Bubble Charts
Data Visualization for Peer Graded Systems

A bubble chart can be used to represent 4 dimensions, x-axis, y-axis, color of the bubble and the size of the bubble.

Input data:

JSON Structure

The common properties for all the JSON files are described below:
x_label denotes the entity represented on the x-axis, its label being specified by ‘label_name’ and the URL is given by ‘url’. Upon clicking this URL link the user is redirected to another visualization of, say, Scores received on that particular criterion by different students for each stage of submission of an assignment.
y_label denotes the entity represented on the x-axis, its label being specified by ‘label_name’ and the URL is given by ‘url’.
configuration denotes all the kinds of settings that the user is allowed to configure. This includes the
color_scheme, the color code used to display variation from good to bad.
svg-height denotes the height of the page, in pixels, take up by the visualization.
svg-width denotes the width of the page, in pixels, take up by the visualization.
font-family denotes the font family used to display the labels and text in the visualization.
font-size denotes the size of the font used to display the labels and text in the visualization.
bubble has the display configuration for bubble in the bubble chart, w.r.t. its minimum and maximum radius.
is_mean_user_provided it is a flag to specify if the input given to the visualization are individual scores ( whose mean and variance are calculated programtically), or if the user provides mean scores and variance. Note that, here the mean can be a weighted means, calculated as per some algorithm.
In addition, the “content” property differs across the files, and therefore described below
student_criteria_bubble_with_means.html

In this visualization, the student ids are represented along the y-axis, the criteria along the x-axis, the color denotes the Mean of the scores and the size of the bubbles represent the Standard Deviation/variance.
Content Structure

content

This denotes all the content to be visualized.
It is a 3 dimensional array structure.
The outer most dimension represents the students across all the criteria.
The intermediate dimension represents the criteria.
And finally, the inner most dimension denotes the scores awarded by the reviewers to the given student. The input can either be the mean score & variance, ( i.e. if “is_mean_user_provided” is set to true), in which case, “weighted_mean_score” denotes the weighted mean for all scores and the “variance”. On the other hand, if “is_mean_user_provided” is set to false, then an array of “scores” need to be provided, each element of the scores array, has the score value, the reviewer Id who has awarded the score and the comments, if any.
reviewer_student.html

This visualization opens when the user clicks on any of the links on the x-axis labels. In this visualization, the reviewers are represented on the x-axis and the students are represented on the y-axis. The size of the bubble denotes the value of the score and the color denotes the submission.
Content Structure

The content is a 3 dimensional array, the outermost dimension represents reviewers, the intermediate represents the submission number, while the innermost dimension represents the student.

For example,

// scores by 0th reviewer for all submissions.
reviewerData.content[0];

// scores by 0th reviewer for the 2nd submission.
reviewerData.content[0][2];

// scores by 0th reviewer for the 2nd submission for the 4th student.
reviewerData.content[0][1][4];
In general, we can say…
// Data for ith review
reviewerData.content[i] ;

// All the data(scores, comments) for ith review for the jth criteria.
reviewerData.content[i][j] ;

// The score for the ith review for the jth criteria
reviewerData.content[i][j].score ;

// Comment given by the ith reviewer for the jth criteria.
reviewerData.content[i][j].comment ;
Each score consists of the actual score value and the comments received, if any.
criteria_reviewer.html

This visualization opens when the user clicks on any of the links on the y-axis labels. In this visualization, the criteria are represented on the x-axis and the reviewer ids are represented on the y-axis.
Content Structure

The content is a 2 dimensional array, the outer dimension represents criteria, while the inner dimension represents the student. The size of each bubble denotes the actual score values.

For example, we have..

// scores for 0th student across all criteria.
reviewerData.content[0];
// scores for 0th student for the 2nd criterion.
reviewerData.content[0][2];

In General,
// Data for ith review
raw_reviewer_data[i];
// All the data(scores, comments) for ith review for the jth criteria
raw_reviewer_data[i][j];

// The score for the ith review for the jth criteria
raw_reviewer_data[i][j].score;

// Comment given by the ith reviewer for the jth criteria.
raw_reviewer_data[i][j].comment;

References

https://d3js.org/
https://github.com/PMSI-AlignAlytics/dimple/wiki