applicant. And, perhaps surprisingly to women in the audience, both female and male faculty were equally likely to show gender bias favoring male applicants and to offer the female candidates a lower salary. One possible explanation? Male stereotypes meet the expectation for the lab manager position.

Slide from Karen Fleming presentation showing male and female stereotypes.Expectation bias occurs when we hold different groups accountable to different standards. When a man performs well in a traditional male-type task, this performance is expected. When a woman performs well in a traditional male-type task, this conflicts with stereotypic expectations. As a result, her performance is closely scrutinized, and she is required to repeatedly prove her competence.

We can check our biases using tests set up by Project Implicit. “Project Implicit is a non-profit organization and international collaboration between researchers who are interested in implicit social cognition – thoughts and feelings outside of conscious awareness and control. The goal of the organization is to educate the public about hidden biases and to provide a ‘virtual laboratory’ for collecting data on the Internet.” Look for tests on bias concerning gender, race, religion, disability, age, sexuality, weight, skin tone, and more  in the Social Attitudes section.

Fleming then discussed the issue of stereotype threat. For students, performance in academic contexts can be harmed by the awareness that one’s behavior might be viewed through the lens of stereotypes. And, culturally-shared stereotypes suggesting poor performance of certain groups can disrupt performance of an individual who identifies with that group. She referenced a study done by Steele and Aronson in 1995 [CM Steele and J Aronson (1995) Journal of Personality and Social Psychology 69: 797-811.] At Stanford, 114 undergraduate students volunteered for the study. They were given a test composed of Verbal GRE questions. First they took a test that was presented as diagnostic, a measure of intrinsic ability, and then a test presented as non-diagnostic, a laboratory tool for studying problem solving. The results showed that invoking stereotype threat (diagnostic test) affects performance of students who identified as African-Americans in an evaluation setting. Further, having to identify their race before taking the test, “race-priming,” increases the results of stereotype effect on performance in an evaluation setting.

How can instructors be more inclusive? Fleming offers these suggestions.

  • Establish a Growth Mindset
  • Set clear, equal and high but achievable expectations for all students
  • Include material created by people of different backgrounds
  • Pay attention to group dynamics (especially compositions of small groups).

As both Fleming and Brodsky reminded the audience, instructors can change the way they teach to be more inclusive, identify their biases, and work to establish a positive classroom climate.

Macie Hall, Senior Instructional Designer
Center for Educational Resources

Image Sources: Lunch and Learn Logo (© CER), slides from Brodsky and Fleming presentations

 

Grading in the fast lane with Gradescope

[Guest post by Scott Smith, Professor, Computer Science, Johns Hopkins University]

Three speedometers for quality, grades per hour, and efficiency.Grading can be one of the most time consuming and tedious aspects of teaching a course, but it’s important to give prompt and meaningful feedback to your students. In large courses, aligning grading practices across multiple teaching assistants (TAs) necessitates a level of coordination that includes scheduling grading meetings, reviewing materials for correct answers, and calibrating point evaluations, all of which can take up valuable time during the semester.

In courses that teach programming, we typically assign students projects that require them to write programs to solve problems. When instructors grade this type of assignment, they not only have to observe the program’s results but also the student’s approach. If the results are not correct or the program doesn’t run, we have to spend time reviewing hundreds of lines of code to debug the program to give thoughtful feedback.

In the past, my method for grading assignments with my TAs may have been arduous but it worked. However, last year, no TAs were assigned to my Principles of Programming Languages course. Concerned that I wouldn’t have enough time to do all the work, I looked for another solution.

Consistent grading and providing meaningful feedback for student’s every submission, especially with multiple teaching assistants (TAs) can be challenging. Typically, when grading, I would schedule a time to sit down with all of my TAs, review the assignment or exam, give each TA a set of questions to grade, pass the submissions around until all were graded, and finally calculate the grades. When a TA had a question, we could address it as a group and make the related adjustments throughout the submissions as needed. While this system worked, it was tedious and time consuming. Occasionally, inconsistencies in the grades came up, which could prompt regrade requests from students. I kept thinking that there had to be a better way.

About year and a half ago, a colleague introduced me to an application called Gradescope to manage the grading of assignments and exams. I spent a relatively short amount of time getting familiar with the application and used it in a course in the fall of 2016, for both student-submitted homework assignments and in-class paper exams. In the case of the homework, students would upload a digital version of the assignment to Gradescope. The application would then prompt the student to designate the areas in the document where their answers can be found so that the application could sort and organize the submissions for the ease of grading. For the in-class exams, I would have the students work on a paper-based exam that I set up in Gradescope with the question areas established. I then would scan and upload the exams so that Gradescope could associate the established question areas to the student submissions automatically. The process of digitizing the completed tests and correlating them to the class roster was made easy with a scanner and Gradescope’s automatic roster matching feature. Gradescope became a centralized location where my TAs and I could grade student work.

There are a few ways to consider incorporating Gradescope into your course. Here is a non-exhaustive list of scenarios for both assignments and exams that can be accommodated:

  • Handwritten/drawn homework (students scan them and upload the images/PDFs)
  • Electronic written homework (students upload PDFs)
  • In-class exams (instructor scans them and uploads the PDFs)
  • Coding scripts for programming assignment (students upload their program’s files for auto-grading)
  • Code assignments graded by hand (students upload PDFs of code)

The real power of Gradescope is that it requires setting up a reusable rubric (a list of competencies or qualities used to assess correct answers) to grade each question. When grading, you select from or add to the rubric to add or deduct points. This keeps the grading consistent across multiple submissions. As the rubric is established as a part of the assignment, you can also update the point values at any time if you later determine that a larger point addition/deduction is advisable, and the grade calculations will update automatically.

Screenshot from Gradescope--Review grade for assignment feature.

Screenshot of Gradescope’s Review Grade for an assignment

After being informed that I wouldn’t have any TAs for my Principles of Programming Languages course the following semester, I was motivated to use one of Gradescope’s [features, the programming assignment auto-grader platform. Being able to automatically provide grades and feedback for students’ submitted code has long been a dream of instructors who teach programming. Gradescope offers a language-agnostic environment in which the instructor sets up the components and libraries needed for the students’ programs to run. The instructor establishes a grading script that is the basis for the analysis, providing grades and feedback for issues found in each student’s submitted program.

Overall, the use of Gradescope has reduced time spent grading and improves the quality of feedback that I am able to provide students. For instance, when I release grades to the students, they are able to review each of the descriptive rubrics that were used when grading their submissions, as well as any additional comments. Auto-grader was really the star feature in this case. Students were able to submit their code, determine if it would run, and make corrections before the deadline to increase their chances of a better grade. There are features to reduce the number of allowed submissions, but I choose not to set a limit so that the students could use an iterative approach to getting the right solution.

Gradescope is only effective if your rubrics and grading criteria are well thought out, and the auto-grading scripts require some time to set up.  Creating the grading scripts for the programming assignments may seem time intensive, but by frontloading the work with detailed rubrics and test cases, more time is saved in the grading process. The value of this preparation scales as enrollment increases, and the rubrics and scripts can be reused when you teach the course again. With more time during the semester freed up by streamlining the grading process, my TAs and I were able to increase office hours, which is more beneficial in the long run for the students.

Screenshot showing student's submission with rubric items used in grading.

Student’s submission with rubric items used in grading

The process for regrading is much easier for both students and instructors. Before Gradescope, a regrade request meant determining which TA graded that question, discussing the request with them, and then potentially adjusting the grade. With the regrade feature, students submit a regrade request, which gets routed to that question’s grader (me or the TA) with comments for the grader to consider. The grader can then award the regrade points directly to the student’s assignment. As the instructor, I can see all regrade requests, and can override if necessary, which helps to reduce the bureaucracy and logistics involved with manual regrading. Additionally, regrade requests and Gradescope’s assignment statistics feature may allow you to pinpoint issues with a particular question or how well students have understood a topic.

I have found that when preparing assignments with Gradescope, I am more willing to create multiple mini-assignments. With large courses, the tendency would be to create fewer assignments that are larger in scope to lessen the amount of grading. When there are too few submission points for students who are deadline oriented, I find that they wait till the last few days to start the assignment, which can make the learning process less effective. By adding more assignments, I can scaffold the learning to incrementally build on topics taught in class.

After using Gradescope for a year, I realized that it could be used to detect cheating. Gradescope allows you to see submissions to specific questions in sequence, making it easy to spot submissions that are identical, a red-flag for copied answers. While not a feature, it is an undocumented bonus. It should also be noted that Gradescope adheres to FERPA (Family Educational Rights and Privacy Act) standards for educational tools.

Additional Resources:

  • Gradescope website: https://gradescope.com
  • NOTE TO JHU READERS ONLY: The institutional version of Gradescope is currently available to JHU faculty users through a pilot program. If you are faculty at Johns Hopkins University’s Homewood campus interested in learning more about how Gradescope might work for your courses, contact Reid Sczerba in the Center for Educational Resources at rsczerb1@jhu.edu.

 

Scott Smith, Professor
Department of Computer Science, Johns Hopkins University

Scott Smith has been a professor of Computer Science at Hopkins for almost 30 years. His research specialty is programming languages. For the past several years, he has taught two main courses, Software Engineering, a 100 student project-based class, and Principles of Programming Languages, a mathematically-oriented course with both written and small programming assignments.

Images Sources: CC Reid Sczerba, Gradescope screenshots courtesy Scott Smith

Quick Tips: How to Manage Your Teaching During a Personal Crisis

There are a number of resources available to help instructors deal with teaching when an institution or our country has experienced a disaster or crisis [see TheSilhouette of a man seated with elbows on his knees suggesting despair. Innovative Instructor post Quick Tips: Teaching in Challenging Times and Facilitating Difficult Discussions, November 15, 2016], and this blog has covered how to handle student crises, real and not [The Dead Grandmother Syndrome and How to Treat It, December 7, 2016], but what happens when the death or serious illness in the family is your family?

A recent article in Faculty Focus, Teaching During a Personal Crisis (Elizabeth Barnett, February 2, 2018) deals with these issues. Barnett outlines 7 strategies, recommended by experienced colleagues and mental health professionals for teaching during a time of personal difficulty. These are: acknowledge the crisis, triage, consider telling your colleagues, consider telling your students, adjust your expectations, try new routines, protect yourself.

Telling your students may seem like the most controversial piece of advice, but Barnett says:

Wait, you say, this is totally inappropriate. And it would be if you told them in the same way as you did your colleagues, as peers. But if your performance and availability are limited due to the crisis, you can reduce guilt and minimize questions by letting them know something is going on. …[Emphasize] to your students that you don’t need help from them, you have that support elsewhere, and you only want to inform them of the situation.

In other words, you don’t need to go into the gory details, just let your students know that you are experiencing some personal issues that may affect your teaching, for example, that you may be absent, or that someone else may be filling in as instructor for some classes or handling grading of some assignments.

Barnett also offers some tips on dealing with a colleague in crisis, particularly how to listen respectfully and things that you can offer to do to help.

The article is short, but useful. If you have suggestions from your own experiences with dealing with personal crisis, please share them in the comments.

Macie Hall, Senior Instructional Designer
Center for Educational Resources

Image Source: Pixabay.com