Is Computer Science More or Less Suitable for Distance Learning?

In this blog, I discuss how undergraduate computer science students perceive the discipline of computer science by analyzing their distance learning experience during the 2020 Spring semester – the Corona Semester. This perception is derived based on an analysis of students' comparison of distance learning of computer science with a) other sciences and engineering and b) social sciences and humanities.

Data was collected during the 2020 Spring semester at the Rachel and Selim Benin School of Computer Science and Engineering of the Hebrew University of Jerusalem (where I was on sabbatical) using two questionnaires. The purpose of both questionnaires was to give the school management feedback on their students' experiences (including learning processes and habits, feelings, needs, and concerns) and, based on the lessons learned from the data analysis, to design the upcoming semesters, regardless of whether or not on-campus learning will be permitted. The feedback received in the first questionnaire, right after the onset of the semester, after the students had experienced distance learning for only three weeks, also enabled the school management to apply the lessons learned from the data analysis to the continuation of the semester. In the second questionnaire, the students were asked specifically to propose how they prefer teaching to be organized during the 2020-2021 academic year if the pandemic continues and on-campus learning is limited (e.g., in small groups) or is not possible at all.

As it turns out, these questionnaires guided the students to reflect on and analyze their learning processes and behaviors before and during this semester, habits of mind they are not usually encouraged to apply during regular semesters. In fact, the analysis of the data gathered by these questionnaires enabled the school management to understand the students' perspective, not only on distance learning of computer science specifically, but also in a broader sense, how students perceive the essence of computer science and computer science learning.

Specifically, in this blog, I describe students' perceptions of computer science by analyzing their comparison of the suitability of computer science for distance learning with the suitability of other domains, specifically (a) natural sciences and engineering and (b) social sciences and humanities for distance learning.

Additional details about the questionnaires can be found in my previous blog on Determining CS Student Preferences During the Corona Semester posted on November 2, 2020.

Data Collection

Approximately 1,800 undergraduate students are enrolled in a variety of study tracks offered by the School of Computer Science and Engineering. The first questionnaire was distributed on April 21, 2020, during the third week of the semester and was answered by 493 students (27%); the second questionnaire was distributed on June 14, 2020, about three weeks before the end of the semester and was answered by 290 students (16%). Responders were distributed as follows (to avoid excessive data, averages for the responders from both questionnaires are presented): 41% of the responding students were freshmen (of which about 3% began their academic studies in the Corona semester), 28% were sophomores, 23% were juniors, and the rest, 8%, were seniors; 60% were men and 40% were women (compared with an overall 70%-30% gender ratio at the school). One-third of the students who answered the questionnaire had previous experience with distance learning. Both questionnaires were in Hebrew and were distributed by email.

The second questionnaire included the following two questions: "To what extent, in your opinion, is distance learning suitable for computer science, in comparison with natural sciences and other engineering domains?" and "To what extent, in your opinion, is distance learning suitable for computer science in comparison to humanities and social sciences?" In both questions, the students were asked to rate their opinion on the following scale: Less suitable, Equally suitable, More suitable, and I don't know. We note that the computer science students study on the same campus of the Hebrew University of Jerusalem as the natural sciences (chemistry, physics, biology) students, while the humanities and social sciences study on a different campus of the university. Table 1 shows the distribution of students' answers to these questions.

 

Table 1: Students' comparison of the suitability of computer science for distance learning relative to (a) natural sciences and engineering domains and (b) humanities and social sciences

Distance learning suits computer science	Compared with: Natural sciences and other engineering domains (286 responses)	Compared with: Humanities and social sciences (282 responses)
More	96 (33.6%)	106 (37.6%)
Equally	104 (36.4%)	39 (13.8%)
Less	46 (16.1%)	76 (27%)
I don't know	40 (14%)	61 (21.6%)

 

As can be seen, students' opinions are divided both with respect to natural sciences and other engineering domains and with respect to humanities and social sciences. However, their opinions are divided differently in the two cases. Specifically, with respect to:

• Natural sciences and other engineering domains: About one-third of the students (33.6%) believe that distance learning suits computer science more, but at the same time, about the same proportion – one-third (36.4%) – believe that distance learning suits computer science to the same degree that it suits natural sciences and other engineering domains.
• Humanities and social sciences: Although about one-third of the students (37.6%) believe that distance learning suits computer science more, at the same time, about one-fourth (27%) of them believe that distance learning suits computer science less than it suits humanities and social sciences.

In addition to the above closed questions, the students were asked to explain their opinion in open questions. As described below, the data analysis of the answers to the open questions reflect students' perception of computer science and how this perception shapes their opinion with respect to the suitability of computer science for the distance-learning mode.

Data Analysis

A. Computer science is less suitable for distance learning

On the one hand, in both cases – natural sciences and other engineering domains, and humanities and social sciences – the students explained that distance learning is less suitable for computer science because computer science courses are more difficult. With respect to humanities and social sciences, one of the students explained: "Computer science is more difficult than humanities and therefore is more challenging in distance learning." With respect to natural sciences and other engineering domains, the students explained this opinion by distinguishing between the mathematics-oriented course and the computer-oriented course. One of them explained: "Although we are dealing with computers, half of the degree is mathematical and studying mathematics this way is much less successful, in my opinion." Another student wrote that "it is harder to teach mathematics (formulas etc.) by remote teaching."

Similar explanations emerged from the data analysis of the first questionnaire, in which students were asked to answer the following open question: "In your opinion, does distance learning have any unique characteristics in the case of computer science?" Their answers further support the above analysis of the students' perceptions of the suitability of computer science for distance learning, by comparing distance learning of mathematical courses ("theoretical") with that of programming courses. We illustrate this by presenting several quotes that reflect how students perceive these differences:

• Mathematical courses are much more difficult through distance learning than courses that are based more on speech or programming.
• Yes, it is for the better. Programming courses integrate beautifully into distance learning because it [distance learning] reflects the independent learning and the independent work aspects of programming. Conversely, mathematical courses are more complex to perform through distance learning.
• It depends on the course. In math and theory courses it is very very challenging and difficult; in programming courses, it can be an opportunity to use additional tools for learning.

B.    Computer science is more suitable for distance learning

In each of the above cases – natural sciences and other engineering domains, and humanities and social sciences – students gave different explanations why distance learning is more suitable for computer science. One student wrote: "I think that in computer science, the option of distance learning is ideal. If we compare [computer science] with other sciences, biology for example, it is clear that there are frameworks that require that students arrive [on campus], such as labs. In social sciences, learning through discussions is very important (but less important than in biology)."

In addition, students attributed the suitability of distance learning for computer science to the tremendous amounts of material available online, to the fact that the teaching staff have more technological skills, and to the fact that "most of the computer science degree (unfortunately) consists of self-learning. Many hours of code programming and solving exercises. In class we learn very little, and we learn most of the material at home." In a similar spirit, another student wrote: "The work of computer science exercises in many cases continues till the middle of the night and because [distance] learning takes place at home, we don't have to worry about getting home."

As before, similar explanations emerged from the data analysis of the first questionnaire, in which students were asked to answer the following open question, as mentioned above: "In your opinion, does distance learning have any unique characteristics in the case of computer science?" Here are several illustrative quotes that further support students' perceptions that distance learning is more suitable for computer science:

• Yes, in most of the computer science courses I attended so far, most of the learning was autonomous any way. In distance learning, it is easier to do that.
• Yes, in my opinion, it is a domain that can be relatively learned well from the distance since a lot of material is available on the Internet, and sometimes the ability to adjust my study time to when I am more concentrated helps me understand the material.
• The lecturers and the TAs have less problems with the technology.
• Yes, you can write code and try different things during the lecture and so the material is grasped better. Furthermore, the recordings are given in comfortable times and suit people with prior knowledge (you can skip parts that you know/advance faster).

Summary

As can be seen, on the one hand, students perceive computer science as a difficult topic for learning and therefore as less suitable for distance learning compared with other subjects, while on the other hand, since computer science requires less face-to-face interaction (either in terms of discussions or physical facilities) as well as many self-learning hours, it is more suitable for distance learning.

Students are not, however, giving up on on-campus learning. As I showed in my previous blog, Determining CS Student Preferences During the Corona Semester (posted on November 2, 2020), the Corona Semester sharpened students’ understanding of what is worth coming to the campus: learning-based social interactions that take place either in formal on-campus settings (class rooms, etc.) or in informal on-campus settings (in the learning spaces).

Orit Hazzan is a professor at the Technion’s Department of Education in Science and Technology. Her research focuses on computer science, software engineering, and data science education. For additional details, see https://orithazzan.net.technion.ac.il/ .