Designing Report Data

Before using Learnosity's Large Group Report, customers should consider the guidance in this article to help design the report for their needs.

There are two kinds of Large Group Report, to support different methods of selecting the desired assessment data. The design of your assessment implementation will determine which of the two reports suits your use case. Aside from the data selection criteria, the two reports offer the same analytical capabilities.

The "activity-summary-by-group" report summarises data for a cohort of students who received the same activity_id:

  • Use the activity summary report to analyse assessment data where the same activity_id has been issued to a cohort of students who undertook an identical or directly comparable learning activity. If all students were served the same semester exam, with identical questions or comparable questions drawn from a calibrated pool, and each assessment was initialised using the same activity_id, the activity summary report can be used to select the most recent attempt of that activity for each student.
  • The activity summary can accept up to 10 activity identifiers.
  • Each user score will be aggregated as the sum of their scores on the specified activity_id values. If multiple sessions were recorded for an activity, the score of the most recent session will be used. This supports more complex scenarios, like where users have been issued multiple assessments which can be summarised together, eg. viewing the total score from the last 10 weekly tests, or from all 3 parts of a multi-part exam.
  • A mintime and maxtime may be specified to consider only sessions completed within that timeframe.

The "sessions-summary-by-group" report summarises user scores from a set of specific sessions per user:

  • Use the Sessions Summary Report to specify the set of sessions to be aggregated per user.
  • The activity summary can accept up to 50 session_id values per user. The report will aggregate each user's score from the sessions specified for that user (ie. their total score will be the sum of scores across the sessions specified for that user).
  • A mintime and maxtime may be specified to consider only sessions completed within that timeframe.

Design the levels at which user scores will be aggregated. Each level has an arbitrary label and up to 1000 group names. Example groups for a report of Australian schools might include State, Region, School and Class. In this example, different schools have different conventions for naming each class; remember, all group names are just arbitrary strings:

Group label: State Region School Class
Group names: Queensland North
Central
South
West
Cairns State High School
Cooktown State School
Lockhart State School
...
9-Gregory
9-Leighton
9-Willoughby
...
New South Wales North
Central
South
West
Fort Street College
Oxford College
St Andrew’s Cathedral High
...
yr9_A
yr9_B
yr9_C
...
Victoria East
Central
West
Belmont High School
Melbourne Girls College
Williamstown High School
...
C612
C678
C690
...

Example groups for a report to analyse assessment scores by age and sex might be:

Group label: Age Sex
Group names: 16 Female
Male
Intersex
17 Female
Male
Intersex

Consider the following when designing your groups:

  • Every user must be allocated to exactly one group at every level in the hierarchy. For example, it must be possible to allocate each student to exactly one State, Region, School and Class for a given instance of an aggregate report. If a particular student transferred between two different schools, their score must be attributed to a single School group for the purposes of summarising their result.
    • The scores for all users in a group will be aggregated to create a summary row for that group.
    • It may make sense to create two or more group reports with different group hierarchies and a subset of the activities.
  • Customers will need the relevant data sources within their own systems to be able to associate users to the desired groups. Learnosity only stores the user_id provided with each session. We do not store any demographic information about a user session.
  • Refer to the limitations on the maximum size and number of groups.

The scores for each group of users will be aggregated to calculate and display statistics about that group. For example, the report can display the average, median and 75th percentile score for each group of users, as well as many other statistics. If all the lowest level groups of the hierarchy contain <= 1000 users, raw scores can also be displayed for those users.

Refer to the list of available fields for groups and user level data.