In which subjects are students making the most progress? (2023)

Editor's Note:This work was completed as part of Duolingo.summer internship program.

Everyone learns at their own pace, and Duolingo students are no exception. Some Duolingo users study a few lessons a day on their way to work, before bed, or when they have a free moment. On the other hand, there are users who complete the course in a few days, such asthis motivated Duolingo userwho completed the Esperanto course just six days after its launch! But is there a standard for the amount of progress Duolingo users make based on the course they're taking? If so, which courses do we see users making the most progress on? Can this tell us something about our users? And what other factors contribute to the progress achieved?

about the data

We collected data from new users who created a Duolingo account in the month of February 2016. At that time, Duolingo had released a total of 44 courses, ranging from an Irish course for English speakers to a German course for English speakers. . of Portuguese (now there are 62 and counting). For each course, we analyze the progress users make over the next 90 days. All Duolingo courses are made up of lines of 1-3 skills each (see Figure 1). We measure progress based on the number of lines in the course users have completed. In the analysis, we refer to the rank the user achieved at the end of the 90-day period as their "maximum rank".

Does course duration affect users' progress?

Because each Duolingo course is customized to the language you teach, the length of each course will vary. The number of lines ranges from 29 (English for Spanish speakers) to 60 (Norwegian for English speakers).

Does the number of rows affect the number of users who complete this course in 90 days or less? It is reasonable to think that, given the same amount of time, fewer users would complete a longer course than a shorter one. Indeed, Figure 2 shows that the longer courses (German for English speakers and Norwegian for English speakers) have a very low proportion of users completing the course in the first few months. However, there is great variability and the number of rows is only moderately correlated with the proportion of users completing the course. For example, both the English course for Greek speakers and the Spanish course for English speakers have 31 lines, but the former has the highest proportion of users who have completed the course, while the latter has the lowest.

So what does that mean exactly? One hypothesis is that certain Duolingo users have external factors that motivate them to learn or allow them to progress through courses more quickly. Let's go deeper.

Which courses have users with more prior knowledge?

Intuitively, one of the most important factors affecting course progress may be the amount of prior knowledge in the language being learned. Can our dataset shed light on this intuition? Conveniently, when starting a course on Duolingo, users choose to take a placement test or start all over again.

The results show that, on average, users who completed the placement test finished more than one and a half times more lines in the course than users who did not.

So which users completed the location test? Figure 4 compares the proportion of users who completed the placement test for a given course with the proportion of users who completed that course. Now, taking the placement test doesn't mean you're placed outside the rankings, but as shown in Figure 4, a higher proportion of users who take the placement test is strongly correlated with a higher proportion of users who complete the course. within 90 days.

English skills are valued around the world, so it's no surprise that the courses with the highest proportions to complete the placement test are all English language learning. Some of these courses have more than half of their users completing the placement test! Furthermore, none of the courses with the lowest ratio for completing the placement test are learning English. Interestingly, most of the courses with the highest placement test completion ratios in Figure 4 are in Central/Eastern Europe. This can be attributed to the fact that in that region it is mandatory to learn English in school, and knowledge of English is a common skill needed to pursue higher level careers. Whatever the reason, their dedication to learning English through Duolingo is impressive!

English speakers who learned Irish and Turkish were the least likely to complete the placement test, suggesting that these courses attract first-time students. This is not surprising, as in most countries it is difficult to find classes in any of these languages.

Which courses have the most motivated users?

Globalization has increased the importance of being able to communicate with people outside your native language. So which languages ​​are Duolingo users most dedicated to learning? We compare all courses by looking at the maximum average user rating for each course. Because placement test completion is strongly correlated with tree completion rate, we compared separately for users who completed the placement test versus those who did not.

Figure 5 shows that, regardless of whether or not the placement test was completed, the course with the lowest average maximum number of lines over 90 days was Irish, while the course with the highest average maximum number of lines was German (The course English course for Japanese speakers had a higher average maximum rating than the German course for those who didn't complete the placement test, but more on that below!).

The slow progress of the early grades can be explained by a variety of factors. One hypothesis is that users may have difficulty learning a language with a completely different writing system or grammar (English for Arabic or Hindi speakers comes to mind). Another is that users may simply not be as motivated to progress quickly through a course as may be the case with less popular languages ​​in general like Dutch or Irish. The Spanish course for English speakers is an atypical case worth mentioning:is by far the most common language to learn in the United States, but users show relatively slow progress on average.

As for the best courses, the rapid progress can be due to a greater command of the language at the beginning. This is likely to happen with English courses, as English is often required in schools around the world. High starting proficiency can also be a factor for courses where the language of the course and the user's native language are very similar, such as the French course for Italian speakers. Users' rapid progress in the German course for English speakers is more of a mystery, but therecent course reviewcould have been a contributing factor.

An interesting statistic: even though the English for Japanese Speakers course was still in beta for only part of the period the data was collected, this course still had the highest average maximum rating for users who did not complete the beta test. second highest for those who completed it). These users seem to be very engaged in using Duolingo! Perhaps the upcoming 2020 Summer Olympics, which will be hosted in Japan, will provide a partial explanation for this.

What is the importance of the specific course?

As we've seen, it seems that users who are learning English generally progress faster, but how much more motivated are they?

Let's look at courses in terms of the proportion of users in each line of courses. Figure 6 compares courses where users are learning English (in green) with courses taken by English speakers (in red) and courses where users are neither learning English nor learning English (in blue). We can observe that the courses that teach English have a higher proportion of users in the more distant rows. Courses that don't teach English, on the other hand, still have a lot of users in the front rows.

Users learning English, on average, are not as advanced in the course as those learning a different language. But are there Duolingo-related factors that could be contributing to this difference in motivation?

Of the 44 Duolingo courses available at the end of February, 19 were courses to learn English from another language! At that time, seven of those 19 courses were the only ones offered in that specific base language. On the other hand, if English were your base language, you would have 12 courses to choose from. So, instead of focusing on just one course, English speakers can use Duolingo to learn multiple languages ​​at the same time and therefore not make as much progress in each individual course as users who only have one course to choose from. . We leave this issue for future research.

Ultimately, the amount of progress a user makes in a course seems to be influenced by their amount of prior knowledge and their desire and motivation to learn. In the end though, it's not the speed with which you finish a Duolingo course that matters, but the simple fact that you're making an effort to learn another language! Happy learning!