Poll context
The second batch (2018) of the Academy started with 49 accepted students. About 60% of these students completed all the requirements and graduated. But what happened with the remainder 40%? This corresponds to 20 students that dropped out or failed to graduate.
The pie chart bellow summarises how far did these 20 students actually go in the Academy. The chart groups the students by the percentage of work units they successfully completed. Each BLU, each SLU and the Capstone was considered as a single unit (32 units in total, 19 SLUs, 12 BLUs and 1 Capstone). So, for example, 65% of the people who didn’t graduate (13 students) did less than 50% of the expected workload.
The following charts give more insight about which work units were left incomplete. A great majority of 75% (15 students) didn’t finish half of the SLUs. Comparatively, the BLUs were more frequently finished. Only one person submitted and successfully completed the Capstone.
A poll was sent to all these students to figure out what caused them to fall behind. We got 9 (45%) responses and some clues on how to prevent this on next editions. Here are the results.
Poll results
1. How much of the academy did they believe having completed?
| Percentage | No. of Students |
|---|---|
| Less than 10% | 1 |
| 10% - 30% | 2 |
| 30% - 50% | 1 |
| 50% - 70% | 2 |
| 70% - 90% | 2 |
| More than 90% | 1 |
2. What was their goal when they enrolled on the LDSA?
| Goals | No. of Students |
|---|---|
| Curiosity, to be up-to-date with data science trend | 5 |
| Academic reasons | 3 |
| Career change | 3 |
| Career progression | 3 |
| Networking | 2 |
3. What problems did they face?
| Problems | No. of Students | |||
|---|---|---|---|---|
| Issues not related with the academy (unexpected changes at work or university) | 9 (100%) | |||
| The Bootcamp modules were taught too fast, not enough time to keep up | 3 | |||
| Not enough python programming experience | 2 | |||
| Not enough personalised feedback from the instructors | 2 | |||
| Not enough time to finish the SLUs from the Bootcamp and falling behind | 2 | |||
| The hackathons were too challenging | 1 | |||
| The programme contained too much material for the available time | 1 | Not enough assistance from the instructors | 1 | |
| Although having completed most of the academy, students didn’t feel obliged to finish because there was no actual advantage or reward | 1 | |||
| The assignments were too time consuming | 1 | |||
| Not happy with the quality of resources made available to students | 1 | |||
| Difficulty attending hackathons due to distance | 1 |
Preventive actions
Let’s summarise this feedback in some main areas and think about preventive actions:
| Area | Suggestions |
|---|---|
| External factors | We can’t directly do anything about this, but we can use the buddy system to know when this happens and encourage them to stay in the academy. |
| Python programming | Hopefully this is fixed with the Python Prep Course. |
| Bootcamp and SLUs | The bootcamp is very intense and for some students is difficult to grasp so much new information in such a short time. Since there is a gap of about a month between closing the list of accepted students and the beginning of the Academy, there could be a recommended reading list (or videos, etc.) to help them prepare the bootcamp. If possible, it would be nice to have more time to do the SLUs during the bootcamp. If not, maybe an extra day of instructor’s support (in person/online) could be scheduled before the first hackathon. |
| Instructors | In order to increase personalised feedback, within the buddy system, besides ensuring motivation and that everything’s OK, the instructors could also keep track of their students background and performance (e.g. notebook grades and hackathon participation) and give some specific advice/compliments/suggestions periodically. |
| Hackathons | After each hackathon, it would be nice to open the leaderboard to all students individually and encourage them to compete until the following week. Then, maybe the work with best scores could be made available to the rest of the group. Or maybe an advanced solution could be given by the instructors. |
| Rewards | Having an official certificate publishable on LinkedIn would be good to increase commitment. Maybe with a distinction for the best students. |
| Time | Make sure students are aware of the necessary time commitment to be successful in the academy. Encourage them to create a study schedule early on and to finish their exercise notebooks on time. |