Unit 4 Study Notes: Social Media and Data

Analytics
These study notes cover the specified topics for Unit 4, providing detailed explanations for your exam
preparation in social media analytics and data analytics. Each section includes definitions, key concepts,
applications, and examples, tailored for quick review and understanding.

Social Network and Modelling

1. Social Contexts: Affiliation and Identity

• Definition: Social contexts are environments where individuals interact, and affiliation and identity
describe how group memberships shape self-perception and behavior (Social Identity Theory).
• Key Concepts:
o Social Identity Theory: Individuals categorize themselves into in-groups (e.g., X
communities) and out-groups, influencing self-esteem and behavior.
o Affiliation Networks: Two-mode networks linking individuals to groups (e.g., X users in a
hashtag group).
o Homophily: Tendency to connect with similar individuals, based on status (e.g., age) or
values (e.g., interests).
• Applications:
o Predicting community formation on social platforms.
o Analyzing echo chambers in X discussions.
• Example: X users joining a fandom group form an affiliation network, shaping their identity through
shared interests.
• Key Takeaway: Affiliations drive identity and behavior, critical for understanding social media
dynamics.

2. Social Capital

• Definition: Social capital is the value derived from social networks, including trust, reciprocity, and
access to resources (Social Capital Overview).
• Key Concepts:
o Bonding Social Capital: Strong ties within close groups (e.g., family).
o Bridging Social Capital: Weaker ties across diverse groups (e.g., professional contacts).
o Linking Social Capital: Connections to authoritative figures (e.g., industry leaders).
o Measurement: Uses network metrics like centrality and clustering coefficients.
• Applications:
o Enhancing business opportunities via LinkedIn connections.
o Improving community resilience through X interactions.
• Example: A LinkedIn user leverages bridging social capital to access job opportunities through
distant connections.
• Key Takeaway: Social capital enhances network value, supporting collaboration and opportunity.

3. Structural Holes

• Definition: Structural holes are gaps between unconnected individuals or groups in a network,
offering opportunities for brokers to gain advantages (Structural Holes).
• Key Concepts:
o Brokerage: Bridging gaps to control information flow.
 o Information Advantage: Access to diverse, non-redundant information.
o Network Efficiency: Reduces redundancy in connections.
• Applications:
o Identifying influencers who connect disparate X communities.
o Enhancing innovation by linking diverse ideas.
• Example: An X user connecting two unrelated hashtag groups gains influence by sharing unique
insights.
• Key Takeaway: Bridging structural holes provides strategic advantages in information and influence.

4. Structural Balance

• Definition: Structural balance theory predicts that social networks tend toward balanced
relationships, where triads (three-node groups) are stable (Structural Balance).
• Key Concepts:
o Balanced Triads: All positive ties (friend-friend-friend) or two negative and one positive
(enemy-enemy-friend).
o Unbalanced Triads: Cause tension, driving network change.
o Signed Networks: Edges are positive (friend) or negative (enemy).
• Applications:
o Predicting group stability in X communities.
o Analyzing conflict in social networks.
• Example: In an X network, if A follows B and B follows C, balance suggests A likely follows C.
• Key Takeaway: Balanced networks reduce tension, shaping stable social structures.

5. Predictive Modeling

• Definition: Predictive modeling uses historical social network data to forecast future behaviors or
outcomes (Predictive Modeling).
• Key Concepts:
o Algorithms: Regression, classification, neural networks.
o Features: Network metrics (e.g., centrality), user attributes (e.g., activity level).
o Evaluation: Metrics like accuracy, precision, recall.
• Applications:
o Predicting viral X posts based on engagement patterns.
o Identifying potential influencers.
• Example: A model predicts which X users will share a post based on past sharing behavior.
• Key Takeaway: Predictive modeling enables proactive strategies in social media marketing.

6. Descriptive Modeling

• Definition: Descriptive modeling summarizes network structure and properties to understand

organization and dynamics (Descriptive Modeling).
• Key Concepts:
o Community Detection: Identifies densely connected groups.
o Centrality Measures: Degree, betweenness, closeness for node importance.
o Clustering Coefficients: Measures local connectivity.
• Applications:
o Mapping X community structures.
o Analyzing network evolution over time.
• Example: Descriptive modeling reveals clusters of X users discussing a trending topic.
• Key Takeaway: Descriptive models provide a foundation for network analysis and insights.
7. Community/Anomaly Detection

• Definition: Community detection identifies cohesive groups, while anomaly detection spots outliers
in networks (Community/Anomaly Detection).
• Key Concepts:
o Community Detection: Algorithms like Louvain or modularity optimization.
o Anomaly Detection: Statistical or machine learning methods to identify unusual patterns.
o Metrics: Density, centrality, edge weights.
• Applications:
o Detecting fake X accounts (anomalies).
o Identifying X hashtag communities.
• Example: Anomaly detection flags an X user with unusually high posting rates as a potential bot.
• Key Takeaway: These techniques enhance network security and understanding of group dynamics.

Facebook Analytics
1. Introduction

• Definition: Facebook Analytics, now part of Meta Business Suite, tracks audience behavior, content
performance, and marketing impact (Facebook Analytics Guide).
• Key Features:
o Metrics: Reach, engagement, likes, followers, demographics.
o Tools: Meta Business Suite, Creator Studio, third-party platforms like Sprout Social.
• Importance:
o Optimizes marketing strategies.
o Enhances audience targeting.
• Example: A business uses Meta Business Suite to track post engagement and adjust content strategy.
• Key Takeaway: Facebook Analytics drives data-driven marketing decisions.

2. Parameters

• Definition: Parameters are key metrics for analyzing Facebook Page performance (Facebook
Parameters).
• Key Parameters:
o Reach: Unique users seeing content.
o Engagement: Likes, comments, shares, clicks.
o Impressions: Total content displays.
o Click-through Rate (CTR): Percentage clicking links.
o Conversion Rate: Percentage completing actions (e.g., purchases).
o Referral Traffic: Visits from external sources.
• Applications:
o Measuring campaign effectiveness.
o Optimizing ad spend.
• Example: A high CTR on a Facebook ad indicates effective call-to-action design.
• Key Takeaway: Parameters quantify performance for strategic adjustments.

3. Demographics

• Definition: Demographic insights reveal audience characteristics like age, gender, and location
(Audience Insights).
• Key Metrics:
o Age and gender distribution.
 o Geographic locations (cities, countries).
o Language preferences.
o Interests and behaviors.
• Applications:
o Tailoring content to audience segments.
o Targeting ads effectively.
• Example: A brand targets 25-34-year-old females in urban areas based on demographic data.
• Key Takeaway: Demographics enable precise audience targeting.

4. Analyzing Page Audience

• Definition: Analyzing the page audience involves understanding who engages with content and their
behaviors (Page Audience Analysis).
• Key Insights:
o Audience growth trends.
o Engagement patterns (e.g., peak times).
o Content preferences (e.g., video vs. text).
• Tools:
o Meta Business Suite: Audience tab for demographics.
o Sprout Social: Detailed reports and competitor analysis.
• Applications:
o Refining content strategy.
o Increasing engagement rates.
• Example: A page finds videos posted at 8 PM get the most engagement, adjusting posting schedules.
• Key Takeaway: Audience analysis informs content and engagement strategies.

5. Reach and Engagement Analysis

• Definition: Reach measures unique users seeing content, while engagement tracks interactions
(Reach and Engagement).
• Key Metrics:
o Reach: Paid, organic, viral, nonviral.
o Engagement: Likes, comments, shares, clicks.
• Analysis:
o Identify high-performing content types.
o Optimize posting schedules.
• Tools:
o Meta Business Suite for real-time data.
o Sprout Social for competitive insights.
• Example: A post with high organic reach but low engagement prompts a review of content quality.
• Key Takeaway: Reach and engagement analysis drives content optimization.

Google Analytics
1. Introduction and Working

• Definition: Google Analytics 4 (GA4) is a platform for tracking website and app performance,
providing insights into user behavior (How GA Works).
• How It Works:
o Add JavaScript code (gtag.js) to website pages.
o Collects data on interactions, browser details, and traffic sources.
o Processes data and generates reports.
 • Key Features:
o Real-time reporting.
o Customizable dashboards.
o Integration with Google Ads.
• Example: A website tracks page views to identify popular content.
• Key Takeaway: GA4 provides comprehensive insights for performance optimization.

2. Brief Implementation Technology

• Definition: Implementation involves adding tracking mechanisms to collect data (GA4

Implementation).
• Key Technologies:
o gtag.js: Standard JavaScript tracking code.
o Google Tag Manager (GTM): Simplifies code management.
o APIs:
▪ Data API: For reports and audience export.
▪ Admin API: For account configuration.
▪ Measurement Protocol: For server-side events.
▪ User Deletion API: For data privacy compliance.
• Implementation Steps:
o Create GA4 account and property.
o Add gtag.js or use GTM.
o Configure data streams.
• Example: A site uses GTM to manage GA4 tracking for multiple pages.
• Key Takeaway: GA4 implementation is flexible but requires careful setup.

3. Limitations

• Definition: GA4 has constraints that impact its usability (GA4 Limitations).
• Key Limitations:
o No data migration from Universal Analytics.
o New interface may slow workflows.
o No views functionality for traffic isolation.
o Limited custom dimensions (25 user-scoped, 50 event-scoped).
o Data-driven attribution model with limited flexibility.
o No scheduled email reports.
• Applications:
o Requires workarounds like BigQuery for advanced reporting.
• Example: A user must recreate UA reports in GA4 due to non-migratable data.
• Key Takeaway: GA4’s limitations necessitate additional tools for full functionality.

4. Performance Concerns

• Definition: Performance issues arise from GA4’s tracking mechanisms (Performance Concerns).
• Key Concerns:
o Page Load Time: gtag.js loads asynchronously but may add milliseconds.
o Bandwidth: Script (30.92kB, 12.63kB gzipped) and data transmission use bandwidth.
o Asynchronous Requests: May block site functions on heavy async sites.
o Execution Time: First load ~330ms, subsequent loads 35-130ms.
• Mitigation:
o Place code at page bottom.
o Use caching (86,400s max-age).
 • Example: A site with heavy async requests optimizes GA4 placement to avoid delays.
• Key Takeaway: Proper implementation minimizes performance impacts.

5. Privacy Issues

• Definition: GA4 faces privacy challenges due to data collection and transfer practices (Privacy
Issues).
• Key Issues:
o Collects PII via third-party cookies.
o Data stored on US servers, violating GDPR in some rulings.
o Legal risks: Ruled non-compliant in France and Austria.
o Reputational risks: 77% of consumers value data transparency.
• Mitigation:
o Use anonymization features.
o Consider alternatives like Matomo.
• Example: An EU site switches to Matomo to comply with GDPR.
• Key Takeaway: Privacy concerns require careful compliance strategies.

6. Google Website Optimizer

• Definition: Google Website Optimizer, now Google Optimize, was a tool for A/B testing and
conversion rate optimization, discontinued in 2023 (Google Optimize).
• Key Features (Historical):
o Editor for creating page variations.
o Reporting integrated with Google Analytics.
• Status:
o Discontinued due to limited features.
o Alternatives: VWO, Optimizely.
• Example: A site tested button colors to increase conversions before Optimize’s sunset.
• Key Takeaway: Users must adopt alternative tools for website optimization.
Summary Table
Topic Key Concept Example Application
Group memberships shape self-
Affiliation and Identity Predicting X community formation
perception
Accessing job opportunities via
Social Capital Value from network relationships
LinkedIn
Structural Holes Gaps offering brokerage advantages Connecting X hashtag groups
Structural Balance Balanced triads for network stability Analyzing X follower relationships
Predictive Modeling Forecasting behaviors with data Predicting viral X posts
Descriptive Modeling Summarizing network structure Mapping X community clusters
Community/Anomaly
Identifying groups and outliers Detecting fake X accounts
Detection
Tracks audience and content
Facebook Analytics Intro Optimizing ad campaigns
performance
Facebook Parameters Metrics like reach, engagement Measuring post effectiveness
Facebook Demographics Audience characteristics Targeting ads to specific groups
Analyzing Page Audience Understanding engagement patterns Refining content strategy
Reach and Engagement Measuring visibility and interactions Optimizing post timing
GA Introduction Tracks website/app performance Analyzing page views
GA Implementation JavaScript and APIs for tracking Setting up GA4 with GTM
GA Limitations Constraints like no UA migration Using BigQuery for reports
GA Performance Concerns Impacts on load time, bandwidth Optimizing code placement
GA Privacy Issues GDPR non-compliance risks Switching to Matomo for compliance
Google Website Optimizer Discontinued A/B testing tool Using VWO for testing