Teaching Programming Fundamentals to Modern University Students

Anton Bogdanovych and Tomas Trescak

School of Computing, Engineering and Mathematics, Western Sydney University, NSW, Australia

Keywords:

Programming Education, Learning by Doing, Clara, Gamiﬁcation.

Abstract:

In this paper we investigate how teaching programming to the modern generation of students, “digital natives”

who grew up with Google and Facebook and do not know the world before the Internet, can be improved

through a highly visual game-like approach. Many programming teachers report that modern programming

students have short attention span, lack concentration and have poor motivation to learn programming. We

show how we were able to improve the motivation of students and their marks by changing the study program

so that the entire entry-level programming course (Programming Fundamentals) is being taught using a visual

set of in-class examples and assignments. The paper presents a set of successful teaching patterns that helped

to convert one of the most hated subjects in our school into a subject that many students loved and were able to

master. The corresponding statistics suggests that one of the key achievements of our approach is a dramatic

change in students’ motivation to learn programming, which has resulted a signiﬁcant improvement in their

overall results and was noticeable in the follow-up subjects.

1 INTRODUCTION

In recent years many programming teachers have no-

ticed that teaching introductory level programming

courses is associated with great difﬁculties for stu-

dents (Bergin and Reilly, 2005). Educators are con-

cerned with high failure and drop-out rates, poor aca-

demic performance and lack of motivation to study

by modern students. Some researchers believe that

the reason for this change in students’ attitude to pro-

gramming is a global phenomenon and is not only

programming related.

In our digital age technology surrounds students

from early childhood; they play video games, watch

movies and use the Internet more frequently than

reading books or going to the library (Prensky, 2007).

As the result, the information becomes “cheap” and

readily available, so educators have to compete for

student attention with too many distractions intro-

duced by technology as well as with other online ma-

terial (Small and Vorgan, 2008).

Indeed, the amount of quality educational con-

tent on the Internet is increasing at an astonishing

rate. Platforms like Coursera

, provide open on-

line access (sometimes for free) to courses from the

world’s best universities. Many universities them-

https://www.coursera.org/

selves e.g. Massachusetts Institute of Technology

and Stanford

choose to upload their course materials

online (including lectures videos, tutorials and solu-

tions to exercises) and allow anyone to access them.

This presents a great opportunity for students and, at

the same time, creates a huge problem of how to man-

age the learning process in this “ocean” of informa-

tion - vast, disorganised, and continually in a state of

ﬂux between updating and stagnating. The problem

is further complicated by the new generation of stu-

dents, who have grown up with Google and Facebook

and who do not know the world before the Internet.

The way these “digital natives” learn and how their

brain processes information is believed to be different

to the “pre-Internet” generations (Vassileva, 2008).

Easily operating with multiple and diverse informa-

tion streams at the same time, “digital natives” have

learned to frequently multitask, but as the result of

this multitasking they are believed to process informa-

tion in a more superﬁcial manner (Vassileva, 2008).

In general, educators observe that contemporary stu-

dents are less efﬁcient in their school work, have

shorter span of attention and tend to struggle with

in-depth analysis (Small and Vorgan, 2008). This is

not a reﬂection of being less capable to learn, but

rather being disengaged with traditional instruction

http://mit.edu

http://stanford.edu

308

Bogdanovych, A. and Trescak, T.

Teaching Programming Fundamentals to Modern University Students.

In Proceedings of the 8th International Conference on Computer Supported Education (CSEDU 2016) - Volume 2, pages 308-317

ISBN: 978-989-758-179-3

(Prensky, 2007). Constant interaction with techno-

logy made modern students comfortable with multi-

ple information streams; they prefer inductive reason-

ing, want frequent and quick interactions with con-

tent, and have good visual literacy skills, but this is

not something that is easy to offer through traditional

instruction (Prensky, 2007).

Despite the changing nature of the students, mod-

ern education often fails to adapt to their new learning

habits and capabilities. In a traditional classroom the

changing demands of digital natives are often ignored

as teachers expect the same levels of focus, attention

and self-motivation as from the pre-Internet students

and are resilient to change their style of delivery and

to engage with new technologies. We believe that this

is one of the key problems that has to be addressed

in order to improve the situation in teaching program-

ming. Our key hypothesis is that modern educators

have to adapt their teaching practices with a particular

focus to improve the motivation of modern students.

We strongly believe that in the world, where infor-

mation is cheap, the role of educational institutions

and teachers has to shift from simply being knowl-

edge providers toward motivators and mentors (Pal-

freyman, 2008).

The advancement of video games and internet

technology shows a huge drift towards entertainment

in modern societies. This growth in the entertainment

industry has certainly affected education as well as

many other aspects of our lives (Pine II and Gilmore,

1998). Students demand engaging learning experi-

ences in traditional learning environments and ex-

isting literature suggests video games as one of the

best mediums to provide such experiences to stu-

dents (Vassileva, 2008). The unique advantage of

these games is the player’s engagement: (Prensky,

2007) demonstrates it as the ability to keep people

in their seats for hours and hours, actively trying to

achieve new goals, determined to overcome their fail-

ures and immersed as part of this interactive entertain-

ment. While digital natives are being criticised for

their lack of concentration and poor motivation, at the

same time they show these very characteristics when

playing video games. In this work we will investigate

the range of factors that make playing video games

so effective and how to integrate these factors into the

classroom when teaching programming fundamentals

to university students.

The remainder of the paper is structured as fol-

lows. In section 2 we analyse the problem and attempt

to understand the reasons causing it. Section 3 out-

lines existing research on teaching programming to

digital natives with a particular focus on using game-

based approaches. Section 4 outlines and provides

motivation for the changes in teaching Programming

Fundamentals in our school. Section 5 presents the

research methodology and other details of the study

that was conducted in our school. Section 6 outlines

the results of this study and Section 7 presents further

discussion of results and concluding remarks.

2 PROBLEM STATEMENT

In order to understand how teaching introductory pro-

gramming courses could be improved, we ﬁrst have to

formulate the key problems that need to be addressed,

understand the reasons for these problems and pro-

duce research hypotheses that can be tested in regards

to successfully developing an appropriate solution.

2.1 Problems

To understand the problems associated with teaching

introductory programming, we have conducted an ex-

tensive analysis of academic feedback in this respect.

At the level of the dean of our school the problem

with programming has been raised as an alarming

phenomenon and the dean has initiated a regular dis-

cussion group among academics involved in teaching

programming, as well as academics whose subjects

require programming as a pre-requisite. Below is the

summary of key problems raised by the academics:

• Lack of engagement and motivation to study pro-

gramming

• Poor academic performance

• High dropout and failure rates

• Poor programming skills after the course

In order to obtain a more complete picture it was also

important to understand the perspective of students.

The student perspective was analysed by a number

of qualitative and quantitative surveys that were con-

ducted by two lecturers in their subjects, among the

students who have completed Programming Funda-

mentals. The analysis of student survey responses

showed that many students simply developed a strong

sense of dislike to programming in general and could

not really explain the rationale behind these strong

emotional reactions. Apart from the emotional com-

ponent, the key concerns of the students can be sum-

marised as follows:

• The subject is too difﬁcult

• Poor engagement in the classroom

• Fighting errors and compiling code

Teaching Programming Fundamentals to Modern University Students

309

• Writing code and sticking to the confusing syntax

of the programming language

• The textbook is hard to understand

• The lectures are hard to understand

• Not clear why learning programming is important

• Boring and difﬁcult assignments

2.2 Reasons

Having the feedback from lecturers and students was

important in understanding the key concerns of both

sides, but even more important was to understand

the deep roots of these problems that could be re-

sponsible for the identiﬁed problems. In an attempt

to understand the reasons behind these problems we

have initiated informal talks with academics and stu-

dents, where the aim of establishing the high-level

reasons for problems with programming was declared

and an attempt to understand these was made through

a brainstorming session in each group (a group of

academics and a group of students). The discussion

with students was the least productive one as students

kept repeating the problems and were unable to of-

fer something constructive beyond blaming program-

ming to be too hard and useless, having bad teachers

and the course being too difﬁcult. Talking with aca-

demics proved to be more productive and the follow-

ing set of reasons was formulated there:

• Digital natives require different instruction prac-

tices

• Low entry requirements for students starting at

university

• Most modern students have to work and have little

time for assignments

The initial analysis of problems in teaching program-

ming and underlying reasons for those showed that

further analysis was needed to understand how to im-

prove the situation. So next we discuss our ﬁndings in

the relevant literature and summarise successful prac-

tices in teaching programming to modern students.

As educators, we realise that from the list of the

identiﬁed reasons for the problems associated with

teaching programming we will not be able to offer a

solution to addressing low entry requirements, which

is the job of university management rather than edu-

cators. We will also not be able to change the working

patterns of the students. Therefore, the key focus of

our further analysis is on improving the instruction

practices, so that they are more suitable for digital na-

tives and improving student motivation is the prime

focus of our work.

3 BACKGROUND

When looking at developing better teaching practices

for educating digital natives some educators claim

that the focus of modern universities has to change

from providing information to providing mentoring

services (Palfreyman, 2008). The times when uni-

versities have actually possessed the knowledge and

could easily restrict the outsiders from accessing it are

long gone. Now most of the necessary information is

freely available and is only one click away, so the role

of modern universities has to change from simply giv-

ing the information to students (hoping that they are

motivated enough to absorb it) to selecting the rele-

vant subset of information and making sure they are

focusing on the right things and are progressing in the

right direction, which essentially is mentoring rather

than lecturing. The lack of such mentoring services is

believed to cause education to become less “higher”

and more “tertiary” as it shifts from a pedagogical

emphasis on liberal education to a skills/curriculum

dominated agenda (Palfreyman, 2008). Such shift in

education, in the authors’ view, fails to properly pre-

pare graduates for rapidly changing economic condi-

tions and related employment opportunities (Palfrey-

man, 2008).

Offering personalised mentoring and tutoring ser-

vices is not something that is only relevant to dig-

ital natives and not something that has to be fully

re-invented for the modern age. Building education

around mentoring was one of the key features of bet-

ter world universities in the late 1800s and is often

named as the key reason for high quality and price

of the degree provided by these institutions (Palfrey-

man, 2008). Each student back then was provided

with a personal tutor whose key role was in mediat-

ing students’ access to knowledge. One of the fun-

damental works on assessing the tutorial system in

(Moore, 1968) claims that the tutor is not teacher in

the usual sense: it is not his job to convey informa-

tion, but the role of the tutor is to act as a constructive

critic and help in sorting out the knowledge of the stu-

dent. The usefulness of inquisitive nature of the tutor

is supported by the French philosopher Peter Abelard,

who said “the key to wisdom is this — constant and

frequent questioning. . . for by doubting we are led to

question and by questioning we arrive at the truth”

(Gaskin, 1998).

While we can rely on the teaching practices of the

past, we should not ignore the new demands of the

modern age. Universities can no longer afford hav-

ing small classes and offering personalised mentoring

services to each of the students. The fact that the vast

majority of the students are digital natives also forces

CSEDU 2016 - 8th International Conference on Computer Supported Education

310

us to integrate corrections in the teaching practices.

Apart from mediating access to knowledge and ques-

tioning, one of the key demands of our times is to help

students with their motivation to learn. There is strong

evidence suggesting that intrinsic motivation to study

programming has a strong correlation with academic

performance (Bergin and Reilly, 2005).

As per ﬁndings from (Prensky, 2007), a good way

of improving motivation of digital natives is to use

highly visual interfaces and employ game-like assign-

ments and lecture examples. Graphical Frameworks

in teaching Programming Fundamentals have become

an effective motivational mechanism (Moskal et al.,

2004). This approach is growing in popularity, espe-

cially in primary school programming classes. Sys-

tems like ALICE (Dann et al., 2011) and Scratch

(Resnick et al., 2009) have been successfully used

in primary and secondary schools, but they are often

abandoned by university educators for their simplic-

ity, speciﬁc appearance and due to the fact that they

teach students framework speciﬁc languages that can-

not be used beyond these tools.

Another framework that is being used at some uni-

versities is Karel (Pattis, 1981). Karel represents a

Java plugin that can be used together with classical

Java editors and, essentially, teaches pure Java rather

than some other framework speciﬁc language. The

Karel approach is to teach programming in a discre-

tised environment, where students control a robot us-

ing a small set of standard commands: move(), turn-

Left(), putBeeper(), removeBeeper(). Figure 1 shows

an example of a typical interface of a Karel assign-

ment. An extensive explanation of the Karel frame-

work is presented in (Pattis, 1981).

In most Karel-based courses students use this in-

terface for a short period of time (e.g. three weeks

like in Stanford) and then move on to pure Java. In

our study we use a similar approach to Karel, but

extend it to the entire course. Karel is rather out-

dated and its black-and-white graphics lack modern

appeal. A more advanced framework is Greenfoot

(Koelling, 2010). We found Greenfoot to be suitable

for our purposes, but we have decided to further en-

hance it with gamiﬁcation and automatic student feed-

back features, so we have developed our own frame-

work called “Clara”. All assignments worked both in

Greenfoot and Clara and students had a choice to se-

lect one of the two frameworks for working on their

assignments.

Improving academic performance through moti-

vation was one of our key goals, but another goal was

to improve the situation with high dropout and failure

rates. These, apparently, are not easy to ﬁx. While

high motivation is certainly connected with dropout

and failure rates, introductory programming requires

dedication and a lot of work and high dropout and

failure rates are a norm. A study in (Bennedsen and

Caspersen, 2007) shows that indeed the failure rates

are signiﬁcant (33% on average from 80 institutions

who participated in the study). The authors also men-

tion that they do not consider 33% being a very high

number as their study has also shown that the number

of people enrolling vs the number of people graduat-

ing from a university is around 27%.

4 SUMMARY OF CHANGES

Based on the analysis of the data obtained from staff

and students, as well as using the results of the liter-

ature review we have decided to introduce a number

of changes in the introductory programming unit in

our school (Programming Fundamentals). In order for

us to be able to measure the results of the introduced

changes we made a decision to maintain a high degree

of similarity in terms of lecture material, the degree of

difﬁculty of assignments and tests as well as the struc-

ture of the new course and the original course. In this

way if there is indeed any difference in the course out-

come it can be attributed to speciﬁc changes we made

rather than the new structure of the course or its dif-

ﬁculty level. Next, we explain the key changes that

were introduced in the new Programming Fundamen-

tals course.

4.1 Highly Visual Programming

Examples

One of the key novelties of our approach was to make

all examples shown in lectures and assignments be-

ing highly visual. We wanted to beneﬁt from the suc-

cess of highly visual systems like ALICE (Dann et al.,

2011) and Scratch (Resnick et al., 2009), but after dis-

cussions with other academics and potential employ-

ers, conversations with students and the analysis of

the job market we have decided to keep a strong focus

on predominantly teaching a popular programming

language and minimise teaching other things that are

not directly related to this language. In case of Scratch

and ALICE, while they offer highly visual and intu-

itive frameworks, a lot of the knowledge gained as

the result is the knowledge of the framework itself,

as well as the knowledge of speciﬁc commands that

are only used in those environments. Furthermore,

our analysis of the teaching resources offered with

Scratch suggests that this framework is more suitable

for primary and secondary school and is not a good ﬁt

for teaching at university. As the result, the approach

Teaching Programming Fundamentals to Modern University Students

311

Figure 1: An example of a Karel exercise. Picture source (Pattis, 1981).

we took is similar with that of Karel the Robot used

at Stanford (Pattis, 1981). The framework we use

(Clara) is similar to Karel, but has a modern graphical

interface that is more appealing and also includes var-

ious gamiﬁcation elements. Unlike most Karel based

courses our entire course was taught using the visual

Clara approach.

Figure 2 shows an example of one of the practi-

cal assignments as well as the interface being used

there. On the right hand side of the ﬁgure there is a

practical assignment showing the main character (la-

dybug Clara) located in a discrete environment. Here

Clara must ﬁnd her way out of a maze. The exit is

marked by a leaf, which after ﬁnding she has to eat.

On the left hand side there is a programming solu-

tion for this problem. The code is written in pure

Java and uses a small set of additional Clara speciﬁc

commands (move(), turnLeft(), turnRight, onLeaf(),

removeLeaf() ).

Using the Clara approach we were able to translate

the entire programming course to be based on prob-

lems with Clara. All programming assignments and

lecture examples are Clara problems.

4.2 Problem-Oriented Lectures

From our previous teaching experience we know that

students respond well to stories that touch upon things

that interest them. So, to better engage with students

and capture their attention we built all lecture mate-

rial around interesting and challenging practical prob-

lems, which can be translated into terms of a lady-

bug manipulating leaves and moving around a dis-

crete environment. Each lecture would start with a

story or a video focused on some interesting internet

phenomenon, a popular hobby topic or just some fun

story from a TV series. After discussing the story fur-

ther lecture material was presented as new knowledge

that could help to solve the problem discussed through

the story in the beginning of the class. Most of such

stories that we used were from the domains of gam-

ing and entertainment. Such problems ranged from

building a robot vacuum cleaner to implementing the

friendship algorithm from the Big Bang Theory

4.3 Learning by Doing and Interactivity

in the Classroom

One of the important elements of the digital natives

is their predisposition to learning by doing (Schank

et al., 1999). To address this need each of the lectures

was structured so that a bigger problem presented

in the beginning of the class was broken down into

smaller sub-problems and these smaller sub-problems

were interactively solved during the lecture using the

new knowledge that was learned in the class. Solving

the sub-problems in the class represented a moderated

process, where the lecturer presented the problem and

asked the students for help with continuing to solve it.

The moderation helped to make sure that the problem

is being solved in a focussed manner. Apart from sim-

ply asking students for help and typing their proposed

commands, we have also experimented with more en-

gaging forms of interaction. One of such forms is to

translate the process of problem solving into interac-

tive theatre. In this interactive theatre the students en-

acted the elements of Clara’s world (e.g. each student

enacting a leaf, a tree or Clara) and other students con-

trolled the changes in the simulated environment by

proposing commands to be enacted.

4.4 Generating Flow and Making

Programming Fun

A good way to motivate students is to generate ﬂow,

which is the state associated with a high degree of ex-

citement. One of the conditions that allows for expe-

riencing the ﬂow state in the classroom is to structure

http://the-big-bang-theory.com

CSEDU 2016 - 8th International Conference on Computer Supported Education

312

Figure 2: Clara’s World: An online assignment in Programming Fundamentals.

the practical assignments so that they incrementally

increase in difﬁculty and the difﬁculty of each new

problem is such that the amount of skills required to

solve it is slightly above the current level of skills of

the person solving it (Csikszentmihalyi, 2002).

Flow is associated and is considered to be much

easier to achieve in video games as games environ-

ments often represent puzzles or problems to solve

and can be easily structured by the aforementioned

principle of incrementally increasing the difﬁculty of

the problem together with the growing level of player

skills. An important work that documents this phe-

nomenon in video games is (Koster, 2013). This work

does not explicitly explore the concept of ﬂow, but in-

vestigates the principles that make video games fun

and enjoyable. One such principle (that can poten-

tially address the issue of high dropout rates) is that

players will quit playing the game if it is too hard to

win or if it is too easy to win. So, the increase in difﬁ-

culty should be carefully tested on an average player

to make the game fun (Koster, 2013).

Both lectures and practical assignments were care-

fully designed to potentially be fun and generate the

ﬂow. To be able to do so, we have modelled an av-

erage student and made practical assignments and ex-

amples being solved during the lectures incrementally

more difﬁcult, so that the skills required for solving

each problems are slightly above the current set of

skills of an average student.

While we had clear cases of reaching the state of

ﬂow by many students that we were able to witness

in the tutorial rooms, this approach is also associated

with the drawback that students who for some rea-

son missed some of the assignments showed much

lower engagement thereafter due to the growing gap

between their skills and skills required to solve the

problem.

4.5 Maintaining Motivation via

Gamiﬁcation and Social Experience

Apart from generating ﬂow, having highly visual as-

signments and facilitating learning by doing (Victor,

2012) presents a number of other recommendations

for building programming frameworks that could im-

prove student motivation. Providing quick (or even

automatic) feedback in case of errors, reﬂecting on the

quality of the student code are some of the factors that

are claimed to be responsible for better understanding

of the programming concepts and improved motiva-

tion to learn programming. Additionally, (Prensky,

2007) states that offering a competition element to

the coding and using gamiﬁcation principles are also

known to be associated with an improved motivation

to study.

Therefore, in our online Clara framework

have incorporated some of these features. The Clara

framework employs game-based learning and gamiﬁ-

cation approaches to motivate students to learn, con-

tinuously improve their solutions and possibly design

their own in their spare time (Figure 2 depicts the on-

line editor). We studied and evaluated effects and

http://pf.scem.uws.edu.au

Teaching Programming Fundamentals to Modern University Students

313

Figure 3: Programming attitude.

signiﬁcance of various gamiﬁcation approaches. To

achieve our goal, we decided to employ star ratings,

badges and challenges with user created content.

To motivate students to submit better code we

used the approach popular in games like Angry

Birds

. Each submitted solution received up to ﬁve

stars for the shortest and most effective code. Five

stars were awarded when Clara used the least amount

of steps to complete the solutions. Completed solu-

tion received hints on how much the solution needs

to be improved to receive a better star rating. Results

were published in leaderboards and often were the tar-

get of heated discussions between students ﬁghting

for the best result. Anecdotal evidence obtained from

the tutors suggests that students who have used the

Clara framework were less eager to share their solu-

tions with their colleagues and they preferred to not

to reveal their solutions but to guide other students to

come up with their own.

Another form of using gamiﬁcation was to award

badges to the best performing students with the most

efﬁcient code, as well as for the ﬁrst ﬁve solutions

turned in the given week. This approach encouraged

students not to procrastinate, and to complete their so-

lutions as fast as possible. Badges could be “stolen”

by creating a better solution than the badge holder. In

this case, badge holder would be notiﬁed to try to get

the badge back by revisiting and improving the solu-

tion, effectively improving student engagement.

Challenges with user generated content aimed at

https://www.angrybirds.com

highly motivated students that completed their assign-

ments fast and wanted to improve their expertise by

creating their own set of problems, publishing them

online and challenging their colleagues. Often, such

students created problems that were then used in the

class to explain a speciﬁc topic, which further encour-

aged students to actively participate in the educational

process.

Our online system also improved communication

between students and tutors, where solutions could

be discussed remotely, not only in class. In this sce-

nario, both student and tutor connect to the same page,

which is automatically updated, allowing for collabo-

rative coding and presentation of results.

5 USER STUDY

In order to ﬁnd a solution for improving the situa-

tion with teaching Programming Fundamentals in our

university we have combined the knowledge obtained

from academics and students together with the ﬁnd-

ings from the literature. As the result, we have for-

mulated the following research hypotheses:

5.1 Research Hypotheses

• H1: Supplying visual and animated game-like ex-

amples throughout the course would help to in-

crease interest in the subject and motivation.

CSEDU 2016 - 8th International Conference on Computer Supported Education

314

• H2: Generating ﬂow is associated with increased

motivation.

• H3: Student academic performance will improve

as the result of increased motivation.

• H4: Adding Gamiﬁcation and social elements

would further increase student engagement and

their academic performance.

5.2 Research Methodology

The introductory programming unit at our university

is composed of the two hour lecture and two hour

laboratory session every week over one semester. In

general, most students in Australia do not study pro-

gramming before entering university. The majority

of students in the Programming Fundamentals course

(58.7% of students in Autumn 2013) had no prior pro-

gramming experience.

Performance in this module is based on continu-

ous practical assessments (30% of the overall mark),

major assignment (10% of the overall mark), mid-

semester test (10% of the overall mark) and the ﬁnal

examination (50% of the overall mark).

A number of studies were carried out in the aca-

demic years 2013-2015. Students enrolled in the ﬁrst

year “Programming Fundamentals” course in our de-

partment voluntarily participated in this study.

The study aimed at testing H1-H3 has been car-

ried out in Autumn 2013. Overall 458 students were

enrolled in this course in Programming Fundamen-

tals in Autumn 2013. Only 92 students have decided

to complete the study survey. This survey was con-

ducted after the ﬁnal lecture, when all the practical

assignments have been completed and only the ﬁnal

exam was pending.

Testing H4 was carried out throughout 2014-2015

with 160 students in total. The students had a choice

to employ either the Greenfoot framework (90 stu-

dents) or our experimental Clara framework (70 stu-

dents) that featured gamiﬁcation and social elements.

The performance of the students was evaluated to see

whether the choice of the Clara framework would

have any impact in regards to H4. An additional qual-

itative survey with overall 90 students has been con-

ducted to understand particular beneﬁts of using the

Clara framework.

6 RESULTS

The outcomes of the study conﬁrm the validity of hy-

potheses H1-H4. The evidence obtained in favour of

H1-H2 comes from a survey that was conducted in

Autumn 2013 where students explicitly report on vi-

sual aspects of the course contributing to their engage-

ment and better understanding of the course material.

Some students explicitly report about them reaching

the state of ﬂow and programming becoming “addic-

tively enjoyable” as the result of this.

6.1 Impact of Visual Delivery and

Assignments Designed to Generate

the Flow

The results of our study show that the attitude toward

programming has dramatically changed after the in-

troduced changes. From the students who were inter-

viewed after completing the previous version of the

course 43.5% have reported that they hate program-

ming and only 13% said that they love it. In an online

survey that was completed by the students enrolled in

the new version of the course (after sitting through all

lectures and ﬁnishing all their assignments) indicates

that only 5.4% of the students claim to “hate program-

ming” and 38% said that they love programming. Fig-

ure 3 shows the comparison between student attitude

toward programming in the previous course (labelled

as “Before changes”) and the course after we’ve in-

troduced the changes to the study program (labelled

as “After changes”).

6.2 Changes in the Academic

Performance

Due to maintaining a similar level of difﬁculty in the

assignments and in the ﬁnal exam across the new

course and the old course we were able to measure

the impact of the introduced changes on the aca-

demic performance of the students (testing the H3).

Ensuring the similarity across practical assignments

was a difﬁcult task, as the nature of assignments has

changed from printing numbers on the screen to mov-

ing a lady bug and eating leaves, but we ensured that

each practical problem in the new course requires a

comparable amount of time for an average student

to solve it and targets the same topics of the course.

Maintaining a similar level of difﬁculty in the exam

was achieved by reusing the same exam questions that

were used in previous years of teaching Programming

Fundamentals. The new exam paper was a compila-

tion of exam questions from multiple past exam pa-

pers, but all questions had to be adopted to Java from

C++. It is important to note that despite teaching the

entire course in the visual framework all exam ques-

tions were pure Java questions and did not feature

Clara. This was done so that we could achieve better

Teaching Programming Fundamentals to Modern University Students

315

Figure 4: Marks comparison.

comparison. Given that in order to pass the subject

students had to pass the ﬁnal exam we can claim that

students graduating from this unit had pure Java skills

and not only Java in the context of Clara.

One of the key achievement that we would like to

report is the difference in the grade distribution that

we connect to the difference in the student motivation.

Figure 4 features a comparison between two versions

of the Programming Fundamentals course with clear

indication of the percentage of students receiving the

corresponding mark in the course “before changes”

and “after changes”. The marks are coded as follows:

“HD” stands for High Distinction, “D” stands for Dis-

tinction, “C” stands for Credit, “P” stands for Pass and

“F” stands for Fail.

Figure 4 shows that in the previous version of the

course the majority of the students (32.2%) received

the pass mark and only a small minority (5.7%) re-

ceived High Distinction and Distinction. In contrast.

after introducing the changes the mark distribution

looks rather ﬂat. Many more people have achieved

higher marks with 16.6% receiving High Distinction

and 15.7% receiving Distinction. The number of stu-

dents receiving Pass and Credit has signiﬁcantly de-

clined. Our explanation of this phenomenon is that

in the updated version of the course the students have

shown an improvement in their level of motivation to

study programming and, hence, they were much more

keen to obtain higher grades.

6.3 Impact of Gamiﬁcation Elements

The impact of the gamiﬁcation elements (H4) intro-

duced in the Clara framework was measured using

quantitative and qualitative approach. The qualita-

tive measure came in the form of surveys, given to ten

groups of students in three semesters. During this sur-

vey, students responded to questions regarding their

experience with both original solution (i.e. Greenfoot)

and the online system. The results show that students

prefer the innovative online solution and ﬁnd the re-

ward system highly motivating, making them revisit

and improve their solutions. This was happening very

sparsely with the original system. Students have par-

ticularly enjoyed the automated feedback in the form

of detecting their errors as they type, helping to iden-

tify inﬁnite loops etc. Students reported that such au-

tomated feedback improved their submitted code.

The quantitative measure evaluates the quality of

the student code. On average, students who used

the gamiﬁed system wrote almost 20% shorter code

and 17% more effective solutions (taking less steps to

complete). Moreover, no online solution has ended in

an inﬁnite loop, while almost 8% of ofﬂine solutions

did. One of the greater successes of the online system

is that 0% of students failed the unit for not achieving

CSEDU 2016 - 8th International Conference on Computer Supported Education

316

required points from practicals. Points received from

practicals were 16% higher than average and comple-

tion of the unit is 15% higher than average.

7 CONCLUSION

We showed how by basing the lectures and all as-

signments of the Programming Fundamentals course

around game-like visual examples structured so that

an average student is likely to encounter the state of

“ﬂow”, we were able to dramatically improve stu-

dents’ motivation to learn and interest in the subject.

To make it a fair comparison the level of difﬁculty

of the practical assignments as well as the difﬁculty

of questions in the ﬁnal exam was made similar to

those that were used in the Programming Fundamen-

tals course prior to introducing the aforementioned

changes. The results show that while maintaining a

comparable level of difﬁculty of the assignments, the

grade distribution has shifted towards High Distinc-

tion and Distinction and the student feedback has be-

come much more positive. Moreover, despite learn-

ing in a highly visual framework and only with visual

animated examples the students were able to success-

fully learn pure Java. The exam questions were delib-

erately designed to include no Clara related questions

and were made comparable with those from previous

years. The students showed good performance on the

exam and experienced no difﬁculty with switching to

console programming in pure Java as reported by the

lecturers from the follow-up courses.

Additionally, we have obtained evidence suggest-

ing that the inclusion of such gamiﬁcation elements

as awarding badges, employing ratings and introduc-

ing challenges helped students to produce better cod-

ing solutions and had a positive impact on their aca-

demic performance (when compared with students

who were not exposed to these features).

REFERENCES

Bennedsen, J. and Caspersen, M. E. (2007). Failure rates in

introductory programming. SIGCSE Bull., 39(2):32–

36.

Bergin, S. and Reilly, R. (2005). The inﬂuence of moti-

vation and comfort-level on learning to program. In

Proceedings of the 17th Annual Workshop of the Psy-

chology of Programming Interest Group, pages 293–

304, University of Sussex, Brighton.

Csikszentmihalyi, M. (2002). Flow: the classic work on

how to achieve happiness. The Random House Group

Ltd, London, UK, 2 edition.

Dann, W. P., Cooper, S., and Pausch, R. (2011). Learning

to Program with Alice (W/ CD ROM). Prentice Hall

Press, Upper Saddle River, NJ, USA, 3rd edition.

Deterding, S., Sicart, M., Nacke, L., O’Hara, K., and Dixon,

D. (2011). Gamiﬁcation. using game-design ele-

ments in non-gaming contexts. In CHI’11 Extended

Abstracts on Human Factors in Computing Systems,

pages 2425–2428. ACM.

Gaskin, R. (1998). The Philosophy of Peter Abelard by

John Marenbon. Cambridge University Press. Philos-

ophy, 73(2):305–324.

Koelling, M. (2010). Introduction to Programming with

Greenfoot: Object-Oriented Programming in Java

with Games and Simulations. Prentice Hall, Univer-

sity of Kent.

Koster, R. (2013). Theory of fun for game design. ” O’Reilly

Media, Inc.”.

Moore, W. G. (1968). The tutorial system and its future, by

Will G. Moore. Pergamon Press Oxford, New York,,

[1st ed.] edition.

Moskal, B., Lurie, D., and Cooper, S. (2004). Evaluat-

ing the effectiveness of a new instructional approach.

In Proceedings of the 35th SIGCSE Technical Sympo-

sium on Computer Science Education, SIGCSE ’04,

pages 75–79, New York, NY, USA. ACM.

Palfreyman, D. (2008). The Oxford tutorial : ’Thanks, you

taught me how to think’, volume 2nd. OxCHEPS, Ox-

ford.

Pattis, R. E. (1981). Karel the Robot: A Gentle Introduction

to the Art of Programming. John Wiley & Sons, Inc.,

New York, NY, USA, 1st edition.

Pine II, B. J. and Gilmore, J. H. (1998). Welcome to the

experience economy. Harvard Business Review, pages

97–105.

Prensky, M. (2007). Digital Game-Based Learning.

Paragon House.

Resnick, M., Maloney, J., Monroy-Hern

andez, A., Rusk,

N., Eastmond, E., Brennan, K., Millner, A., Rosen-

baum, E., Silver, J., Silverman, B., and Kafai, Y.

(2009). Scratch: Programming for all. Commun.

ACM, 52(11):60–67.

Schank, R. C., Berman, T. R., and Macpherson, K. A.

(1999). Learning by doing. Instructional-design the-

ories and models: A new paradigm of instructional

theory, 2:161–181.

Small, G. W. and Vorgan, G. (2008). iBrain : surviving the

technological alteration of the modern mind. Collins

Living, 1 edition.

Vassileva, J. (2008). Toward social learning environments.

IEEE Trans. Learn. Technol., 1(4):199–214.

Victor, B. (2012). Learnable programming : designing a

programming system for understanding programs.

Teaching Programming Fundamentals to Modern University Students

317