The Reading Brain in the Digital Age: The Science of Paper versus Screens

This article looked at the differences between learning by reading text on a piece of paper versus on a screen. Before 1992, there were a lot of studies that concluded that people read slower, less accurately and less comprehensively on screens than on paper. However, now many studies have found few significant differences in reading speed and comprehension.

Such studies show that screens drain more of our mental resources and make it harder to remember what we read when we are finished reading. Therefore, we comprehend less when we read on screens since it’s just more physically and mentally taxing.Paper allows reader to focus on a single page of paper without losing sight of the text, which makes it easy to form a coherent mental map of text that is not doable when reading on a screen. The limitation of navigating a text while reading on screen can impair comprehension. In Norway, a study of 72 10th students, who were given a 1500 word text to read on paper and computer, and later took a reading comprehension test showed that computer users performed a little worse than student who read on paper.

Students who read pdf files on the computer have been shown to have a more difficult time to reference text, and such reading on paper is better suited to absorb text as screens can interfere with user’s sense of control. Additionally, studies have shown that those who really wish to go into a reading, will read it on paper as oppose to on screen because it helps them understand the text with more clarity.

Furthermore, the students’ approach to reading on computers is a state of  mind that is much less conducive to learning. Kate Garland did a study where she made half her students study economics material from a computer and the other from a booklet. After 20 minutes of reading, the students from both groups scored equally well. Reading on different mediums also distinguish between remembering and knowing something. Remembering is a waker form of memory that is more likely to fade unless it is converted to long-term memory. Garlands’ students who read off a computer relied more on remembering than on knowing. The students who read on paper learned the material more thoroughly and quickly.

Other studies show that people who read off a computer screen scored lower on the tests and also reported higher levels of stress and tiredness than people who completed it on paper. While they scored relatively equal on the tests, they did not do as well on the attention and working memory part of the test. Those students who read off a paper scored higher, because researchers concluded that they had a more studious state of mind, and therefor effectively directed their attention and working memory. Further, people who read on screens take a lot of shortcuts, spend more time of browsing and scanning, as oppose to people who just read a paper once.

Nonetheless, in recent surveys, it shows that most people prefer paper, but their attitudes are changing as tablets and e-books are becoming more common, so more studies must be done.

Student Assessment: Measuring Progress Toward Your Goals

Click to access IPD_Ch2_2011.pdf

This article is a guideline for creating assessments and gives examples of good and bad questions to include when creating assessments. There are three types of assessments: diagnostic, formative, and summative. Additionally there are two types of assessment questions: objective and non-objective. Objective questions are those which are not generally open to interpretation. True/false, fill-in, matching/sequencing, multiple choice are examples of objective questions; short answer and essays are examples of non-objective questions.

A valid assessment must allow student to show their actual understanding of the content. Some suggestions to consider when creating questions include:

-Make sure that items don’t give away the answers
-Focus on the item that you’re testing
-Ensure that the incorrect answer choices are not too extreme in that someone who lacks knowledge would think that they are plausible
-Distractors are used to see whether students truly understand the objective or not, and ensures that they did not just guess or eliminate the obvious incorrect choices
-Avoid asking questions in the negative
-Be conscious of potential bias in questions
-Ensure that answers are definitely true or definitely false

Introducing computer science to K-12 through a summer computing workshop for teachers

This paper described a one-week summer computer workshop that is very similar to what we are working on for the Alice project. Again, the researchers noticed that the demand for CS and math will grow more and more, but at the same time there is a bigger decline of interest among students. Therefore, the goal of the workshop was to provide K-12 students with a better education in the field of CS by teaching teachers and educating them on ways to improve CS education.

Scratch and Alice were used for this workshop, and in the end teachers created curriculum materials to use in their classrooms. Previous related work include lecturing and curriculum development on CS education, using Python to emphasize the connections between natural science and math, and efforts to increase the number of students in AP CS. In this research paper, they discussed how their work built upon the successful strategies for teacher workshop from related work. One concern they had was much of the previous work was done with high school CS teachers, but this one was targeted for K-12 students to expose students at an early age.

The workshop provided each teacher with a flash drive that had all the workshop materials, which were created by two undergraduate students. There were sessions on Scratch, gathering user input, Boolean expressions, loops, variables and arrays. Additionally there was a session dedicated to curriculum development so that teacher could take back materials for to their classroom. Most of the participants were teachers from around the area – a district with large population of minority and economically disadvantaged students.

The teachers highly praised the amount of support they received and thought that it was well organized and easy to understand. However, many found Alice to be user unfriendly in comparison with Scratch, and wished that they could work more on Scratch. The workshop was assessed by an outside program evaluator, and the average pretest score was 19/50, and the average posttest score was 42/50, a 121% increase in confidence level.

Diversifying High School Students’ Views About Computing with Electronic Textiles

Sherry Turkle and Seymour Papert once published a paper describing the lack of diversity in approaches to teaching to computing. Further, they worked to promote epistemological pluralism. This standpoint is a discipline that has the views and attitudes about the discipline as well as the conceptualization of nature of knowledge. Thus, in this paper, the researchers worked on observing and expanding students’ views and expectations and attitudes towards cS through designing electronic textiles.

In previous research, people have found that students valued concrete approaches more than abstract. They describe CS as boring, tedious and often not a create field. In this study, researchers worked with 27 high schoolers (many who had spent a semester learning Alice), and worked on expanding their views on computing and broadening their participation in CS through a 10-week e-textiles workshop.

This e-textiles approach introduced computing as designing and programming artifacts. e-textiles combined engineering and computing, often seen as masculine concepts, and crafting, often seen as feminine. Additionally, there was a tangible dimension that made students ultimately see computing as accessible, transparent, and creative.

The main focus of this student was to address relationships between activities used to teach CS and students’ perception of computation, as well as engaging students in a variety of approaches to expand on their knowledge base. They also conducted pre- and post-interviews: the pre-interviews involved questions about willingness to participate and post-interviews included questions about experience with computing, career aspiration, and what they learned.

In the pre-interviews, many students’ understanding of CS was very narrow and limited. Their words were very vague and also has difficulty connecting their programming activities to career-related work. In the post-interviews, there was a much wider range of perspective of computing hands-on. e-textiles highlighted positionally and the students’ appreciated some creativity. Some of the students’ responses included how the project widened possibility of what CS can do, and how it used code to tell a story. Many were excited to share their final project.s

However, the positive result could also be due to the environment of the classroom and the flexibility to create whatever for the project.

Predicting Student Success using Fine Grain Clicker Data

This paper describes a 12-week study at a large university, in which they examined students’ clicker questions performance affects the outcome at the end of a term for an Introduction to Computer Science class in Python. Currently, the first CS course has very extreme grades – high failure and drop rates, and therefore, in most cases, the first exposure student have to CS is most often their last.

Students were given clicker questions in which they answered individually, with their peers, and then as a class, a teaching method known as peer instruction. Additionally, they were given isomorphic questions that were answered after each question individually. For students who were absent, the credit was given for students by dividing the number questions correct by the total number of questions answered.

The study concluded that the first few weeks of a CS course has a strong influence on a students’ end of course performance. Such students who cannot the explain coding, could not code later on the final exam. Those who performed in the top quartile throughout the semester did better on the final exam. Further, these questions best predict the outcome of multiple choice questions, code writing, and the final exam. The isomorphic questions ended up being the top predictors of student outcomes.

Effective and Ineffective Software Testing Behaviors by Novice Programmers

This paper evaluates software testing behaviors in introductory CS classes and looked at the quality of testing at different stages. The process included teaching students, writing software tests, and using a grading system to collect the students’ work. Students submitted their work online and received feedback. Afterwards, they have the option to resubmit their work.

Previously, many students struggled to manage their time wisely, and would often procrastinate. Further, many didn’t anticipate how much time programming takes. This new approach uses a “test-driven development,” which encourages students to test smaller units of code before completing, and testing in increments. This development uses higher-order thinking and builds more confidence in the students.

The students’ work are graded on the quality of code, correctness, the amount of time it took, and the number of lines in the submission. Their initial and final submissions are compared. Additionally, they compared the final correctness score of good to poor solutions, of high and low performing students, and the time of each “development milestone.”

They found that there were strong positive correlations between early test quality and the quantity of other assignments. There was a higher coverage in early development, which meant that there was high quality code and students also completed their work earlier. Early testing and positive outcomes are independent of time management and effect of abilities, however.

Since students have different testing behaviors on different assignments, there needs to be some kind of consistent testing habit among CS students. Students’ testing behaviors must also be observed, but paper concludes that students who reached the “testing threshold milestone” early and later produced higher quality code.

Children learning Computer Science Concepts Via Alice Game-Programming

The content of this paper was very similar to what the Alice team at Duke has been working on – in fact, this paper references some of Professor Rodger’s papers. Again, in an effort to increase diversity in CS-related fields, using Alice in the core curriculum seems to ben effective way to proceed. Previous studies on how game-programming affects students’ ability to learn programming only focuses on the psychological and educational aspects. This study focused more on whether programs were executable or not, and extends on prior research.

Environments like Alice have a drag and drop interface, and include predefined operations. Previous research has found that women are able to be as successful in basic programming, and all the previous work done shows that game-programming has has a positive impact on programming and engages students to do higher-level thinking.

iGame collected students’ work over the course of 2 years, and looked at how game creating and programming can promote computational thinking in middle school students by having students create challenges, similar to the ones used at Duke. Two versions of Alice were used: Storytelling Alice and Alice 2.2.

Students were able to create non-list variables and used if/else statements, but some were unsuccessful. However, still, research shows that making games in Alice can effectively engage students in key features of programming. Future students include the necessity to compare students working with different programming interfaces, and determine what students want to design to incorporate into educational curriculum.


iMpaCT-Math stands for Media-Propelled Computational Thinking for Mathematics Classrooms, this paper discusses how it uses introductory programing to help enforce students’ understanding of math concepts. The studies developed at the University of Texas at El Paso presented in the paper support their hypothesis that programming can provide a path to increase understanding of math.

iMpaCT-Math includes games and project-based learning activities that can be used with high school and math course at the college level. These classrooms combine graphical programming challenges that focus the students’ attention towards exploring math principles to encourage them to pursue STEM. The graphics help the students focus, and at the same time they can adapt programming skills while reinforcing math concepts. iMpaCT-Math rethinks how computation can engage students. Many students without prior interest or exposure to programming increased interest in programming. However, intended CS majors wanted to learn more about how the library was implemented.

Algebra I is a fundamental course for STEM-related studies, and iMpaCT helps students understand these concepts and reinforce them. Most of these exercises are taught in high school, and many students pass math after half a year of impact. Further, the failure rate was halved, and majority of the students had positive attitudes towards programming, more than a quarter of them voluntarily enrolled in AP Computer Science.

Currently iMpaCT-Math is trying to build a foundation of understanding and providing enough evidence to support their long-term goals. They are working on addressing challenges they encountered during their alpha test, primarily working on training teachers. iMpaCT-Math activities are considered “high cognitive demand tasks because they involve reasoning about procedures.” However, teachers want additional materials to help them practice skills, and therefore, they are working to develop a teachers’ workshop to train math teachers to effectively implement the games.


Integrating Computing into Middle School Disciplines Through Projects

Click to access cse12.pdf

In the 2012 publication of Professor Rodger’s research with Alice, they detail their efforts from the two years prior to writing the paper. This paper is more focused on their efforts to integrate computational thinking into middle school science and math education. Much of the earlier research was towards creating curriculum materials for teachers to use in teaching, but this publication focuses more projects that they created based on teacher feedback. Since it is difficult to create a new course, many of projects combine computing with a core subject.

Again, Rodger emphasizes how there is a lack of standards and exposure to CS education. Many students use computers at home and school, but many don’t understand what CS is and how it relates to problem solving. At Duke, the use of Alice in college-level courses showed that there was an improvement in students’ performances in CS, and further an increase in retention in CS.

Most of the paper detailed projects that they had been working on. In previous workshops, the main feedback that they received from teacher was a larger number of sample projects and for there to be longer workshops. Sine there is already a link between science and computer science, it was easy to create projects that created models and collected data. However, there is a need for accurate objects. Teachers can also use Alice as a tool for students to practice a math concept, such as plotting, substituting x, or scientific notation. Additionally, students can create book reports from Alice, and use it to help them learn languages.

In the workshops, teachers were taught the basics of Alice and then were guided to create their own lesson plans. After looking over their lesson plans, many of the teachers plan to use Alice as a supplement to attract more students. They gave positive feedback on the workshops, and expressed their plans to continue to use Alice in the following academic year. Rodger found that the best way to integrate computing into middle school is through the technology teachers and have them collaborate with teachers of different subjects.

Engaging Middle School Teachers and Students with Alice in a Diverse Set of Subjects

This is one of Professor Rodger’s first papers regarding her research with Alice and integrating computing into core subjects in middle school. In the summer of 2008, Rodger and her students held three weeks of training for teachers and two one-week camps of student training.

The primary target audience of this research involves middle school students because it is at this age that middle schoolers start thinking about their careers. Most students being to develop an interest in medicine and education because they are exposed to those occupations daily. Duke University offers an Alice class for non-CS majors, which attracts 50 percent of females, but many do not pursue another CS course at Duke, simply because they do not have an interest in it.

Alice can be used as a tool for middle school students, to further solidify understanding of concepts and to use it as a tool to present projects. It can be used to solve problems in math, create ecosystems and civilization worlds, create a world based on stories/poems that students read, and also create stop-motion animated films. To increase the diversity in CS related fields, Alice appeal to girls because of its story telling aspect.

At the student camps, students were taught Alice, and also got free time to create Alice worlds. Afterwards, Rodger and her students examined their worlds and discovered that many used concepts needed for CS – many used lists, which was an easy concept for the students to grasp. They continued to develop and revise tutorials for students to use. Many of the teachers who attended the workshop at Duke had no prior knowledge of programming, and were more interested in integrating Alice into their lesson plans. Most of the attendees were NC teachers who taught students in the 4th-8th grade. Teachers created lesson plans and planned to use them in the following academic year. They would present their results at a follow-up workshop the next summer.