Douglas Kiang

This post was originally published in EdWeek Researcher.

When I started teaching computer science at Punahou School, there were two computer science classes: an Intro and an AP course, mostly attended by boys. Occasionally, a girl would enroll in the course, and then more often than not, she would drop the course a few weeks in. “I’m not like them,” one of them told me. “I don’t fit in here.”

Over the years, we have managed to increase the percentage of girls enrolled in our courses from near zero to about 35% and trending upward (update: 42% in 2019). More importantly, we have worked to change the way students perceive computer science. Rather than being primarily for nerds and hackers, students now see computer science at Punahou as a rigorous, challenging course that attracts a surprisingly wide range of students from all backgrounds. Based on our experiences and research, here are some strategies for engaging more girls in computer science.

Female role models

Increasing the gender diversity of the computer science faculty should be a priority for all schools as it would provide female students—as well as students of color—with role-models and mentors. In a study conducted in a suburban area near Oklahoma University, researchers examined the correlation between an after-school engineering mentoring program led by female college students and the perceptions of sixth and seventh grade girls (Holmes, Redmond, Thomas, & High, 2012). The researchers found that high quality mentoring relationships with a female role model led to an increase in the girls’ confidence in their mathematics ability (Holmes et al., 2012). Similarly, seeing a confident and successful female computer science teacher could have a positive impact on girls and boys in computer science classes.

Another way to highlight positive female role models is to have guest speakers visit class. We have had a female programmer from the Microsoft Xbox team, a freelance web developer, and a music education app developer talk with our classes over Skype. One of my students told me that she did not realize that so much of your time as a developer is spent meeting and talking with other people on your team. She thought the job of a programmer was just sitting at a computer by yourself, which was not appealing to her. Year after year, we receive feedback from students that the guest speakers were their favorite part of the course.

Seeing other students who look like you being successful

We have found that seeing another student whom they identify with being successful in the course makes it easier to visualize themselves having similar success. Harvard’s highly popular Introduction to Computer Science course, CS50, uses undergraduate teaching fellows to teach its small sections, which are differentiated by “less comfy” and “more comfy” with the subject material. In many cases, students are taught by peers who have just completed the course twelve months earlier, so they are familiar with the most current material, and they understand where students are likely to struggle.

In our high school computer science classes, we have followed a similar model, choosing two students to be teaching assistants the following year. Rather than simply choosing the best students, we look for students who are approachable and representative of the class as a whole. We also try, whenever possible, to give students the freedom to choose which walkthrough or section to attend based on their comfort level with what we are currently covering.

Context over tools

We have found that presenting courses in the context of what problem they are trying to solve, or whom they are trying to help, makes the course more appealing to girls and boys. “Building Socially Responsible Websites for Non-Profits” is more attractive than “Introduction to HTML/CSS”, even if the core skills being taught are the same. 

David Nassar is the Computer Science Chair at Winchester-Thurston School. He created different “flavors” of computer science, teaching students the same basic skills, but giving students the option of learning through a humanities, music, or a math and science focus. He found a dramatic increase not only in the numbers of students choosing to progress beyond the required intro course, but also in the numbers of girls choosing to explore further computer science courses. Emphasizing the purpose of the course (create music) over the tools and language (learn EarSketch or Python) made a big difference.

Community and collaboration

We know that girls prefer to learn in a collaborative environment (Kirk & Zander, 2002). Changing the seating in the room to groups of desks, rather than lecture-style seating, is a good start. But we also work to change the nature of the learning to emphasize, even require, collaboration in order for us to move forward as a group and as a community.

We emphasize “code review” where students take turns showing each other their work once the assignment has been turned in, but before it has been graded. It is helpful for students to see that there are different ways to solve many problems. When we assign independent projects, we often group students who are doing similar projects. Although each individual has to turn in their own project, often students are solving the same problems and can help and support each other.

Encouragement by other adults

The influence of gatekeepers, or those who control course selection and advising for boys and girls in schools, is a key part of the puzzle. They can be valuable allies in building girls’ confidence in taking a computer science course, but they can also harbor outmoded, and often unconsciously biased, ideas about what computer science is today.

An encouraging note from a teacher, a guidance counselor, or a parent can give girls the confidence that if they try computer science, they can succeed, and can give specific reasons why a particular girl would do well in a computer science class, depending on how they know her personally. 

At our school we also regularly do outreach to our deans and college counselors to communicate that computer science is a form of patient problem-solving, that it focuses on concepts and context, rather than on tools or specific programming languages, and that encouragement by others in the community can make a big difference.

Changing culture takes time. There are still too many negative stereotypes of computer “geniuses” and hackers in the media and popular culture, and we could certainly use a greater number of positive portrayals of smart, creative women using their computing skills for good. However,  by focusing on providing positive female role models for computer science in our schools, and supporting girls and young women in their endeavors, we can send a strong message that computer science is for everyone.

References

Holmes, S., Redmond, A., Thomas, J., & High, K. (2012). Girls helping girls: Assessing the influence of college student mentors in an afterschool engineering program. Mentoring & Tutoring: Partnership in Learning, 20(1), 137–150.http://doi.org/10.1080/13611267.2012.645604

Kirk, M., & Zander, C. (2002). Bridging the digital divide by co-creating a collaborative computer science classroom. Journal of Computing Sciences in Colleges, 18, 117–125. Retrieved from https://dl-acm.org

Use the Four Gamer Types to Help Your Students Collaborate

“You are a liar and I will never, ever, trust you again!” The student rose to his feet at the conclusion of a role-playing simulation in our negotiation class, pointing across the room at another student who had misled the other team and then changed his own vote.

The other student, a little shaken, raised his hands in protest. “Dude, relax, it’s just a game.”

“That’s exactly what I mean!” the first student countered. “If you’re willing to do this in a game, what would you be willing to do if the stakes were real?”

The way we play games mirrors how we act in real life. A 2008 study by the Pew Internet and American Life project found that 97% of teens play computer games. On top of that, half of all teens reported playing a video game “yesterday.” Some of these are simple puzzle games on their phone or tablet, but many are surprisingly sophisticated role playing games such as League of Legends or The Elder Scrolls Online. In my own classroom, I have often found that students who don’t seem to have a terribly profound insight into their actions in the classroom are often very self-aware of the choices they make in online gaming situations. 

A new area opens up in the online game world. What do you do? 

A. Explore all the new places? or B. Get all the new loot?

The Bartle Test of Gamer Psychology is an online multiple choice test that asks questions such as the one above. On one of the first days of class, I ask my students to take the Bartle Test online, and report their results to me on a Google Form. It takes about twenty minutes. Based on the students' answers, it assigns a weight in four separate categories: Socializer, Achiever, Explorer, or Killer. Students will often be a mixture of more than one category, although most of the weight will usually be in one category. I use these results to inform how I make groups and how I ask kids to collaborate. The test is based on how students play games. It offers useful insight into how students will react in different situations, and also provides a starting point for me at the beginning of the semester.

Explorers

Students who are Explorers love to explore and wander. In the game world, Explorers want to map the entire area and discover all of its hidden areas. In the classrooms, Explorers love to amass large quantities of knowledge. Their sense of achievement comes from knowing more facts (often obscure ones) and finding secret shortcuts or mnemonics. They love to demonstrate their knowledge to others. I often ask my Explorers to “map out” a new unit of curriculum a couple of weeks before the rest of the class. Can they scout out where the pitfalls might be, and where others might be confused? Can they blaze a trail for the rest of us? Explorers are the closest to what we consider to be the ideal “academics”: those who value knowledge for knowledge’s sake. Yet, Explorers are often those kids who will do all of the work for a project — then forget to turn it in. They’ve already gotten the value from the project. Turning it in for the grade is just not at the forefront of their minds.

Achievers

By contrast with Explorers, students who are Achievers love to gain levels, badges, and awards. In the game world, Achievers gain a sense of accomplishment by being the first to complete something, or by being one of the only ones to win a trophy or reward. They are often very motivated to "beat the game" and move on. In the classroom, Achievers are often most concerned with grades as a measure of their achievement. They want to know the quickest, fastest, or shortest way to the prize. They are often detail-oriented, and I can count on the Achievers in my classroom to help move the others along, simply for the thrill of succeeding.

Socializers

The greatest percentage of my students are Socializers who are motivated by the desire to form meaningful connections and relationships with others in class. In the game world, the Socializers are often the connectors, who help to form questing parties and seem to know everyone online at any given moment. Socializers often form clans, which are groups of like-minded gamers, and when a particular game gets old, Socializers will simply switch games, keeping the clan intact. The game is simply a backdrop for the chatting and interactions that are the true draw. Socializers judge their accomplishments by how many friends, or how many followers, they have. How many of us know students for whom the classroom is simply a backdrop for their friendships and social life?

Griefers

The smallest percentage of students are known as Griefers, or Killers. In the online gaming world, griefers willfully damage and vandalize other people’s creations. They take pleasure in the turmoil they cause in the game world, and in the damage they wreak on others. In the classroom, those students often are the first ones to see if they can “hack” the system, and are often willfully oblivious to the consequences to the community. However, Griefers are often your risk-takers, the ones who are used to starting over with nothing, over and over again — because they’re always being killed — and who don’t mind being wrong. They are likely to have a growth mindset, rather than a fixed one, and if you can draw them into the classroom community, they can have a positive influence on others.

Groups 

What can you do once you have a list of your students and their percentages in each of these categories? You can sort the list by any of the categories and assign groups accordingly. If I want collaborative working groups, I will sometimes ask students to form their own groups with one of each type of gamer in each group. Or, sometimes, I will ask for my Explorers to form a group to do some preliminary research, my Achievers to formulate a plan, my Socializers to publish and share our process, and my Griefers to look for flaws in the game plan.

My Explorers are those kids who can scout ahead for potential problems and pitfalls.

Sorting my students in this way allows me to focus on those areas where I most need to provide support. I know at the beginning of the year, I need to actively bring Griefers into the fold, before they get themselves in trouble. I need to keep the Explorers from wandering off into the woods. I need to broaden the perspective of my Achievers and not let their razor-sharp focus keep them from seeing the big picture. All of my students need to appreciate each other’s strengths, and learn from each other. The Socializers can keep everything moving, and let me know when they sense trouble in the working groups, since they are usually closely attuned to who’s working with whom.

Achievers will make the grade, no matter what.

This activity works best with middle and high school students. The Bartle Test is not scientific, and it’s just a small test with subjective results. But it’s written in a language my gamer students understand, and the results are fun and spark discussion. In my experience, there is no such thing as behavior that is “just a game.” How we play mirrors how we live, and the choices we make. My students generally feel the Bartle Test results are pretty accurate, and it’s a simple starting point for the complex task of adapting my learning environment to meet the needs of every learner in the room.

How would you score? Take the free Bartle Test yourself: http://4you2learn.com/bartle

NEW! One of my students created a free iPhone app that allows you to take the Bartle Test: https://itunes.apple.com/us/app/bartle-test/id887069888?mt=8

NOTE: MMORPG is short for Massively Multiplayer Online Role Playing Game, an online game world that is running consistently, like World of Warcraft or Everquest.

This is some research I conducted at the request of my Principal.

Word Processed vs. Handwritten Essays: What Does the Research Say?

Our school implements a traditional final exam period. Over the course of three days, students are excused from classes and sit for timed exams. In some cases, the final exam counts as much as the grade for an entire quarter, and is worth up to 33% of the student’s semester grade.

As our school reexamines our current practices of assessment, I was asked to examine current research and studies contrasting handwritten and word processed final essays. Our school is a one-to-one school, so all students have access to a personal laptop throughout their school careers. The school is concerned about the validity of handwritten essays in assessing students whose work all semester has been word processed.

There are other questions the school wishes to investigate, as well. If given the choice to handwrite or type, are students who choose to handwrite their essays at a disadvantage when the essays are scored? Do students perform differently when handwriting vs. typing their answers? What are students’ concerns when choosing to handwrite or type?

The University of Edinburgh did a study in which students were given the choice between handwriting or typing essays. Typed and handwritten versions of each essay were produced and graded by a group of four graders. Surprisingly, the researchers found a small but noticeable grading bias toward handwritten essays. Students who chose to word process their essays received scores that were several points lower, a result that has been confirmed in a number of other studies (Russell & Tao 2004, MacCann et al 2002, Sweedler-Brown).

Powers et al (2014) provide two possible explanations in their report on a similar study they performed at Rio Hondo College. When grading a typed essay, readers may inadvertently expect a higher level of polish because the essay is typed out, rather than considering it as the rough draft produced under time pressure and anxiety that it is. The authors also noted a “Reader Empathy Assessment Discrepancy” in which readers tended to identify more closely with the writer if the work was handwritten, citing a “closer identification with the writer’s voice.” 

This suggests that in order to ensure that handwritten and typed essays are graded equivalently, schools should either require that all examinations be either handwritten or typed, or in cases where students are given the choice, educate graders about the presentation effect. Mogey et al. 2006 found that when teachers were made aware of this type of grading bias, it reduced the effect. It is also interesting to note that papers that were printed in a cursive font also seemed to reduce, but not eliminate, the bias (Powers et al 2014). When asked if they wanted to be able to choose handwritten or typed essays, only students who were faster handwriters suggested the need for choice. (Mogey et al 2014)

A second difference between handwritten and typed essays is the quantity of the finished output. Thomas, Paine, and Price (2003) found that students who typed their essays produced considerably more words than those students who chose to handwrite, a finding confirmed by Mogey, Paterson, Burk and Purcell (2006) although another study found the difference to be very small — 9 words more on average (Horkay et al. 2006) In schools where most assignments are given on the computer, students rarely use handwriting to do their work. Connelly, Dockrell and Barnett (2005) found that first-year undergraduates had a handwriting fluency level similar to what would be expected from an 11-year old child. Consequently, requiring students to handwrite their final essays when they have typed all their previous work calls into question the very validity of the assessment.

A number of studies also show that students who are very familiar with technology do better on typed tasks than those who are not (Wolfe, Bolton, Feltovich, and Bangert 1996 and Wolfe, Bolton, Feltovich, and Niday 1996). This suggests that for a number of our students entering as ninth graders, if they have not had the same access to technology, they will be at a disadvantage until they develop their word processing skills (not just typing, but editing and rearranging text too.)

The biggest question of all is how we might use computers more effectively to assess what students have learned. The nature of the word processed essay is essentially a substitution-level task (Puentedura, 2014) and is subject to many of the limitations that handwritten essays face. We should look into the possibility of using computers to generate and display animations, screencasts, video clips, live links, and other digital artifacts to demonstrate learning in ways that would be impossible in a handwritten essay. We might also examine the value of using time constraints on the task of demonstrating learning, and consider piloting untimed tests. 

In the end, the goal of the final exam should be to gather information about what is most important to us to know about our students’ learning. Horkay et al. 2006 suggest that which mode you use depends on what you want to know — do you want to know whether students write well on paper, or digitally, or how well they write in the mode of their choice? How relevant is the ability to write well on paper? Once we answer those questions for ourselves the path should be clearer.

References

Horkay, N., Bennett, R. E., Allen, N., Kaplan, B., & Yan, F. (2006). Does It Matter if I Take My Writing Test on Computer? An Empirical Study of Mode Effects in NAEP. Journal of Technology, Learning, and Assessment, 5(2).

MacCann, R., Eastment, B., & Pickering, S. (2002). Responding to free response examination questions: computer versus pen and paper. British Journal of Educational Technology, 33(2), 173–188. http://doi.org/10.1111/1467-8535.00251

Mogey N., and G. Sarab. 2006. Essay exams and tablet computer – trying to make the pill more palatable. Paper presented at the 10th CAA Conference, Loughborough, UK. 

Mogey, N., Paterson, J., Burk, J., & Purcell, M. (2010). Typing compared with handwriting for essay examinations at university: letting the students choose. Alt-J, 18(1), 29–47.http://doi.org/10.1080/09687761003657580

Powers, D. E., Fowles, M. E., Farnum, M., & Ramsey, P. (2014). Will They Think Less Of My Handwritten Essay If Others Word Process Theirs? Effects On Essay Scores Of Intermingling Handwritten And Word-processed Essays. ETS Research Report Series, 1992(2), i–15.http://doi.org/10.1002/j.2333-8504.1992.tb01476.x

Puentedura, R. R. (2014). SAMR: A contextualized introduction. Retrieved November.

Russell, M., & Tao, W. (2004). The influence of computer-print on rater scores. Practical Assessment.

Sweedler-Brown, C. O. (1991). Computers and assessment: The effect of typing versus handwriting on the holistic scoring of essays. Research and Teaching in Developmental Education.http://doi.org/10.2307/42801814

Thomas, P., C. Paine, and B. Price. 2003. Student experiences of remote computer based examinations. Paper presented at the the 7th CAA conference, July, Loughborough, UK. 

Wolfe, E. W., Bolton, S., Feltovich, B., & Bangert, A. W. (1996). A Study of Word Processing Experience and its Effects on Student Essay Writing. Journal of Educational Computing Research, 14(3), 269–283. http://doi.org/10.2190/XTDU-J5L2-WTPP-91W2

Wolfe, E. W., Bolton, S., Feltovich, B., & Niday, D. M. (1996). The influence of student experience with word processors on the quality of essays written for a direct writing assessment. Assessing Writing, 3(2), 123–147. http://doi.org/10.1016/S1075-2935(96)90010-0

Link to complete sketchnote graphic