Go to Step 1 labeled "Sections". If you haven't already done so, the app will ask you to import the sections and students into the app. If you're not sure, you can always import the data again. Also, if you think you have new students who registered late to your course, you can click import again and that student will also be added to the app.

Go to Step 2 labeled "Students". All the imported student information will show up there. You can always come back to this tab after the teams are formed to search which team a particular student belongs to.

"Attributes" is the generic term we use to refer to the types of information that you may wish to elicit from students. These include: personal traits, attitudes, demographic information, characteristics, expertise, friends preferences, enemies preferences, time slot availability, and project preferences. Once you have defined the attributes you want to use, you can add them to the survey and ask your students to complete the survey. Similarly, if you want to match students to projects based on their expertise, you can add the expertise-related attributes as the projects requirements.

Since attributes are being used as part of a survey for eliciting information from students, you will need to write out the question text that the students will see for that attribute.

Since attributes are being used as part of a survey for eliciting information from students, an attribute has a question and possible answers that students may respond with. Our app comes with some commonly used answer groups so that you can use them if you like. If you create a new answer group, you can also use it at a later time for another attribute without having to re-define all the details again. Note that for privacy reasons, some students may not wish to answer sensitive questions. In these cases, we suggest you include an option for "Prefer not to say".

Yes, when you go to Step 3 labeled "Attributes", you will see at the bottom there is a "List of Available Attributes to Import". These are the commonly used attributes that we found in the research. You can view them to see if you want to use them as part of your elicitation process. After you import them, you can also modify them.

Yes, you need to create an attribute for all the information you wish to elicit and use to generate teams. Aside from the student name, the student ID, and the course section, we do not get any other student information from the learning management system. While it may save you more time if we do grab basic student information, we believe it is important to that students know exactly what information is being gathered about them.

You can skip the "Projects" step since this information is optional.

In Step 4 labeled "Projects", you can give the set of projects a name first. Next, you can define all the projects that are available in the class. If it is important to match student skills to project requirements, you will want to make sure you identify the appropriate attributes as requirements for the respective projects so that the

You can first define the project, then clone it, and give it a different name. Then when you generate teams, one team will be assigned to the original project and another team will be assigned to the cloned project.

Unfortunately, once the teams have been generated, you cannot modify the requirements. If you need to change the projects, you will need to create a new project set with new projects, and then re-generate the teams.

The purpose of the survey is to help you elicit student characteristics and preferences. If you have already created all the necessary attributes that correspond to information you wish to elicit from your students, you can simply add them to the survey.

No, once you are done building the survey, you can publish it in the learning management system. In Canvas, the published survey will show up as a quiz. Students can complete the quiz in Canvas.

Currently, Teamable Analytics has 4 algorithms available: the weight algorithm, the priority algorithm, the social algorithm, and the random algorithm.

1. If you have a lot of constraints in how you want to build teams, the weight algorithm is the most suitable option because it will try to satisfy a preference for everyone to make everyone happy in some respect.
2. If you have strong feelings about one constraint being satisfied as much as possible first before trying another constraint, you should select the priority algorithm because it lets you indicate the priority of which constraint to handle first.
3. If you wish for the students to be in their social groups, the social algorithm is the best choice for this scenario because it will maximize the students' preferences for stated friends and enemies.
4. If you only need teams to be of a certain size, then the random algorithm is the easiest to setup because it lets you define teams of specific sizes and does not use any other constraint.

Yes. When you view a set of generated teams, you will see at the top of the page the parameters that were used to generate these teams. There is also some summary statistics about the overall teams. You can use this summary to compare which set of teams is better for your class. We will be extending on the available summary statistics in the near future.

In addition to the summary statistics of the team set, our app provides visual analytics to help you quickly inspect how well a team adheres to the criteria you selected for forming teams. When you view a team set, you can select the attributes that you wish to use as part of this diagnosis, the type of graph to use for an attribute, and click "Visualize". Each team will have appear with the set of visual analytics. You can also expand a team and the same visual analytics will appear for each individual team member. This will also help you detect if a specific member should be moved to a different team or not.

You can either identify specific students to move to a different team, or re-generate all the teams again.

Yes, you can select grades from the learning management system's gradebook and display them in the app to better visualize how the teams are performing. The data will be displayed as a box plot graph, so you can see each team's score relative to the class average performance. You can also expand each team to inspect how each student is doing within the team.

You can click on a spot in the graph, keep the mouse down and drag it to another spot in the graph to select an area that you want to zoom into. When you let go of the mouse, that area will appear in closer details. You can double click to zoom back out.

When you have multiple milestones, you may find it useful to turn on the line toggle option so that you can track the progress of a single team or the progress of an individual within a team. The line helps you quickly see if the scores are trending up or down over time.

There are two ways to check how individual students are doing. First, you can use the monitoring function to visualize the performance of a student in terms of their gradebook information. Second, if you have peer evaluations in your course, you can see a summary view of the evaluations that a student received from others in the class.

If the students completed peer evaluations that indicated they do not want to work together again, you can use the re-generate feature and the new social preferences will be considered for the next milestone. However, there is no way to guarantee it because the algorithms just try to do the best they can given all the constraints specified.

Yes, you can go to Step 7 labeled "Peer Evaluations" and follow the steps to create a peer evaluation for students to complete. The steps are similar to creating a survey in the sense that you will need to define attributes first. When you are done, you can preview the peer evaluation from the perspective of the student to verify it looks right to them.

The students will need to use Teamable Analytics to complete the peer evaluations. They should see a link to "Teamable Analytics" in the left navigation bar in the learning management system. Once they click on it, the only options available to them in this app is to complete peer evaluations. Unlike surveys, students are expected to complete one peer evaluation for each of their team members. Teams may also have different number of people in a class. For these reasons, we kept the peer evaluations within the app so that we could manage them better.

As the instructor, you can check the opening and closing date of the peer evaluation.

When you create the peer evaluation, you can decide if you want the responses to be shared with the student receiving the evaluation. This is a setting you can make for each question in a peer evaluation.

Currently, you can create either a multiple choice question or an open-ended text question.

In Step 6 labeled "Teams", you can view the "List of Team Sets" and find the one you were using. Then you can select the icon for re-generate teams. As part of this process, if there are specific peer evaluation feedback you want to use to consider how future teams are formed, you can select those. In addition, if there are certain teams you want to stay the same in the next round, you can lock those teams so that the app won't change the members of those teams.

Yes, when you create a peer evaluation, you can indicate whether you want the student to evaluate their own teammates or evaluate other teams. This option is useful if you want students to be aware of what other teams are doing and to provide feedback to them.

Unfortunately, that feature is not available yet.

Yes. There is an export function in Step 7 for you to export and download the data from Teamable Analytics.

Teamable Analytics interacts with the learning management system via the LTI protocol. As such, our app can be tweaked so that it works with another LMS that also uses the LTI protocol. At present, we have only integrated and tested with the Canvas learning management system.