Skip to content
Canadian Immigration Dashboard [ CID ]
Civility Analysis

Discourse Quality

Toxicity and constructive discourse scores from Google's Perspective API (via Communalytic).

Comments Scored
55,769
Embeddings
55,418
Avg Toxicity
0.2446
Avg Constructive
0.4192
Highly Toxic (>0.5)
7,830
14.0% of scored

Toxicity Breakdown by Video (Top 10)

Average toxicity scores across comments on each video — bars show how commenters' language scores on each toxicity dimension.

Constructive vs Toxic

Each point is a comment. X-axis = toxicity, Y-axis = average constructive score (affinity, compassion, curiosity, nuance, personal story, reasoning, respect).

27,099
Constructive & Civil
2,759
Passionate but Toxic
19,440
Neutral / Low Signal
4,931
Purely Toxic

Based on 54,229 comments with both toxicity and constructive scores.

Toxicity & Constructiveness Over Time

Monthly average toxicity vs constructive scores — track how discourse quality shifts around key events.

Discourse Quality Score by Video (Top 10)

Composite score = avg constructive - avg toxicity, computed from comment-level Perspective API scores. Green bars = net constructive discourse; red bars = net toxic discourse.

Discourse Quality Score by Channel (Top 10)

Same composite score aggregated by channel — based on comment toxicity/constructiveness scores, not the channel's own content. Which channels' comment sections foster constructive vs toxic discussion?

{# Expects: explorer_rows, explorer_total, explorer_pages, current_page, page_range, filter_opts, f_q, f_polarity, f_tox_min, f_tox_max, f_sort, f_cluster, f_scope, explorer_reset_url #}

Comment Explorer

Browse comments with toxicity & constructive scores. Filter by keyword, polarity, toxicity range, or cluster.

Search & Filter

Search comment text, filter by category or toxicity level
Clear all | Quick: Toxic (>0.5) Most constructive High toxicity (>0.7) Most liked
0 comments
No comments found matching your filters.
Data source: comment_perspective_scores (Google Perspective API via Communalytic) · Scores are probability values (0-1) indicating how likely a reader would perceive the interaction as toxic or constructive.