Fabulous video showing remote labs to complement regular labs. Thanks for sharing this! Love the idea of labs become homework and then data is discussed in the classroom. How many classes are currently using it? Can students at different locations share lab results and discuss it with each other?

Thank you! The data have generated some very interesting class discussions about things like sample size, the need to be able to replicate results, experimental design, and naturally occurring variations.

We recently launched a new user platform at ilabstudio.org which is starting to see significant activity. Our original user platform has had more than 11,000 users all over the world design and run their own experiments.

Students are able to export their results and share them via Excel files or graphs, but we do not yet have a feature that allows student to student sharing within the system. That might be an interesting item for our next release!

Thank you for your comment – we are indeed excited to focus on new areas now that the ilabstudio.org site has been launched. We always welcome comments and suggestions!

Thank you so much for attending last week and for your enthusiasm! Please let me know if there is anything I may do to be of assistance when you try iLabs out in your classroom.

Thank you for your question. When designing the student interface that controls the remote instruments, we make sure to include the opportunity for students to make mistakes in their experimental design such as not collecting enough data points for comparison, not measuring for a long enough period to see results, and not including enough trials to make a definitive statement about the natural phenomenon in question. This gives teachers implementing remote labs in their classrooms the opportunity to discuss data collection and experimental design as a class using individual or aggregated results as examples, or include scaffolding materials within the lab journal itself to guide students towards improved experimental design. Students can then perform the experiment again with a (hopefully) improved design.

Thank you for your comments and inquiry! Thus far recruitment has not been too difficult as many instruments located at university labs have significant periods of time where they are not in demand and most belong to labs with some sort of broader impacts or outreach mandates. The challenge lies more in finding instruments that meet our criteria for being a useful remote lab, namely that students are able to design and run an individual experiment, that the process can be automated to run with minimal human intervention, and that the data can be returned in a format that is meaningful over the internet. We carefully evaluate and design each addition to make sure that it is scalable, and that it will be able to keep up with increased demand as the project grows.

Thank you for your input and your questions! We currently have instruments located at the University of Queensland in Australia, Northwestern University in Evanston, Illinois, and at Stanford University in Stanford, California. Individual instruments are maintained by the labs that house them. Many of the instruments are shared, meaning that K-12 students can access them remotely but students at the universities use them as well, so they are kept in good working order. Costs incurred by remote usage (supplies, wear and tear) are usually covered by the hosting lab as broadening impact or outreach costs.

There are occasional traffic jams caused by multiple users submitting experiments at one time, but we encourage teachers to include time for the experiment to run when planning the assignment. Students submit their experiments into a queue, and then the queue is processed one at a time. Our user interface is designed so that students can log out and come back at any time, and have their work saved. It also provides an estimate of how long you will need to wait to receive results. Teachers often have students work to the point of experiment submission in class, let the experiments run over night, and then come back the next day to discuss results. The lab journal assignment can also be given as homework, with the caveat that students will need to manage their time to allow sufficient time for the experiment to run and for them to finish the rest of the lab journal before the due date.

Thank you, Tammy! We appreciate your comment.

Hi, Jeremy!

There are a few critical features we emphasize in the design of a “great” remote lab:

1) Students should be able to design and run individual experiments that answer a specific research question they have developed. This means that the lab can’t just be something you “watch” – students need to have control over the experimental parameters that allow them to manipulate the instrument and explore the behavior of the phenomena under investigation.

2) They should be able to make mistakes in experimental design, be able to observe in the data whether their design was effective or not, and then do the experiment over again once they figure out what those mistakes are. This means that the interface has to allow for those mistakes and not prevent them (even if we as designers know ahead of time that certain inputs aren’t that useful).

3) Students should get back real data with naturally occurring variation that they must then make sense of and use to support some sort of claim or explanation about the phenomenon in question.

4) On the teacher side, we have also found that it is crucial to allow teachers the flexibility to customize each remote lab experience for their students (local adaptation).

5) Finally, the best labs are ones that can be run asynchronously – that is where the student does not need live control of the instrument. By being able to queue up each student’s experiment, we not only use the instrument time more efficiently, but it is much more user-friendly for students who don’t have to worry about reserving a specific time slot.

Some of the best classroom discussions that we have observed in our pilot classrooms using remote labs have focused on experimental design and data analysis. For example, in the radiation verses distance lab, many students initially make the mistake of taking samples at only two distances from the radiation source. When asked about the mathematical relationship between radiation and distance, they will conclude that there is a negative, linear relationship between the two variables (in fact, the relationship is an inverse-square, i.e. one over the distance squared). The teacher will display one of these graphs and have the class discuss it. Then the teacher will display the graph of a student with a better experimental design, sampling at four or five distances, and ask the class to explain the difference between the two graphs. Students immediately see the difference, and how inadequate sampling can lead to significant errors in data analysis. The teacher can then display the cumulative data for the class, and the discussion can shift to reliability and reproducibility of results, and students can see how a few outliers due to natural variation can cause them to misinterpret results if the sample size is not large enough.

The best discussions we have seen are primarily student led, with the teacher acting as a facilitator while students draw their own conclusions about what they are seeing. Our cyberlearning tools for teachers now includes new analytics-based screens that aggregate whole-class experimental designs and results specifically for the teacher to project on a projector and use to facilitate discussion and debate among his or her students. This way, all the students get to see a roll up of the data and the range of approaches to experimental design.

Thank you for your comment! This approach has also been very popular in Australia.