COPUS stands for Classroom Observation Protocol for Undergraduate STEM. It is reliable, meaning different well-trained observers are able to independently log nearly the same data. High reliability is not difficult to achieve because the tool is designed to simply describe what is happening among teachers and students across the class period. This low-inference feature makes it possible for undergraduates to reliably take COPUS data in an unbiased way.
While COPUS was developed within larger STEM classrooms, the categories of data that are logged are general enough to be informative for a number of class settings and styles.
COPUS was developed and validated in 2013 by former CU Science Education Initiative postdoc Michelle Smith, SEI Director Sarah Gilbert, and CU Nobel Prize winner Carl Weiman (). It has since has been used in recent education research (; ). COPUS is being used in a number of institutions of higher ed, including BU and UC-Davis, to promote the advancement of teaching and learning. .
听
听
The online technology that streamlines the collection, visualization, and analysis of COPUS data is known as GORP (Generalized Observation and Reflection Protocol). In addition to COPUS, GORP hosts a number of other observation tools. New observation tools can be built within GORP as well.
This is what the user interface of COPUS looks like, on the GORP platform. The left hand section (orange / rust) displays codes related to what the students are doing. The middle section (blue) displays codes for high, medium and low levels of class-wide student engagement. The right hand section (green) displays codes related to what the instructor is doing. The images help the observer quickly find the relevant code button.
During an observation, the GORP platform begins a 2 minute timer, located in the center grey area of the screen. During that interval, the observer clicks on the relevant code buttons, and enters notes in the comment box in the bottom center. The code data is automatically saved into a CSV file. At the end of the 2 minute interval, the page refreshes and the code buttons return to their original state. The observer continues to code each 2 minute interval until the class session is over. Classes of any length can be observed. If the observer makes a mistake, they can note it during the next interval, and adjust the data accordingly by hand, after class.
A CSV data file for each class session can be exported for analysis and visualization. The dataset for each class session is also automatically visualized in the GORP platform. These automated visuals are still under development. This is an example of a class timeline that GORP currently generates.
