Caliper

Introduction

Learning data or "learning analytics" is generated when students interact with online learning tools. These data can be combined with student record and other data points to support student success. These data can be used by faculty for proactive intervention with a student, or to infer how people use the tool and thus can help to improve the learning process. Instructors can also see which behaviors and content produce the desired learning outcomes.

What It Is

Caliper is a specification developed by IMS Global Learning Consortium. The purpose of Caliper is to define a standard to measure and share learning data. Most learning applications work in silos and separate data has historically been captured by each learning application. One problem that these separate data sources create is that each organization needs to reinvent the analytics wheel. A separate problem is that these separate data sources cannot be merged and therefore, none of the parties have a complete picture. Therefore, a standard way of capturing and sharing learning data enables more efficient development of learning analytics features in learning environments. Caliper attempts to overcome these problems by defining a common language for labeling learning data, and provides a standard way of measuring learning activities.

Following are the terms as defined on the IMS website:

• Metric Profile: This is an information model for Caliper and is organized by learning activity type. For example, the Session Metric Profile and the Reading Metric Profile. A Metric Profile includes Entity, Actions and Events.
• Actor: A Caliper Event includes an actor attribute for representing the Agent. An actor is an Agent capable of initiating or performing an action. For example: a student can be an actor.
• Action: These are the actions performed as part of a Learning Interaction. A Caliper Event includes an action attribute for expressing the associated action. LoggedIn is an example of action performed in a Session Metric Profile.
• Entity: an object that participates in learning-related activity such as Person, Assessment, Video, etc.
• Event: Captures the Entities and Actions performed as a "learning event".
• Sensor: Software assets deployed within a learning application that implement the Sensor API for transmitting Caliper data to a target endpoint. Sensors can be implemented in Java, Javascript, PHP, Python, Ruby, or .NET.
• Sensor API: The standard set of methods and supported parameters that a Sensor implements according to this specification in order to transmit Caliper data in an interoperable way.

End Point

Metric Profiles

• Annotation Profile
• Assessment Profile
• Assignable Profile
• Forum Profile
• Grading Profile
• Media Profile
• Reading Profile
• Session Profile
• Tool Use Profile
• Basic Profile

Entities

Actions

Session Event

        
        {
          "@context": "http://purl.imsglobal.org/ctx/caliper/v1p1",
          "id": "urn:uuid:fcd495d0-3740-4298-9bec-1154571dc211",
          "type": "SessionEvent",
          "actor": {
            "id": "https://example.edu/users/554433",
            "type": "Person"
          },
          "action": "LoggedIn",
          "object": {
            "id": "https://example.edu",
            "type": "SoftwareApplication",
            "version": "v2"
          },
          "eventTime": "2018-11-15T10:15:00.000Z",
          "edApp": "https://example.edu",
          "session": {
            "id":
            "https://example.edu/sessions/1f6442a482de72ea6ad134943812bff564a76259",
            "type": "Session",
            "user": "https://example.edu/users/554433",
            "dateCreated": "2018-11-15T10:00:00.000Z",
            "startedAtTime": "2018-11-15T10:00:00.000Z"
          }
        }
      

The complete list of Metric Profiles can be found in Caliper's Documentation.

Also, click here to see Caliper Sensor APIs.