Digital Media Attention Share Tracker — Methodology

The tracker measures the share of public attention on digital platforms — YouTube, Instagram, and Facebook — received by political leaders and parties during the election period.

The primary metric reported is Attention Share (%) — the proportion of total measurable engagement attributable to a leader or party over a defined time window (e.g., last 24 hours or last 7 days).

The analysis uses publicly accessible data from:

The following engagement signals are considered: Views, Likes, Comments, and Shares. Additional metadata includes content title and description, channel / page name, and publishing timestamps.

No private, user-level, or platform-internal data is used.

The dataset includes content from a curated universe of official political party and leader channels/pages, news publishers and media organizations, political commentary channels/pages, and relevant public content creators.

Only regionally relevant language ecosystems are considered.

Each content unit (video/post/reel) is analyzed to identify mentions of political leaders and political parties. A content item may be associated with one or multiple entities.

Entity detection is performed using title and text parsing, Named Entity Recognition (NER) models, and confidence scoring mechanisms. Only entities above a predefined confidence threshold are included in the analysis.

Attention Share is computed over rolling windows:

Since engagement metrics (views, likes, shares, etc.) are cumulative, incremental engagement within the selected window is calculated using snapshot differences.

For each content item, engagement within a time window is calculated as the difference between the latest and previous engagement counts:

This method is applied consistently across all engagement signals.

For each leader or party, total engagement is the sum of all engagement across content mentioning the entity. Engagement is aggregated across all platforms and all content formats (videos, posts, reels).

Attention Share is calculated as:

This represents the relative share of digital attention captured by each entity.

If a content item mentions multiple leaders or parties, the full engagement value is attributed to each identified entity. This approach reflects that a single piece of content contributes to the visibility of all entities discussed within it.

• Higher Attention Share (%) indicates greater visibility and audience attention
• The metric reflects media presence, not sentiment or public support
• It does not imply endorsement, favorability, or electoral outcomes

The Attention Share metric does not account for sentiment (positive vs negative coverage), audience demographics, watch time or content retention, and platform algorithmic biases.

Results are influenced by the selection of channels/pages, accuracy of entity detection models, and platform-specific engagement dynamics.

Data is updated daily. Metrics reflect the latest available engagement deltas within the defined time windows.