Mobile CSP Students Perform Well on 2018 AP CSP Exam
Congratulations Mobile CSP teachers and students!
The official results are in and Mobile CSP students performed very well on the 2018 AP CSP exam. Out of 73,732 students who took the 2018 exam, 4,031 (5.4%) were part of a course using the Mobile CSP syllabus and 76.2% of Mobile CSP students passed with a grade of 3, 4, or 5, compared to 68.9% for all students. In the following chart Mobile CSP students are shown in blue and are compared to the performance of all students. 
For the multiple choice part of the exam, the College Board divides all students evenly into quartiles and then calculates the percentage of our students who fall into each quartile. As this chart shows, 32% of Mobile CSP students scored in the top quartile, while only 21% or our students scored in the bottom quartile. These are good results.
On the performance task portions of the exam, Mobile CSP students did even better, with 33% of our students scoring in the top quartile and only 17% in the bottom quartile. 
Performance by Content Area
The College Board report (link below) provides a breakdown of student performance by content area (e.g., Algorithms, Abstraction, Programming, etc.). In every one of the 15 content areas reported for the multiple choice exam and the 5 areas reported for the performance tasks, Mobile CSP students performed at or higher than the global average. For example, in the area of Abstraction 1,475 (36.6%) Mobile CSP students scored in the top quartile compared to all students. And in the Create Task Apply Algorithms area 1,609 (39.9%) scored in the top quartile. These are very strong results.
Performance by Demographic Group
The College Board has not yet given us a demographic breakdown of the Mobile CSP cohort. We will update this summary when we receive those data.
For more details, here’s a link to the original report from the College Board.
Congratulations to all our students and teachers!
App Inventor is a Good Programming Language for Learners
In a recent blog post in the Communications of the ACM Mark Guzdial identified Five Principles for Programming Languages for Learners. In a nutshell, the five principles suggest that a good programming language for learners should be low threshold, high ceiling, a phrase coined by Seymour Papert, the creator of the Logo programming language, a language that Mark recognizes as a good model of the five principles.
How well does App Inventor, Mobile CSP ‘s programming language, do in satisfying Mark’s principles?
1. Connect to what learner’s know. Today’s high school and college students practically live on their smart phones. The Mobile CSP course tries to meet them where they live and help them transition from users of mobile technology to creators of real mobile apps. This is part of the low threshold.
2. Keep cognitive load low. As Mark acknowledges, blocks-based languages do “a terrific job of lowering cognitive load.” App Inventor is an excellent example of this. App Inventor’s components and blocks (its programming tools) are powerful abstractions that meet students at a level where they can immediately begin creating their own apps. Students are able to create a mobile app from scratch and run it on their mobile device on their first day of programming. So here too App Inventor presents a low threshold for entry into a learning experience.
3. Be honest. In App Inventor and in the Mobile CSP course, students are building real apps. Students are encouraged (well, required really) to create apps that benefit their friends, families and communities. Currently App Inventor apps run only on Android devices, but we expect that to change soon. The 2018 edition of the Mobile CSP course expects to support both Android and iPhone programming in App Inventor. This is part of the high ceiling.
4. Be generative and productive. A good language for learning to program must enable the learner to generate new ideas and be productive in exploring new domains. In the Mobile CSP course we are trying to promote computer science for all. We don’t expect all students to become professional software developers. They should be able to apply their new knowledge and skills to explore their interests. And this is what we see in the Mobile CSP course: students’ final creative projects span a broad range of interests in art, psychology, science, math, education or other domains. This too is part of the high ceiling — there is no limit on what students can achieve.
5. Test don’t trust. The idea here is to see what works. Our experience has shown that students and teachers respond positively to App Inventor. For example, in an analysis of our pre- and post-surveys from the 2016 course, 85% of students reported that they enjoyed programming and getting their apps to work.
I think it’s pretty clear that App Inventor satisfies Mark’s criteria as a good programming language for learners. But this is not surprising given its genesis. App Inventor was created by Hal Abelson who worked with Seymour Papert in implementing and using Logo in education. So, as described in this video, you can sort of see App Inventor as a Logo for the mobile age!
78% of Mobile CSP Students Passed AP Exam
Congratulations Mobile CSP teachers and students!
The official results are in and Mobile CSP students performed very well on the AP CSP exam. Out of 47,216 students who took the 2017 exam, 3,611 (7.6%) were part of a course using the Mobile CSP syllabus and 78% passed with a grade of 3, 4, or 5, compared to 74% nationally. In the following chart Mobile CSP students are shown in blue and are compared to the performance of all students. 
For the multiple choice part of the exam, the College Board divides all students evenly into quartiles and then calculates the percentage of our students who fall into each quartile. As this chart shows, 30% of Mobile CSP students scored in the top quartile, while only 20% or our students scored in the bottom quartile. These are good results.
On the performance task portions of the exam, Mobile CSP students did even better, with 40% of our students scoring in the top quartile. 
Performance by Content Area
The College Board report (link below) provides a breakdown of student performance by content area (e.g., Algorithms, Abstraction, Programming, etc.). In every one of the 15 content areas reported for the multiple choice exam and the 5 areas reported for the performance tasks, Mobile CSP students performed at or higher than the global average. For example, in the area of Abstraction 1,780 (49%) Mobile CSP students scored in the top quartile compared to all students. And in the Create Task Apply Algorithms area 1,671 (46%) scored in the top quartile. These are very strong results.
Performance by Demographic Group
In an earlier report the College Board gave us a demographic breakdown of the Mobile CSP students who took the exam. This was for a somewhat smaller sample size (3,531 examinees). What these results show is that Mobile CSP students out-performed the average in every demographic group.
Gender and Demographic Breakdown
In terms of gender breakdown, Mobile CSP students track pretty closely to the national group, with 29% female students (compared to 30% nationally).
Demographically, we come in a little lower in terms of our percentages of underrepresented minorities (Black/African American and Hispanic/Latino).
For more details, here’s a link to the original report from the College Board.
Congratulations to all our students and teachers!