Online Mentorship: Building a High-Touch Student Experience

We have been trying to reinvent education with each technological breakthrough since the industrial revolution. First it was mail order curriculum, then radio, then TV, and today, the internet. These movements shared a common message, that through this technology we can improve the quality of education while lowering the cost of educating. It was assumed that the primary cost is employing skilled teachers and the barrier to transforming education was the method of content delivery. However, this radical transformation never took place. We still largely teach to the classroom method, with one teacher to many students, with all students moving at the same pace through the course whether they are ready or not.

Technology did not end up being a replacement for teachers. However, we did not understand why human instruction was so much more effective an any technological medium. In the 1980s, educational psychologist Benjamin Bloom investigated the effectiveness of various methods of group instruction and one-on-one tutoring. The discovery is now commonly known as the 2 Sigma Problem: “The average tutored student was above 98% [2 sigma] of the students in the control class.”

Three curves of final exam scores, showing lower levels for conventional classroom, middle achievement for mastery learning, and highest scores for 1-on-1 mentorship.

Figure 1. Graph of Bloom’s 2 Sigma Problem: Students mentored one-on-one perform two standard deviations above students in the conventional classroom model.

So how might today’s technology be leveraged to address Bloom’s 2 Sigma Problem? We are currently seeing three different models emerging on the Internet. By order of increasingly higher touch:

Open Education – Open education technologies supplement, but do not supplant brick-and-mortar models. Information is on-demand, and sometimes even open-sourced or crowd-sourced. They may contain one-to-many instruction or no “instructor” at all. Examples: Khan Academy, Wikipedia, YouTube, Codecademy.

Massive Open Online Courses – MOOCs provide structured, instructor-led programs of one-to-many—often one-to-thousands. Similar to traditional models of education, but content is made much more accessible through online distribution. Examples: Udacity, Coursera, edX.

Remote One-on-one Mentorship – Remote online mentorship programs are the most involved of all the different online models of education, but they can provide the social accountability necessary to stay focused and can be a launchpad for rapid growth.. These programs are created explicitly with the goal of teaching their students skills to be job-ready upon employment. These programs almost exclusively focus on teaching software development and design, two of the biggest markets right now for talent and does not require certification. Examples: Bloc, Thinkful, The Firehose Project.

In our work at Bloc, we are putting Bloom’s studies to the test. Because we believe in the one-on-one mentorship model, we invest just as much time, if not more, on building the platform tools for our mentors as we do for our students. These mentors of our programs are often design and engineering professionals with many years of experience that is valuable to a career-changing student. But they are not trained in education and teaching methods. In order to empower mentors to smoothly guide a student from beginner-to-hire, we rely on a suite of mentorship tools and processes to facilitate guided learning.

The Platform is Strict, the Mentors are Empathetic

Every student comes in with a unique circumstance, and mentors are excellent at adapting to their individual needs. However, these students share a common goal, to get a job in software development and design upon graduation, and they have a limited amount of time in which to do so. In order to guide both the mentors and students along, the platform standardizes much of the process—curriculum, learning objectives, projects, job prep—in order to maximize outcomes based on what we know hiring managers in the industry look for.

One of the fundamental structures to make this work is breaking up each program into a Foundations phase and Projects phase. Every student must complete Foundations curriculum and assignments to ensure they have fundamental knowledge. Beyond that, a mentor then guides the student in 2-3 projects. These projects allow the student to apply their fundamentals to projects and hone the skills that best meets their career goals.

Standardization of the curriculum allows us to pair a mentor in Atlanta with a student in Detroit, two locations with very different opportunities. If mentors were left to their own devices to decide what to teach and how to teach them, mentors could be biased towards their own experiences and not teach toward the skills that the student needs to get a job in their area. A structured curriculum also allows us to rapidly iterate by continuously assessing student progress and competency against the curriculum.

One example of leveraging the data of the student body is measuring their time in the program against their progress in the curriculum. The original graduation requirement was to complete Foundations and four projects to the mentor’s satisfaction. At the fastest 12-week pace, however, this meant that the student had just over two weeks to complete each phase, often sacrificing quantity for quality. We found that spending at least one-third of their time on mastering Foundations produced higher-quality projects, even if there were fewer of them.

We changed the minimum graduation requirements to Foundations and two projects. While mentors challenge students to complete even more projects, with the requirements for graduation loosened, it is even more critical to make sure the student stays on pace and is making progress week-by-week. Rather than putting all of the responsibility on the mentor to keep an eye on the student, we have built accountability directly into the platform with weekly goals.

The Bloc MOOC Guide points out that one of the challenges of online education is “as a participant you need to be able to self-regulate your learning and possibly give yourself a learning goal to achieve.” Weekly goals break a large challenge down into a step-by-step action plan. Weekly goals help students structure their time and give them a clear line-of-sight on how to accomplish their ultimate goal. This weekly goal UI also adapts to whether the student is ahead or behind pace on completing the minimum graduation requirements, suggesting a more aggressive weekly goal if the student is behind to help them increase their velocity until they are caught up.

Screen shot with three panels for next meeting, weekly goal and progress.

Figure 2. Bloc student dashboard and progress bar showing a student that is ahead of pace.

We found that the student’s progress and weekly goal completion also offers a basis for talking points for mentors. If a student is ahead of pace, the mentor may challenge the student to explore other technologies or complete additional projects. If a student is on pace, a mentor can look ahead to the next week’s goals and guide the student on how to get ahead. If a student is falling slightly behind, it opens up a dialogue for the mentor to ask why. And if a student is severely behind, the mentor steps in to offer alternative ways of catching up, such as slowing down the pace or freezing their program for a week.

Encouraging Synchronous and Asynchronous Learning

Online programs provide a combination of synchronous and asynchronous learning. Even if students start and end an online program at the same time, a student can often access the learning materials or watch the lecture videos on their own time.

However, even with the best curriculum, self-studying independently may cause the student to get derailed, get demotivated, or waste too much time on minor issues. It is the mentor’s role to remove these blocks through frequent communication. Regular synchronous communication, such as teleconferencing calls, allow mentors to quickly identify challenges and provide a short feedback loop to student questions.

As a balance to synchronous communication channels, asynchronous communication between scheduled meetings—including messaging and email—encourages students to not always rely on an immediate helper. By struggling on their own, a student experiences “learning how to learn,” oftentimes solving the problem by themselves.

Neither of these have to be built into the platform—after all, it’s not the institution’s goal to reinvent conference calling and email—but integrating these tools into the platform ensures that the appropriate data can be tracked. There are a variety of in-platform communication tools, including discussion, messaging, assignment submission, and product announcements in order to encourage continuous engagement and track in-platform activity. It is rare for a student to not log on every few days, and when the platform notices that they don’t, the mentor is automatically warned and encouraged to reach out proactively.

Scaling the Logistics of One-on-one

Bloom thought that one-on-one mentorship was not scalable, but his original study was constrained by the geographical locations of the instructor and the student. New communication technologies open up the two-sided global marketplace of mentors and students. At Bloc, the challenges of coordinating mentors and students are exacerbated because we offer the flexibility of time and place as well as control over pace. New students are beginning their program each week at one of three paces (slow, moderate, fast). A mentor may have a roster of a dozen students, located in different time zones, each one at a different point in the curriculum, meeting one to three times a week, and on different tracks to graduation.

In order to allow the mentor to focus on mentoring, our goal is to reduce the cognitive overhead of these practical issues by automating the repetitive tasks of scheduling, chasing down data, and prioritizing students. Once past those details, the mentor can focus on guiding the student to the best outcome by providing individualized attention and guidance for each student.

One of the most important tools for the mentor is their dashboard (see Figure 3). The dashboard automates the tracking of student meetings, pace, progress, messages, student assessments, and assignment submissions to encourage productive asynchronous communication between meetings.

Screen shot of Bloc's mentor dashboard, showing an overview of student statuses, assignment submissions, and unread messages.

Figure 3. Bloc mentor dashboard.

Using the data gathered from weekly goal completion, pace, progress, and student engagement, we also help the mentor automatically identify which students may be struggling. We want to empower the mentor to proactively reach out rather than letting the student continue to fly under the radar.

As a student progresses through their program, their mentor is not the only person that they interact with. From enrollment to beyond graduation, a student may also work with advisors, other mentors, or customer support. At each touch point, the staff members need to be acutely tuned into the student’s goals, their struggles, and their accomplishments in order to provide the student with a seamless and consistent experience.

Much of this rich information lies on the student account page, with key moments and interactions stored as a structured note alongside the student’s timeline. By building a narrative around the student’s program experience, it empowers the whole community to rally around helping the student accomplish their goals.

Engaging Mentors in a Human-centered Design Process

Even though mentors are just as much the end-users of the platform as the students, as a designer, they are also employees. This means that the designers can rely on the mentors to provide short feedback loops, advise on projects as stakeholders, and surface recurring student issues.

As mentors experiment with different methods of teaching and helping struggling students, they can use tools like the Bloc API or Tampermonkey to write and run their own scripts that optimize their personal workflows. These tools act as rapid prototypes, and serve as feedback on the mentoring process. The most valuable features are built into the platform for the benefit of the entire community,

Lin, S. (2016). Online Mentorship: Building a High-Touch Student Experience. User Experience Magazine, 16(2).
Retrieved from http://uxpamagazine.org/online-mentorship/