Slides in programming lectures:

How they are and how they

should be
Igor Delen
(6828736)

Bachelor’s thesis
Information Sciences

Department of Information and Computing sciences

Utrecht University, Utrecht, The Netherlands
1st July 2022

First examiner:
Hieke Keuning
Second examiner
Christof van Nimwegen
 Abstract
There is considerable research that aims to improve education for university students, with some
focussing on the way presentations are used during lectures. After a literature overview in which
the main criticisms and theories on presentation design and the use of presentations in different
disciplines are discussed, an analysis on slides currently used by teachers is conducted. This
analysis shows that teachers use a high number of words in their slides and relatively few
visuals. All teachers display code in their slides, which they differentiate from the other text by
writing it in a monospace font. To support teachers in using these findings on their
presentations, a prototype tool is developed that allows teachers to upload a slide and receive
automatically generated feedback on their use of words and visuals.

Keywords – PowerPoint presentation, education, code visualisation, lecturing

2
Preface
In my role as a student in the last five years, I attended many lectures. Almost every one of
these lectures was aided by a presentation, mostly made in PowerPoint. More often than not,
these presentations were filled with text or bullet points that the presenter would read aloud.
This made either the lecturer or the presentation redundant in my eyes, as they both conveyed
the same information. I was always glad to attend a lecture where the lecturer used the
presentation as a tool to expand on what they talked about e.g., by using pictures. Since I had
such a preference and interest for this type of presentation, I started to try out ways to improve
my own presentations. Using ideas from presentations I liked and from educational videos on
PowerPoint use, I started to develop an increasing idea of what I considered to be good slide
presentations and less effective slide presentations. This only made me more frustrated when I
noticed lecturers using these less effective presentations styles. Therefore, I wanted to use this
motivation in my thesis. I wanted to check if my ideas about a good slide presentation were
shared by researchers, if the slides that lecturers use were always as bad as I made them out to
be, and I wanted to help teachers to improve their slide by writing a simple tool.

3
1. INTRODUCTION 5

2. LITERATURE STUDY 6

3. ANALYSIS OF SLIDES 13

4. PROTOTYPE 20

5. CONCLUSION AND DISCUSSION 23

6. REFERENCES 25

4
1. Introduction
There is considerable research that aims to improve education for university students, with some
focussing on the way that lectures are presented. Along with active learning, using presentations
during lectures is a well-researched topic. There are two main categories of research done
related to presentations during lectures. The first are articles that compare lectures where
PowerPoint is used to lectures where this is not the case. Nearly every one of those papers
concludes that there is no difference between these two forms of lectures [1, 2, 3, 4].
The other category consists of papers that try to find the best way to design a slide to improve
students’ recall and recognition of the information that is being presented. In this field, there
are two important theories used to analyse the quality of slides: the dual coding theory [5] and
the multimedia theory [6]. These theories both promote the idea that presentations should
contain images and pictures to be presented along the speech of the lecturer and that they should
not contain too much text. This might be easier for presentations in certain disciplines than
others. Studies that looked at the differences in presentation use between disciplines found that
some disciplines are more likely to use text than others [7, 4]. In maths one can display a
formula to explain the solution to a problem but in history one can only explain certain
situations because of their relation to other events, which is more easily displayed in text. For
computer and information science, little research is done specifically in this field, which is why
this thesis focusses on presentation slides that contain lines of code.

This thesis consists of three sections. First, in chapter 2, a literature study is conducted to collect
information on presentations aided by slides in education. It also discusses notable theories for
slide design and common criticisms on PowerPoint, which helps to determine what constitutes
and effective or ineffective slide. The information from this literature study allows for an
analysis and review of slides. Chapter 3 is an analysis of existing slides from teachers at
different institutes using a self-created tool. The slides are mainly analysed on their use of words
and slides, as these are the main indicators of quality according to the literature. This shows
how teachers currently design their slides, which is needed to determine where they can
improve their presentation. Chapter 4 deals with a tool written by the author of this thesis to
help teachers in assessing the quality of their slides. This makes it easier for teachers to use the
information found in the literature and analysis, which could lead to more effective
presentations by teachers. The thesis ends with a summarising conclusion.

5
2. Literature study
To find what research was previously conducted on the topic of PowerPoint and its use in
education, a literature study was conducted to answer the following questions, which are all
answered in their own section:
2.2 What presentation software is most popular?
2.3 How is PowerPoint used in education?
2.4 What is considered a good way to present a lecture?
2.5 What theories on slide design exist and how can they be applied?
2.6 What criticism on PowerPoint presentations exists and what do they state?
2.7 How should code be visualised on a slide?
2.8 How do different disciplines, especially information and computer sciences, make use of
presentations?
2.9 What has not yet been researched?

2.1 Literature search method

Since this research paper focusses on the use of slides in programming lectures, the ACM
library was primarily used to search for relevant papers in this regard, as this library contains
articles specifically in the computer and information science disciplines. The Google Scholar
and Worldcat search engines were used to find papers more generally related to PowerPoint
and also to find if they had any relevant papers on PowerPoint in combination with
programming.
I started with the search term ‘PowerPoint’ to get an idea of how many articles were available
on this topic and how specific they could be. This resulted in some interesting articles in the
ACM library but turned out to be too broad for Google Scholar and Worldcat.
Changing ‘PowerPoint’ to ‘slides’, resulted in almost no relevant articles. However, searching
for ‘lecture slides’ resulted in more relevant results. To find articles that discussed the best use
of slides, I used the search terms ‘lecture slide design’, ‘good lecture slides’ and ‘bad lecture
slides’. The resulting articles from these terms often mentioned the same article to explain their
evaluation or recommendations. Therefore, these articles were also included, as they functioned
as groundwork for other articles on presentation design.
To obtain information on lecturing and presenting during lectures in general, I used the search
terms ‘presenting lectures’ and ’traditional lecturing’. Most articles that contained these terms,
however, were less about how lecturing is mostly done these days and more about how lectures
could be improved by adding more interactive teaching methods.
In order to answer the question “How has PowerPoint changed over time?”, I searched between
1990 and 1995 to find articles written in the very beginning of PowerPoints life. Since this gave
no relevant results, I changed the years to 1990 and 2000.
Articles used to answer the question on code visualisation were found by searching for ‘code
visualisation’ and ‘code readability’.
For the question in section 2.8, the terms ‘Computing OR coding OR programming’ were added
to search for articles specifically about presentations about programming. Since this was a more
focussed topic, I used the databases from multiple conferences specifically on the topic of
computer science education: ITiSCE, SIGSCE, and ICER Unfortunately, there was no article
that focussed on its use in computing science, information science, or in combination with
programming. There were two articles that compared the use of PowerPoint between different
disciplines. One of these articles, by Garret [7], referenced other articles that discussed
differences between disciplines, so I snowballed into articles that seemed relevant to this study.

6
2.2 Presentation software
This thesis focusses on presentations used in traditional lectures, where a teacher presents slides
and talks about a topic. Most often, teachers use Microsoft PowerPoint to create their
presentation, as PowerPoint is by far the most used slideware i.e., software for slide
presentations, with a market share of 95%.1 However, there are other programs available. This
section will give a short overview of some slideware programs.

Microsoft PowerPoint
Microsoft PowerPoint was released in 1990.2 Originally, it was not a success on the market, but
this changed when it was included in the Microsoft Office Suite in 1993, when it quickly
became the dominant slideware on the market.3 It was created to be an upgrade from the old
overhead projector used in most presentations up until then. It was meant for group
presentations to be used in business organisations. However, it is currently used in nearly every
field of work and for any kind of presentation.3

Keynote
Apple’s Keynote was developed for Steve Jobs. He used it to create presentations for the
MacWorld conventions a year before the program was released to the public in 2003.4 The
program was meant to compete with Microsoft’s PowerPoint. It is integrated with other
applications from Apple, such as Pages. Keynote promotes a good compatibility with Microsoft
PowerPoint, giving the user the option to save presentations as PowerPoint files i.e., .pptx files
and the ability to import and edit .pptx files in Keynote.5

Prezi
Prezi was created in 2009. It is a slideware program that enables users to zoom in and out from
their slides dynamically, allowing for a more interactive presentation style.6 In 2019, Prezi
launched Prezi Video, which was developed when Prezi noticed that more and more
presentations were given online or in a hybrid style. This was mostly used by teachers during
the corona pandemic.7 Prezi presents itself as slideware that is scientifically better than
PowerPoint e.g., by quoting an article that concluded that participants evaluated Prezi as more
favourable than PowerPoint [8].8 This research had participants view multiple lectures and
1
. G., C. O., D. D., N. H., M. B. and S. L., “Who are the biggest players in the global presentations industry and
what is their share? (by number of users, revenue, etc),” Wonder, 16 7 2017. [Online]. Available:
https://askwonder.com/research/biggest-players-global-presentations-industry-share-by-number-users-revenue-
etc-ee0fb6843. [Accessed 1 5 2022].
2
P. Sen, “Disruption, innovation, and endurance: A brief history of PowerPoint.,” Hult International Business
School, 8 July 2019. [Online]. Available: https://www.hult.edu/blog/brief-history-of-powerpoint/. [Accessed 1
March 2001].
3
Encyclopedia Brittanica, “Microsoft PowerPoint,” [Online]. Available:
https://www.britannica.com/technology/Microsoft-PowerPoint. [Accessed 3 5 2022]
4
D. H. Steinberg, “Apple on Top of Its Game: the Macworld SF 2003 Report,” MacDevCenter, 1 10 2003.
[Online]. Available: https://web.archive.org/web/20160305151820/http://www.macdevcenter.com/lpt/a/3099/.
[Accessed 8 6 2022].
5
Apple, “Keynote,” [Online]. Available: https://www.apple.com/keynote/. [Accessed 9 6 2022].
6
Prezi, “About,” [Online]. Available:
https://prezi.com/about/?click_source=logged_element&page_location=footer_mobile&element_text=about.
[Accessed 3 5 2022].
7
J. Szafranski, Interviewee, There will be a 'massive shift' in how teaching in the workplace is executed: Prezi
CEO. [Interview]. 8 9 2020.
8
Prezi, “The Science,” [Online]. Available: https://prezi.com/the-
science/?click_source=logged_element&page_location=hamburger_menu_why_prezi_dropdown&element_text
=the_science. [Accessed 3 5 2022].

7
evaluate them through a survey. Prezi presentations were evaluated as organized, engaging,
persuasive and effective.

LaTeX
LaTeX is an open-source program that allows users to create documents through a kind of
programming language. The text is typed without any style, but the style is later determined by
the writer in separate lines of code. It is often used in scientific communities, especially when
they use uncommon symbols regularly.9 For example, mathematicians often use it because
LaTeX makes it easier to include mathematical symbols in their text. It can also be used to
display coding lines as if they were written in a terminal, which is useful for computing sciences
researchers explaining code in a paper for example.

2.3 Introduction of PowerPoint in education

In the mid-nineties, Microsoft saw a new market for their system. Primary and secondary
schools were already looking for ways to take advantage of new technology like computers and
the internet, making them a good possible consumer of PowerPoint.10 Since PowerPoint was
used in educational context more often, researchers took an interest in the possible effect
PowerPoint could have on the students’ exam scores or their motivation. In 1998, Ahmed [1]
was the first to indicate that there was no difference between the test scores following lectures
where PowerPoint was used as opposed to traditional overhead projectors. They state that the
most important aspect in the classroom is still a good teacher. Similar results were found by
Szabo and Hastings [3], who concluded that the difficulty of the lecture, but not the method of
lecturing, contributed to a better or worse grade. Others [3, 4] suggested that the best use of
PowerPoint depends on the context and is not universal. Herting et al. also asked teachers why
they used presentations during their lectures. 67% Of teachers responded that they use slides
mainly to illustrate the contents of the course. Others mentioned that they use them to help
students study for the tests (16%), reflect critically on the content of the course (11.8%), and to
memorize key concepts (6%). Regardless of all the research into the best use of PowerPoint in
the classroom, a large study from 2018 by Baker et al. [2], in which they review literature from
the previous twenty years on PowerPopint use in education, found that PowerPoint had no effect
on students’ learning compared to lectures where PowerPoint was not used. They suggest that
research should focus more on how PowerPoint is used instead of comparing the differences
between a lecture with and without PowerPoint. Even though multiple scientific articles stated
that there is no extra benefit to using PowerPoint in lectures [1, 2], 69 percent of teachers who
used Microsoft software used PowerPoint in their classroom in 2001 [9].

2.4 Presenting lectures

The traditional form of lecturing, in which a teacher stands in front of a class and provides
information through speech and a slide presentation, is still one of the most common forms of
lecturing [10]. It is likely that this will remain the case, since traditional lecturing takes a
relatively low amount of preparation time and effort compared to active lecturing for example,
where students are participating and interacting during a lecture [11]. Besides, the preparation
time is the same for large and small groups of students.
Regardless of the popularity of traditional lecturing, research has shown that active lecturing
helps students to retain information better and that they are more effective in conveying
information. These lectures are also often evaluated by students as more effective and more

9
CTAN, “What are Tex and friends?,” [Online]. Available: https://www.ctan.org/tex. [Accessed 3 5 2022].
10
L. Guernsey, "PowerPoint Invades the Classroom," The New York Times, 31 May 2001.

8
 motivating [11, 12, 13, 10]. These articles also mention ways that teachers can make their
classes more interactive, without changing their entire teaching method. In traditional lectures
for large classes, it is difficult to engage every single student through discussion, as there is no
time to let everyone speak. Instead, teachers can use quizzes to engage their students. Popular
tools to create these quizzes are KaHoot and MentiMeter, for example.
Interestingly, Covill [14] suggests that students do not evaluate traditional lectures as
uninteresting. According to her findings, students report being involved in the learning process
and mention that they are engaged in independent thinking. However, the research was
conducted in classes of 30 students, which is relatively small compared to lecture class sizes.

2.5 Theories on slide design

Most modern articles about slide design mainly reference two theories. The dual coding theory
as explained by Clark and Paivio [15, 5] and Mayer’s multimedia theory [16, 6]. Dual coding
theory explains how people retain information better if that information is fed to them through
verbal input as well as non-verbal input [15], as both types of information are handled
differently, and this makes it easier to form connections between the two types. Multimedia
theory is made up of multiple principles that mostly focus on reducing the cognitive load of the
audience. The application of these principles manifests itself in more visuals and less text on a
slide.
Blokzijl and Andeweg [17], as well as Tangen et al. [18] found that using images on slides
increased students’ interest, as it is more visually appealing. Increased interest could be one of
the reasons that students learn better. Pate and Savannah [19], and Issa et al. [20] also found
that students may retain items better when the lectures are presented with slides using
multimedia design, as they performed better on exams than students who had lectures with
traditional slide design. Fig. 1 shows an example of a slide as used in a lecture and an example
of how that slide could be improved by replacing the text with an image. The teacher is
discussing a model, of which Data understanding is the second part he explains. The image in
the new slide shows the relevant context for data understanding in the model that the lecturer is
talking about, allowing the student to receive the information through speech and image. The
slides used in this section are all taken from the data collected for this thesis. The alternative
slides are handmade by the author of this thesis.

Fig. 1: The slide on the left is a slide as used during a lecture. The slide on the right is a possible alternative made using dual coding theory and the
modality principle.

Another important part of multimedia theory is cognitive load: Cognitive load is the amount of
information someone can handle at a given time [6]. According to the verbal redundancy
principle in multimedia theory, presenting textual information while at the same time explaining
something through speech increases cognitive load, which impairs learning. This is supported
by findings in the neurolinguistic field. Horvath states that reading the same information that is
spoken has a negative effect on learning [21]. This is because people read the text on a slide

9
 using the silent reading voice in their head. This information is processed in the same way as
aural speech, leading to two streams of information that must be processed in the same way.
Experiments have shown that this leads to participants remembering only one of the information
streams when paying attention to one of them and remembering very few of both information
streams when they try to pay attention to both. This is known as the cocktail party effect [22].
Fig. 2 shows an example of how a slide can be improved to reduce this effect.
Imperatief programmeren Rekenen Input en output Strings Commentaar Foutmeldingen Imperatief programmeren Rekenen Input en output Strings Commentaar Foutmeldingen

Commentaar in code Commentaar in code

Code bevat commentaar Code bevat commentaar

Wordt genegeerd door Python Wordt genegeerd door Python
Bedoeld voor mensen die de code lezen Bedoeld voor mensen die de code lezen
Andere mensen Andere mensen
Jijzelf op een later moment Jijzelf op een later moment
Commentaar schrijven is een van de belangrijkste Commentaar schrijven is een van de belangrijkste
onderdelen van programmeren onderdelen van programmeren
Guido van Rossum: Guido van Rossum:
Code is read much more often than it is written Code is read much more often than it is written

Fig. 2: The slide on the left is a slide as used during a lecture. The slide on the right is a possible alternative made while keeping the verbal
redundancy principle in mind.

37 / 44 37 / 44

However, not all researchers agree that using the suggestions in Fig. 1 and 2 will improve a
presentation. There is some disagreement on the use of text in a slide. Whereas multimedia
theory and dual coding theory suggest using no text at all or only a couple of words, it has been
shown that short written phrases that the lecturer uses to make full sentences from has a positive
effect on students learning [16]. How the lecturer uses these phrases to form their speech has
been researched by Hallewel and Crook [23]. Although they cannot say what the best way is,
as they did not research what the effect was of different styles, they propose that lecturers should
be consistent in either strictly following the bullet points on the slide or using the bullet points
freely to build their speech around them. These bullet points should not contain copious
amounts of text. Fig. 3 shows an example of how a slide can be improved using these findings.

Fig. 3: The slide on the left is a slide as used during a lecture. The slide on the right is a possible alternative made using the findings
from Hallewell & Crook [23] and Mayer & Johnson [16].

10
 2.6 Common criticism on PowerPoint slides
Not everyone thinks PowerPoint is a good tool to use for presentations. The most notable
example is the booklet “The Cognitive Style of PowerPoint” by Edward Tufte [24]. Tufte said
that PowerPoint should not be used; especially not for scientific analyses, as PowerPoint’s
inherent style would ‘dumb down’ the content for the audience too much. PowerPoint was made
for the presenters, and not for the audience. It offers an auto content wizard and encourages
bullet points to organise information on a slide, but the person who would benefit the most from
these features is the presenter, who could look at the slides to remember what to say. For the
audience, the keywords at bullet points were not helpful, according to Tufte. Tufte was also
very critical of business templates; pre-made templates that contain the company’s logo and
had presenters follow a very strict layout. According to him, this was strictly unnecessary and
could even damage the quality of the presentation. However, his booklet contained relatively
few scientific references and his findings were not based on experiments.
Another notable example of critical papers on PowerPoint is an article by Angela Garber from
2001.11 According to her, presenters should avoid what she called “death by PowerPoint”. It
refers to the “blue screen of death” after a Windows crash. In the earlier templates proposed by
PowerPoint, a blue background with yellow or white words was very common, resulting in a
“blue screen of death” for PowerPoint. According to Garber, “Death by PowerPoint” is caused
by the following issues, among others:
• Data Dumping: When someone puts too much information in a slide. This can cause
information overload in audience members.
• The slide slave: When someone keeps looking at the slides they are presenting. This changes
the focus of the audience from the presenter to the slides.

2.7 Visualising code

Since this thesis focusses on presentations of courses in which programming is taught, slides
often contain lines of code. The way that these lines of code are displayed can add to or decrease
the quality of the presentation. Having an effective presentation therefore means that this code
should be presented in such a way that it is readable and understandable. For example, using
colours to highlight certain operators in the code improves the readability and understandability
[25]. Displaying the code using a monospaced font helps with the alignment of the code, which
also improves readability [25]. A monospaced font is a font in which all characters take up the
same amount of space. Below are the same lines of code first without any special layout,
followed by a monospaced font and finally in a monospaced font and with syntax highlight.
The code lines are taken from one of the collected slides, where it was displayed with
monospacing and syntax highlight.
Lines of code are also more readable if they are shorter and if they contain a small number of
identifiers. These two variables are most determinative when looking at code readability [26].

def raadgetal (): def raadgetal (): def raadgetal():

oplossing = "5" oplossing = "5" oplossing = "5"
while True: while True: while True:
getal = input("getal: ") getal = input("getal: ")
getal = input("getal: ")
if getal == oplossing: if getal == oplossing:
if getal == oplossing: return return
return

11
A. Garber, “Death by Powerpoint,” Small Business Computing, 22 June 2021. [Online]. Available:
https://www.smallbusinesscomputing.com/software/death-by-powerpoint/. [Accessed 9 March 2022].

11
2.8 Using PowerPoint in different disciplines
As stated before, there are multiple studies that agree that the best use of PowerPoint is not
universal and depends on context, like disciplines. Very few studies have investigated the use
of PowerPoint across multiple disciplines. This is relevant however, as this type of research
could show what the differences between presentation use are. The two main works on this
topic are by Garret [7] and Herting et al. [4].
Garret analysed PowerPoint slides that came with textbooks from thirty different disciplines he
used three major categories of descriptive variables:
- Textual variables. This variable represents the use of words in the presentation and per slide.
Garret wrote a tool to calculate the sum of all words in a presentation, the average number of
words in a slide, and the average number of words in a title.
- Readability. This variable was used to determine the reading level based on their Flesch-
Kincaid reading grade [27]. The reading grade is influenced by the number of words in the
number of sentences and the number of syllables in the number of words.
- Graphic/visual count. This variable represents the use graphics of any kind in a presentation.
It includes coloured text boxes, tables, graphs, and images. The graphics were grouped by their
size in bytes.
Garret found that there are some differences between so called “hard” and “soft” disciplines, as
defined by Biglan [28]. Hard disciplines include, among others, math, physics, and computing
science. Soft disciplines include, among others, psychology, history, and philosophy. Hard
disciplines use more words, but the words used are less complex. The difference in word use is
suggested to be due to the nature of the disciplines. Some disciplines, like math, can convey
their key points in short phrases or formulas, whereas a discipline like history must convey its
key points through written out relations between two events. Hard disciplines also use more
graphics, but these are smaller in file size than those of soft disciplines. These graphics are more
often charts or graphs, whereas the graphics used in soft disciplines more often consist of
images.
Garret also found that students from soft disciplines report PowerPoint as more effective than
students in hard disciplines, by linking his findings to those of James et al. [29]. They showed
that students from courses such as accounting and IT perceived the use of PowerPoint as
ineffective, whereas students from courses such as management or education perceived the use
of PowerPoint as effective.

Herting et al. did not analyse the slides of presentations, but asked teacher about their preferred
use of slides through a questionnaire. They did this for twelve courses of the natural, medical,
and social sciences. They found that the natural and medical sciences had more slides with
visuals, and that the social sciences had more slides with text [4]. This is in line with Garret’s
findings that hard disciplines use more graphics.

2.9 Research gap

Multiple articles state that using PowerPoint during a lecture is just as effective as a lecture
without PowerPoint [2, 3, 4]. Because of this, they propose that other researchers look deeper
into how PowerPoint is used instead of only if it is used or not. However, there are very few
studies that do this. For computing science or information science courses, I was not able to
find any article that had exclusively studied the use of PowerPoint or other slideware. This was
only done in studies that evaluated multiple disciplines. Therefore, the following section will
discuss the analysis of presentations in computing science and information science courses.

12
3. Analysis of slides
This chapter contains an analysis of slides that are or have been used by teacher in programming
lectures. This is done to answer the question; “How do teachers make use of slideware in
programming lectures?”.
In 3.1 Material, the source of the slides is discussed. In 3.2 Method, the method and the
motivation for analysing certain variables are mentioned. In 3.3 Results, the results of the
analysis are shown and discussed with examples. In 3.4 Discussion and conclusions, the results
are discussed and connected to the literature discussed in chapter 2.

3.1 Material
To obtain realistic examples of slides used in lectures, I reached out to several teachers at
different universities. Each of these teachers taught a course in which programming was
explained. The institutes are anonymised through a code to ensure anonymity of the teachers of
each course. I also used some freely available slides from courses at MIT. Table 1 shows a
description of each course from which I used slides in this analysis.

Table 1: A short description of each course of which slides were analysed.

Course Description Institute File format

Code
A A course about information retrieval and how C1 LaTeX
to use Python.
B A course about object-oriented analysis and C1 PowerPoint
design in the context of building software.
C A course about computational thinking, data C1 Libre
related problems and data analysis. Office
D A course about functional programming. LaTeX
E A course about the role computation can play MIT .pdf
in solving problems.12
F A course that continues the content of E.13 MIT .pdf
G A course about MATLAB.14 MIT .pdf
H A course about programming in C# C2 PowerPoint

Since, regardless of the file format, some aspects still had to be analysed manually, only three
presentations per course for the manual analysis were used. This would give a good overview
of the way the teacher used the slides during the lectures, without costing too much time to
analyse everything. I used a random number generator to choose the three presentations I would
use. I also excluded the first and last presentation of a course, as the first presentation is usually
meant to inform the students about the structure of the course and possible assignments instead
of course information, and because the last presentation is often used to summarize the
information taught in the course.

12
A. Bell, E. Grimson and J. Guttag, “INTRODUCTION TO COMPUTER SCIENCE AND PROGRAMMING IN
PYTHON,” 2016. [Online]. Available: https://ocwnext-rc.odl.mit.edu/courses/6-0001-introduction-to-computer-
science-and-programming-in-python-fall-2016/. [Accessed 1 4 2022].
13
E. Grimson, J. Guttag and A. Bell, “INTRODUCTION TO COMPUTATIONAL THINKING AND DATA SCIENCE,”
2016. [Online]. Available: https://ocwnext-rc.odl.mit.edu/courses/6-0002-introduction-to-computational-
thinking-and-data-science-fall-2016/. [Accessed 1 4 2022]
14
O. Celiker, “INTRODUCTION TO MATLAB,” 1 2019. [Online]. Available: https://ocwnext-
rc.odl.mit.edu/courses/6-057-introduction-to-matlab-january-iap-2019/. [Accessed 1 4 2022].

13
3.2 Method
First, I coded each presentation based on the course abbreviation and their chronological
occurrence in the course. Some courses already had a shortened course code from their institute,
while some courses did not, these courses were assigned abbreviations.
As not all slides were made in PowerPoint, I either had to export them to .pptx files or analyse
them manually. Analysing slides manually meant that I would not be able to count, among
others, the number of words in the presentation, as that would take too much time. The
preference, therefore, was to export as many presentations as possible to a .pptx file.
The slides made in Libre Office were easily exported to .pptx. The slides made with LaTeX,
unfortunately, could not be exported to .pptx. However, the .pdf files generated from the LaTeX
files were changed to .pptx by using the website ilovepdf.com.15 After I checked to see if the
files were transported correctly, I included them in the analysis. This was also done for other
slides in the .pdf format.

In order to automate some of the analysis process, I wrote a Python tool using the pyhton-pptx
0.6.21 library.16 This tool scans each individual slide in a presentation and calculates the
following information:
- Average number of words per slide
As discussed in chapter 2.4, including too much information in a slide can cause
information overload in audience members. Therefore, the average amount of words
used in a slide are an indication of the quality of that slide. A slide with more words in
generally worse than a slide with fewer words.
- Percentage of slides that contain at least one visual
- Average number of visuals per slide that contain at least one visual
as discussed in chapter 2.4, the use of pictures and images in slides can increase
students’ attention and help them retain more information. Since pictures are the most
noticeable change when applying multimedia design principles, they provide a good
indication of the quality of a presentation. Presentations that contain numerous pictures
are better than presentations that contain fewer pictures.

The export from .pdf files to PowerPoint files did not work as intended. Everything in the .pdf
files got marked as an object by the tool, making it impossible to distinguish between graphs,
images, and textboxes. Therefore, the number of visuals per slide was counted and calculated
by hand for these courses.

Some information could not be analysed by the tool, so this was calculated manually:
- The percentage of slides in the presentation that included code lines
- How the code was presented
e.g., as a screenshot of a terminal or as written text.
- The number of links and their function
e.g., showing a video, getting students to fill in a question or a reference to a used paper.
- Use of animations
This was only done for PowerPoint presentations, as animations do not show in a .pdf
file.
- Other notable use of functions or slide designs

15
https://www.ilovepdf.com
16
S. Canny, “python-pptx,” 2012. [Online]. Available: https://python-pptx.readthedocs.io/en/latest/#.
[Accessed 18 5 2022].

14
The findings on the use of words, visuals and the percentage of slides using code are used for a
quantitative analysis. Findings on other variables are used for a qualitative analysis. The
findings on code visualisations are discussed separately.

3.3 Results
Data
For this analysis, 24 presentations were analysed. Together, these presentations contained 1031
slides.

Quantitative
Table 2 shows the average number of words per slide for the Table 2: the average number of words
per slide for each course.
presentations in each course. The courses from MIT: D, E, and
F, have a similar average of words per presentation. Fig. 4 Course Words
displays the average number of words per slide in a boxplot. It A 64
shows that the highest average of words is 70 words per slide and B 20
that more than half of all slides contain more than 50 words per C 36
slide on average. D 57
Fig. 5 shows a boxplot of the percentage of slides that contain E 56
visuals or code. The same information is shown in Fig. 6, where F 59
the percentages are shown per course. It seems that on average, G 53
more slides contain code than visuals, but that this differs H 24
strongly per course. The low number of slides with code for Total 46
course C can be explained, because this course uses Jupiter
Notebook during the lectures. Nearly all code examples are displayed there. It is also visible
that most courses either show a lot of code or a lot of visuals in their presentations. Course H is
the only exception, with numerous slides containing code as well as numerous slides containing
visuals.

90%

80%

70%

60%

50%

40%

30%

20%

10%

0%

Code Visuals

Fig. 4: A boxplot of the average Fig. 5: Boxplots of the percentages of slides that
number of words in all presentations. contain code and visuals.

15
80%
70%
60%
50%
40%
30%
20%
10%
0%
A B C D E f G H

Percentage of slides that contain code Percentage of slides with at least 1 visual
Fig. 6: A bar graph of the number of slides that contain at least 1 visual and the number of slides that display code.

Visualising code in slides

Table 3: Analysis of code display per course.

Course Coding Monospace Syntax Screenshots

Language font highlight
A Python X X
B C# X X
C Python X
D Haskell X X
E Python X
F Python X
G Matlab X X
H C# X X X

Table 3 shows the results for the way code is displayed. Only course B displays the code in a
separate space from the other text in a slide. All other courses display their code in the same
space as they would otherwise put text. Sometimes, text and code are even displayed alternately.
The columns for monospace fonts and syntax highlights only apply to the written code on the
slides. Course C does not have any written code in the presentations. They use Jupiter Notebook
to shows the majority of code examples. The code in screenshots from course C and course G
is displayed in a monospace font and with syntax highlights.
Course G sometimes uses screenshots, but mostly uses written code in the slides. These lines
are written in blue, but the code on the screenshots is not.
Examples of code in a monospace font and code with syntax highlights can be found in section
2.6.

Qualitative analysis
Hyperlinks are not often used in any presentation. Most courses have very few hyperlinks to no
hyperlinks at all. They are most often used to refer to sources for images or information, as can
be seen in Fig.7, or for a reference to the Creative Commons rules. These links are usually
displayed in a small font. This makes them unreadable from a distance. They are likely meant
for students who look at the slides afterwards.

16
 Only course A and B use hyperlinks to interactive question websites like WooClap or
MentiMeter, as shown in Fig. 8. This is sometimes replaced by only the code that students need
to fill in.

Figure 8: Part of a slide in which a link is used to display Figure 8: A slide with a link and a QR-code to a Wooclap
the source for information in this particular slide. quiz.
The blue boxes cover up the full code and part of the QR-
code, so that they are not accessible through this thesis.

Animations are used more often, but only to display something sequentially within a single
slide e.g., to show how a diagram develops, or by displaying a circle over an image to signal an
important part. These always appear without any special transition. Transitions between slides
are never used. This can only be said for certain for the courses that created slides in
PowerPoint. However, in .pdf files, it is noticeable that several slides are the same, except for
a new line of text. It is assumed that this would have been done with an animation if the slides
were made in PowerPoint.

Course D makes use of a template. All slides display the university logo and show a long,
greyed out picture on the left, as shown in Fig. 9. The blue box in the bottom-left corner covers
up the instance’s logo.

Fig. 9: A slide from course D, using a template. Fig. 10: A slide from course B, where code is displayed in text boxes.

Course B uses text boxes to display code examples, as can be seen in Fig. 10. This is an

17
application of the signalling principle of multimedia theory. It signals to the listeners where
code will appear, which lowers the cognitive load during the presentation.

3.4 Discussion and conclusions

When comparing the results of the analysis to the findings of Garret [7], the most interesting
difference is found in the number of words per slide. Garret’s analysis showed that slides from
computer science contained 36.8 words on average. The analysis in this study showed that this
was nearly 10 words per slide higher at 46, which is also higher than any other discipline in
Garret’s research. This could be due to the difference in data set. Garret analysed slides that
were supplied by the same company that publishes the textbook on which the slides are based.
It could be that slides created by teachers are more likely to contain more words.

The way that code was displayed was relatively consistent. All courses used a monospace font
for their code lines and most courses used syntax highlights. It would be better if all courses
used syntax highlight, but monospaced code does differentiate the code enough from the text
that it is easily recognisable as code. Syntax highlight would mostly help to increase the
readability of the code segments.
Interestingly, there seems to be a negative association between the number of slides that display
code and the number of slides that display visuals. Due to the relatively low amount of data in
this thesis, this could not be confirmed by a statistical analysis. This is something that should
be investigated in further research.

According to the theories discussed in chapter 2, course B and H are considered the best
presentations, as they all use visuals in over 50% of their slides. They also use the lowest
average number of words per slide. Course C is only slightly worse, but the other courses have
a relatively high number of words per slide and few slides that contain visuals. For example, in
the presentations of course A, only 3 visuals were found.

The main improvement point for teachers is deleting words from their slides. Five out of eight
presentations had an average of more than fifty words per slide. This means that there are
numerous slides that contain more than fifty words, which is too much. Teachers could start by
shortening full sentences to just keywords. Ideally, teachers would replace sentences of text by
images or illustrations, which would help students to process the information according to dual
coding theory and multimedia theory [5, 6].
Fig. 11 shows three slides taken from the data set which are used as an example of ways to
improve slides. 11a Shows a common slide from the data set. Although this is a common slide,
it is one of the worst slide designs, as it contains a lot of words and no visuals at all. 11b shows
an improvement on this slide. The long sentences have been replaced by key words and there
are visuals for all key words. 11c shows the best slide of the three. There is no text for the
speaker, but one visual that allows the speaker to explain the topic through example.

18
 11a 11b

11c

Fig. 11 a, b, and c: Three slides from the dataset.

To answer the research question stated at the beginning of chapter 3: teachers who teach
programming courses make use of slideware very differently. Some use many words in their
slides, others don’t. There does not seem to be a standard way that presentations are created.
However, on average, teachers use too many words in their slides and only sparingly make use
of visuals. By removing some words and adding more visuals to their presentations, the quality
of the presentations could be improved.

19
4. Prototype
To support teachers in improving their slides according to the results from the literature search
and the analysis, I created a small prototype tool in Python. This tool scans a PowerPoint file
and gives feedback with possible improvements that can be made based on the number of words
and visuals in the presentation. Counting the number and visuals of words by hand takes quite
some time for most presentations, which is why the tool will be of help to teachers. The tool is
available on GitHub17.
This chapter discusses the requirements and creation of the tool, how a user would use the tool,
and possible ways to improve the prototype. An example run of the program is shown to
illustrate how it works.

4.1 Requirements
The tool has the following requirements:
1. It should be able to receive any .pptx file as input. The library that I used is only able to
analyse pptx files, so slides made in a different format are not usable by this tool.
2. It should be able to analyse and give feedback based on the number of words in each slide.
3. It should be able to analyse and give feedback based on the number of visuals in each slide.
4. It should be able to give an example of a more effective slide i.e., a slide that contains few
words and at least one picture.

4.2 Creating the tool

Since the tool focuses on two points of feedback, I created two functions to analyse the
PowerPoint slide. One for the number of words in each slide and one for the number of visuals
in each slide. Fig. 12 shows the function that calculates the number of words in each slide and
puts them in a dictionary. The function goes over each slide and checks if they contain a text
frame. If there is a text frame, the tool reads the text in the text frame by reading the run objects
that contain information about the text. The runs are then changed to words and counted. For
each slide, the tool creates an entry in the dictionary wordcount_per_slide with the slide number
and the number of words in that slide.

Fig.12: The function that analyses the number of words per slide.

17
https://github.com/Ig0rium/OZP-final-tool.git

20
The function for the number of visuals in each slide is very similar, but it checks if the shapes
in each slide are of the image type.
Two separate functions then scan the dictionaries and provide feedback statements in the form
of a list of numbers that correspond to the slide numbers.
Another function uses the two generated dictionaries and uses them to look for slides that
contain few words and at least one picture and many words and no pictures. This function also
prints two feedback statements.

4.3 Using the tool

The tool should be opened in a terminal, as it has no interface. To let a PowerPoint file be
analysed, the user must input the file location in the variable pr1. The tool can then be executed
and will give the feedback as output. The feedback consists of:
- A list of slides that contain more than fifteen words. Although scientists agree that a presenter
should not use too many words on a single slide, there is no agreed maximum. The number
fifteen was chosen as it allows for a title and a couple of bullet points, but not too many. If the
number of words is above forty, the tool gives the user an extra message to urge them to take
out some text.
- A list of slides that contain no pictures. Although it is understandable if not every slide has a
picture e.g., the first slide with the title of a lecture, it is useful to see how many slides contain
no pictures.
- A list of more effective slides. These contain a maximum of fifteen words and at least one
visual.

It is then up to the user to make use of this feedback and edit their presentation as they see fit.

4.4 Example run

When analysing one of the presentations from course C, the tool gives the following output:

The following slides contain no visuals. Consider adding some

visuals to these slides: 1, 6, 7, 9, 10, 16, 18, 71, 77, 78,
79

The following slides contain more than 15 and less than 40

words. Check to see if all text is necessary: 2, 6, 9, 18, 21,
23, 25, 26, 38, 39, 40, 41, 43, 44, 46, 47, 48, 50, 52, 53,
54, 57, 58, 59, 60, 64, 65, 67, 68, 69, 70, 76, 78, 79

The following slides contain more than 40 words. Strongly

consider deleting some text in these slides: 7, 8, 16, 17, 22,
24, 27, 29, 30, 31, 32, 33, 34, 35, 36, 42, 49, 55, 71

The following slides contain less than 15 words and at least 1

visual, which means they are likely more effective than other
slides: 4, 5, 11, 12, 14, 20, 28, 37, 45, 51, 56, 61, 62, 63,
66, 72, 75

Fig. 13 shows slide 16, which is contains more than 40 words.

Fig. 14 shows slide 5, which is considered a more effective slide by the tool, as it contains no
words and at least one visual.

21
 Based on the differences that a user sees in these slides, they can change the slides. For example,
a possible improvement for slide 16 could be to show an example of an abstraction instead of
the text or to only display the text in bold.

Fig. 13: Slide 16. Fig. 14: Slide 5.

Fig. 15 shows slide 6, which contains between 15 and 40 words. The tool suggests checking if
the text in the slide is necessary. The slide contains a definition of a term that is important to
the course. It might therefore be useful to leave the text on the slide. Fig. 16 shows slide 3,
which does not appear in any list in the output. This shows that the tool is not perfect, and that
the user should always check the content of the slides themselves and consider what changes,
if any, should be made.

Fig. 15: Slide 6. Fig. 16: Slide 3.

4.5 Envisioned tool

The tool is meant to show what is possible by analysing presentations. However, if the tool was
to be developed further, the following functions should be included:
1. It can analyse more file types than .pptx files.
2. It has an interface in which the user can select the slides from their files manager.
3. It can analyse the file on more than the number of words and visualisations. For example, by
putting a language analysis tool within the tool, the Kincaid reading grade of text could be
calculated. Feedback on this could be given in accordance with the relevant findings of Garret
[7], as mentioned in section 2.7.
4. It can give feedback on the way code is displayed in the slides.
5. The tool can recognise if a business or organisation template is used in the slides.
6. The feedback that is given is based on literature in such a way that the user is most likely to
reflect on their slides.
7. The output is not given as a list of slide numbers, but as images of the relevant slides.

22
5. Conclusion and discussion
The aim of this thesis consisted of three parts:
- To look into research done into using slideware in lectures in general and into courses that
teach programming to students. This was done to find out what constitutes as an effective or
ineffective presentation.
- To analyse existing slides from courses where programming was being taught. This showed
how presentations were currently used.
- To write a tool that could analyse a PowerPoint presentation and give feedback on it. This tool
is meant to help teachers to use the information found in this thesis.

Current literature showed that the two main theories referenced in relation to presentation
design were the dual coding theory by Paivio [5] and the multimedia theory by Mayer [6]. Dual
coding theory explains how people retain information better if that information is fed to them
through verbal input as well as non-verbal input. Multimedia theory consists of multiple
principles that focus on reducing the cognitive load of the audience. The application of these
principles manifests itself in more visuals and less text on a slide. However, there is also
research that disagrees with this. Tufte [24], for example believed that presentations did not
usually contain enough information and that more specific text and clearer graphs should be
included in presentations. Others state that text can also be useful for providing structure to a
presentation [16, 23]. Studies that looked at the differences in presentation use between
disciplines found that some disciplines are more likely to use text then others [7, 4]. In maths
one can display a formula to explain the solution to a problem but in history one can only
explain certain situations because of their relation to other events, which is more easily
displayed in text. For computer and information science, little research is done specifically in
this field, which is why this thesis focusses on presentation slides that contain lines of code.
The analysis of the slides showed that there are many teachers who mainly use text in their
slides, even more so than other research showed. Most presentations would benefit from
reducing the number of words in each slide, but also from adding more visuals to the slides. A
positive outcome of the analysis was that almost all presentations had good visualisations of
code, using monospaced fonts and syntax highlights.
The tool written is able to analyse a PowerPoint presentation on the use of words and visuals.
It gives feedback on slides that contain either too many words or no visuals at all. It is also able
to give a list of effective slides. The tool helps to make the information found in this thesis more
available to and easily applicable by teachers. By using the tool, they can analyse their own
slides and use the feedback to improve their presentations. This can lead to improved lectures
and therefore improved learning and even better grades for students. Although the tool does not
give feedback related to coding, the thesis includes an analysis of code visualisation in slides
that can be especially useful for presentations from information and computer science.

The research in this thesis could be improved upon. Because of limited time available, the
dataset that was used was relatively small and therefore not very representative. The number of
courses that was analysed could be larger and more diverse e.g., by including courses from
more institutes or education levels. Also, instead of just three, all slide presentations belonging
to a course should be analysed, so possible differences between courses or instances can be
stated with more certainty. Both these improvements could allow future researchers to make
statistically significant statements regarding the use of presentations. Another way to improve
the quality of the analysis would be to analyse everything by hand instead of through a tool, as
the tool does not always label everything correctly. Having multiple coders also helps to
improve the reliability of coding visuals on slides. Lastly, this paper only focusses on the use
of words, visuals, and code visualisation. More variables could have an impact on the quality

23
of a presentation, like the relation between the visuals and the topic of the slide or the
complexity of code examples.

Future research on this topic is still needed, as there are almost no studies done that evaluate
the use of slide presentations specifically in the domain of computer or information science.
This could help to improve the quality of the presentations and thereby the quality of the lectures
given at universities for example. Another useful topic for future research is the difference in
presentation use between disciplines, as researchers agree that there is no single best way to
make use of slideware. There are a couple of studies that have investigated this, but more
information on this topic could be very useful for other studies. It could give more insight into
which disciplines make similar use of slideware and therefore which disciplines should or
should not be compared to each other. As stated before, the tool created for this thesis is only a
prototype. Future research could look at further development, implementation, and effect of
this tool in different areas of education e.g., primary, secondary, and further education.

24
6. References
[1] C. Ahmed, “Powerpoint versus traditional overheads. Which is more effective for
learning?,” Conference of the South Dakota Associotion for Health, Physical Education
and Recreation., 1998.
[2] J. P. Baker, A. K. Goodboy, N. D. Bowman and A. A. & Wright, “Does teaching with
PowerPoint increase students’ learning? A meta-analysis.,” Computers & Education,
vol. 126, p. 376–387, 2018.
[3] A. Szabo and N. & Hastings, “Using IT in the undergraduate classroom: should we
replace the blackboard with PowerPoint?,” Computers & Education, vol. 35, no. 3, pp.
175-187, 2000.
[4] D. Chávez Herting, R. Cladellas Pros and A. & Castelló Tarrida, “Patterns of PowerPoint
Use in Higher Education: a Comparison between the Natural, Medical, and Social
Sciences,” Innovative Higher Education,, vol. 45, no. 1, pp. 65-80, 2019.
[5] A. Paivio, in Mental representations: A dual coding approach, New york, NY, Oxford
University Press, 1990.
[6] R. E. Mayer, in Multimedia learning, Cambridge, Cambridge Univeristy Press, 2001.
[7] N. Garret, “How do academic disciplines use PowerPoint?,” Innovative Higher
Education, vol. 41, pp. 365-380, 2016.
[8] S. Moulton, S. Turkay and S. Kosslyn, “Does a presentation’s medium affect its
message? PowerPoint, Prezi, and oral presentations,” PLoS ONE, vol. 12, no. 7, 2017.
[9] L. Guernsey, "PowerPoint Invades the Classroom," The New York Times, 31 May 2001.
[10] M. LoPresto, “A New Comparison Of Active Learning Strategies To Traditional Lectures
For Teaching College Astronomy,” Journal of Astronomy & Earth sciences education,
vol. 3, no. 1, pp. 59-76, 2016.
[11] N. A. Alaagib, O. A. Musa and A. M. Saeed, “Comparison of the effectiveness of lectures
based on problems and traditional lectures in physiology teaching in Sudan,” BMC
Medical Education, vol. 19, no. 1, p. 365, 2019.
[12] A. Revell and E. Wainwright, “What Makes Lectures ‘Unmissable’? Insights into
Teaching Excellence and Active Learning,” Journal of Geography in Higher Education,
vol. 33, no. 2, pp. 209-223, 2009.
[13] v. A. Dijk, v. d. G. Berg and v. H. Keulen, “Interactive lectures in engineering education,”
European Journal of engineering education, vol. 26, no. 1, pp. 15-28, 2001.
[14] A. Covill, “COLLEGE STUDENTS' PERCEPTIONS OF THE TRADITIONAL LECTURE
METHOD.,” College Student Journal, vol. 45, no. 1, pp. 92-101, 2011.
[15] J. Clark and A. Paivio, “Dual coding theroy and education,” Educational psychology
review, vol. 3, no. 3, pp. 149-210, 1991.
[16] R. E. Mayer and C. I. & Johnson, “ Revising the redundancy principle in multimedia
learning,” Journal of Educational Psychology, no. 100, p. 380–386, 2008.
[17] W. Blokzijl and B. Andeweg, “The effects of text slide format and presentational quality
on learning in college lectures,” IPCC 2005. Proceedings. International Professional
Communication Conference, pp. 288-299, 2005.

25
[18] J. Tangen, M. Constable, T. Durrrant, C. Teeter, B. Beston and J. Kim, “The role of
interest and images in slideware presentations,” Computers and education, vol. 56, no.
3, pp. 865-872, 2011.
[19] A. Pate and A. Posey, “Effects of applying multimedia design principles in PowerPoint
lecture redesign,” Currents in Pharmacy teaching and Learning, vol. 8, no. 2, pp. 235-
239, 2016.
[20] N. Issa, M. Schuller, S. Sentacaterina, M. Shapiro, E. Wang, R. Mayer and D. DaRosa,
“Applying multimedia design principles enhances learning in medical education,”
Medical education, vol. 45, no. 8, pp. 818-826, 2011.
[21] J. C. Horvath, “The neuroscience of PowerPoint,” Mind, brain and education, vol. 8, no.
3, pp. 137-143, 2014.
[22] E. C. Cherry, “Some Experiments on the Recognition of Speech, with One and with Two
Ears,” The Journal of the Acoustical Society of America, vol. 25, no. 5, 1953.
[23] M. J. Hallewell and C. Crook, “Performing PowerPoint lectures: examining the extent of
slide-text integration into lecturers’ spoken expositions,” Journal of further and higher
education, vol. 44, no. 4, pp. 467-482, 2020.
[24] E. R. Tufte, “The Cognitive Style of PowerPoint,” Conn: Graphics press, Cheshire, 2003.
[25] R. Roels, P. Mestereaga and B. Signer, “Towards Enhanced Presentation-based
Teaching of Programming - An Interactive Source Code Visualisation Approach,”
Proceedings of the 7th International Conference on Computer Supported Education -
Volume 1: CSEDU, vol. 1, pp. 98-107, 2015.
[26] R. Buse and W. Reimer, “Learning a Metric for Code Readability,” IEEE TRANSACTIONS
ON SOFTWARE ENGINEERING, vol. 36, no. 4, pp. 546-558, 2010.
[27] J. P. Kincaid, R. P. Fishburne, R. L. Rogers and B. S. & Chissom, “Derivation of new
readability formulas (automated readability index, fog count, and flesch reading ease
formula) for Navy enlisted personnel,” Naval Air Station Research Branch Report, vol. 8,
1975.
[28] A. Biglan, “Relationships between subject matter characteristics and the structure and
output of university departments,” Journal of Applied Psychology, vol. 57, pp. 204-213,
1973.
[29] K. E. James, L. A. Burke and H. M. & Hutchins, “Powerful or pointless? Faculty versus
student perceptions of PowerPoint use in business education.,” Business
Communication Quarterly, no. 69, p. 374– 396, 2006.

26