Invitation to participate in research study

Using "subgoal labels" to improve learning how to code

Subgoal labels have been shown to be an effective method for novices learning computer programming.

We've partnered with researchers to create a version of CSP - Unit 3 that uses subgoal labels.

What's a subgoal label? (see below)

  • Please consider participating in the research study by clicking the orange button below (it goes to a consent form)
  • This is an exciting opportunity to help you, your students, and computer science education research

Got questions? email: research@code.org


More Information About The Study

What's a subgoal label?

A subgoal label is a brief bit of text (a label) that describes a step in some problem solving process. When these labels are provided to the novice it helps clarify the steps needed to achieve some outcome.

The example on the right shows subgoal labels applied to a small piece of code (in a different language than we use in CSP). It can seem simplistic to the coder with experience, but for the novice these labels can really help get them over initial feelings of not even knowing where to begin.

Sounds great! Sign me up for the Subgoal Labels study

Benefits for your students:

Besides helping inform best practices in computer science education, your participation may improve your students’ performance. In preliminary research in computing classes, students who learn using subgoal labels have been shown to:

  • Perform 10-15% better on problem solving tasks (this is the equivalent of a letter grade on a test)
  • Retain more knowledge over time
  • Because of more efficient learning, complete problem solving tasks in less time

Prior research across a variety of disciplines such as statistics, algebra and chemistry suggests that using subgoal labels improves students’ problem solving abilities by helping to break down problems into manageable parts.

By participating you can help to verify these initial results with a wide range of students.

Sign me up! I want to participate in the Subgoal Labels study

What will you have to do to participate?

Very little!

  • There is NO different or additional teacher prep required - if you prepped for Unit 3 already, fantastic.
  • There is NO difference in the Unit 3 structure, lesson sequence, learning objectives, or projects
  • The only differences students will see are (1) how some material is presented to them through the App Lab programming environment (2) some assessment questions

You may be asked to provide feedback on how you perceive your students are doing with unit 3, and we may ask you additional demographic data about your class.

You also must click here to opt-in to participate in the Subgoal Labels Study

How does the study work?

Code.org, in collaboration with the researchers, are conducting a scientific study about how effective subgoal labels are for students learning how to program.

  • For teachers who opt in to participate in the study, all will be assigned to a redesigned version of Unit 3 that includes improvements to instructions and sequencing.
  • A number of those teachers will be randomly selected to receive the subgoal "intervention". Others will still receive a redesigned version of the unit but without the intervention. This type of blind experiment is important for scientific rigor.
  • all students in all your classes will see the same thing, all your students see the same version of the study – your students won’t be split between different versions of the study.
  • Your students will be automatically routed to a "research" version of CSP Unit 3 which will have a different URL that might look something like https://studio.code.org/s/csp3-research-MXGHYT
  • Simply follow along with that version of unit 3, and that's it!

Data collection and protection

  • Code.org does not store student contact information on its servers. All student activity data collected by Code.org is anonymized with randomly generated student and teacher IDs and further stripped of any identifying information before sharing it for research purposes.
  • Results will never be tied back to any individual student, teacher or school.
  • Any data collected directly by the researchers will have no identifying information
  • The researchers will never have direct access to any student generated data – only aggregated data prepared and cleaned by Code.org. An example of this might be how many students per class get a specific question correct on the first try. While data might be aggregated and generalized based on class groupings, there will be no way to know the identity of a student who got the question right or wrong, nor the identity of the class they were part of, nor the identity of their teacher.

The Researchers:

The study is being conducted as part of continuing research on subgoal labels by Dr. Briana Morrison from the University of Nebraska at Omaha and Dr. Lauren Marguliex from Georgia State University

If you have any questions about the research study, subgoal labels, or your participation in the study, please contact Dr. Morrison or Dr. Margulieux:

Dr. Briana B. Morrison, Ph.D.
Assistant Professor
Computer Science Department
College of IS&T PKI 285B
University of Nebraska at Omaha
bbmorrison@unomaha.edu

Dr. Lauren Margulieux, Ph.D.
Assistant Professor
Learning Technologies Division
Georgia State University
lmargulieux@gsu.edu