10. Network Analysis

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/10

encourage image

There's no tags or description

Looks like no tags are added yet.

11 Terms

1
New cards

The Importance of Network Analysis

Always plays a fundamental role in understanding social structures and relationships.

  • Basic concepts:

    • Ego networks = networks formed by a single focal node and its direct ties.

    • Complete networks = all nodes and all ties within a defined boundary

  • Steps of network data collection

    • Questionnaire construction — name generators and questions wording for collecting relational data

    • Data recording

    • Analysis

    • Visualization tools — using Gephi and Pajek to visualize networks

    • Indices, models — applying statistical and visual methods with tools such as Gephi, Pajek, R and Python


2
New cards

Graph Theory — The Bridges of Königsberg

  • 1736 Leonhard Euler

  • Question — Is it possible to walk over all the bridges in such a way that I cross each bridge exactly once and return to the starting point?

    • If every node has an even degree, then such a walk would be possible: an Eulerian graph

  • Nodes (vertices)

  • Edges (links)

    • Directed edges

    • Undirected edges

  • Degree

  • Paths


3
New cards

Two Important Milestones: I. Sociometry

A method for exploring an individual’s system of social relationships

  • Jacob L. Moreno: spontaneous choices of affinity

    • Background: Group-dynamics studies, 1917-19, Vienna: “The Theatre of Spontaneity” (Stegreiftheater)

  • Method: Sociometric Questionnaire

    • A questionnaire given to all members of a community: Whom would they choose / whom would they reject in important life situations?

    • Our conclusions are always valid only within the given community

    • Determines an individual’s position within the social structure and relational patterns

    This is How it Started: Moreno’s Sociometric Matrix


<p><strong>A method for exploring an individual’s system of social relationships</strong></p><ul><li><p><strong>Jacob L. Moreno: spontaneous choices of affinity</strong></p><ul><li><p>Background: Group-dynamics studies, 1917-19, Vienna: <em>“The Theatre of Spontaneity” (Stegreiftheater)</em></p></li></ul></li><li><p><strong>Method: Sociometric Questionnaire</strong></p><ul><li><p>A questionnaire given to all members of a community: <em>Whom would they choose / whom would they reject in important life situations?</em></p></li><li><p>Our conclusions are always valid only within the given community</p></li><li><p>Determines an individual’s position within the social structure and relational patterns</p></li></ul><p><strong><em>This is How it Started: Moreno’s Sociometric Matrix</em></strong></p></li></ul><p></p>
4
New cards

Sociogram

  • A “map” of groupings and connections in the social field (Mérei Ferenc, 1983):

    • “Informal” connections

  • The network of spontaneous associations alongside the institutional system of formal statuses

    • Every person is marked by a circled number

    • Gender can also be indicated

    • Mutual choices: double line

    • Leaders, pairs, excluded individuals

    • Main groups and subgroups


<ul><li><p><strong>A “map” of groupings and connections in the social field</strong> <em>(Mérei Ferenc, 1983):</em></p><ul><li><p>“Informal” connections</p></li></ul></li><li><p><strong>The network of spontaneous associations alongside the institutional system of formal statuses</strong></p><ul><li><p>Every person is marked by a circled number</p></li><li><p>Gender can also be indicated</p></li><li><p>Mutual choices: double line</p></li><li><p>Leaders, pairs, excluded individuals</p></li><li><p>Main groups and subgroups</p></li></ul></li></ul><p></p>
5
New cards

Theory of Social Influence (Cialdini, 1984; Asch, 1955)

What determines an individual’s opinion?

  • Opinions of other key individuals

  • Strength of one’s relationship to these individuals

  • Degree of attachment to one’s initial opinion

  • Structure of the group

How do others’ opinions influence us?

  • In the direction of people we like/who are relevant to us

  • In proportion to the strength (closeness) of the relationship


6
New cards

Network Data Collection

Steps of Network Data Collection (Statistical methods; Visual methods)

  • Defining research topic

  • Defining research question

  • Formulating research hypotheses

  • Deciding on the type of network: ego vs. complete network

  • Questionnaire construction: drafting questions and name generators

  • Recording data: using matrix tables or edge lists

  • Analysis: statistical visualization methods


I. Start

  • What kind of network do we want to collect?

    • Ego network

    • Complete (whole) network

    • Defining the boundaries of the studied group (sample)

      • e.g.: school class, department, village, etc.

II. Questionnaire construction, Questions and Blocks

  • Professional trust

    • Whom do you consider a good professional?

    • Whose opinion do you rely on in professional decision-making situations?

    • From whom do you usually ask for professional help?

  • Personal trust

    • From whom would you ask for personal (private life) advice?

    • Whom would you help with their work?

    • With whom do you usually have lunch?

  • Trust in leadership

    • When the boss goes on holiday, whom do they appoint as deputy?

    • When the boss goes on holiday, whom would you appoint as deputy?

  • Constructing the questionnaire

    • Name generators

    • Listing (how names are elicited)

III. Recording the Data

  • Creating the data-recording file

    • Matrix table — adjacency matrix

    • Edge list


7
New cards

The data can be analyzed on two levels

  1. Software for Social Network Analysis (SNA):

    • Gephi — for visualizing and analyzing networks

    • R, RStudio — visualizations and statistical analyses

    • Python — visualizations and statistical analyses

    • Pajek — for analyzing large networks

    • UCINET — for analyzing small networks

      • Steps of Network Analysis:

        • Statistical analysis of the network (descriptive statistics) = Shape analysis (morphological analysis): examining network indices

          • Network size

          • Degree, out-degree, in-degree

          • Paths (longest, average)

          • Network density

          • Centrality measures

          • Clusters

        • Comparing networks (dyads, triads)

        • Analysing dynamic ahnges (ERGM, positional analysis, SIENA)

        • The Shape of the Network…

  2. Central Actors Centraility

    • Degree Distribution in Social Networks: Main Actors in Networks — Positions:

      • Formal leaders — formal influencers who shape group norms and group behaviour

      • Bridge actors — information and knowledge transfer cannot take place without them

      • Central actors, “opinion leaders” — they become visible in the informal hierarchy, and they’re influencers who shape group norms and group behaviour


8
New cards

The Role of Attributes

  • Elements of statistical analysis in network research — at the individual (and group) level:

    • Independent variables (e.g.: gender, smoking, ethnicity, type of education, place of residence)

  • Groups (Community Detection), Cliques, Clusters


<ul><li><p>Elements of statistical analysis in network research — at the individual <em>(and group)</em> level:</p><ul><li><p>Independent variables <em>(e.g.: gender, smoking, ethnicity, type of education, place of residence)</em></p></li></ul></li><li><p>Groups <em>(Community Detection)</em>, Cliques, Clusters</p></li></ul><p></p>
9
New cards

Analysis of Dyadic and Triadic Relations

  • Dyad: mutual, non-mutual, no tie (3 types)

  • Triad: 16 possible types (based on structural variation, e.g.: mutual, transitive, cyclic, etc.)


<ul><li><p><strong>Dyad:</strong> mutual, non-mutual, no tie <em>(3 types)</em></p></li><li><p><strong>Triad:</strong> 16 possible types <em>(based on structural variation, e.g.: mutual, transitive, cyclic, etc.)</em></p></li></ul><p></p>
10
New cards

Notation of Triad Types: xyzL

  • The triad codes (e.g.: 003, 021C, 300) describe the pattern of relationships among three actors.
    The digits xyz indicate the types of ties:

    • x = number of mutual ties (A ↔ B)

    • y = number of non-mutual (one-way) ties (A ↔ B)

    • z = number of dyads with no tie in either direction

  • L = a letter that distinguishes the exact structure:

    • C = cyclic

    • D = “downward”

    • U = “upward”

    • T = transitive

  • Example: 030C

    • 0 mutual ties

    • 3 one-way ties

    • 0 pairs without a tie

    • C = cyclic: the ties form a cycle — A → B → C → A


11
New cards

Notions of Dyad Types

  • Reciprocity
    The importance of mutual ties:

    • Network theory (Moreno, 1934): Patterns of social relations and their reciprocity.

    • Social exchange theory (Emerson, 1972): Maintaining mutual exchange relations between individuals

    • Game theory (Axelrod, 1984): The role of mutual cooperation and feedback in the stability of social relationships

  • Transivity
    If actor A has a strong tie to both B and C, the probability is high that a tie will also form between B and C.

    • Granovetter (1973, 1983): According to the strength of weak ties theory, weak ties play a key role in the flow of information and resources in social networks.

  • Homophily — Heterophily

    • Homophily: Individuals with similar characteristics tend to connect with each other (e.g.: similar social background, interests).

    • Heterophily: Ties between individuals with different characteristics (e.g.: different nationality, educational background).