Social Capital and Social Networks: developing the research

agenda

Mike Savage & Gindo Tampubolon.

ESRC Centre for Research on Socio-Cultural Change (CRESC) &

Sociology, University of Manchester

Issues we address

1. Why social networks are important to social capital

2. Network dynamics: what a case study of the ties in three organisations in the NW of England tells us about the mechanisms which produce social capital

3. The changing social capital of corporate elites (Gindo)

4. Network analysis using sample surveys: the Cultural Capital and Social Exclusion Survey

An aside: Promoting Social Network Analysis project

Mike Savage, Nick Crossley, John Scott, Gindo Tampubolon, Alan Warde, ESRC funded 2002-2006.

Aims are to create a high profile team to promote SNA which will– Produce research exemplars in the areas of social

capital; elite structures; consumption; innovation– Run training courses and hold conferences – Develop mailing lists and ‘network’– Develop international links– Collaborate with inter-disciplinary research centres

(e.g. CRIC, CRESC) to cross fertilise with research ‘leaders’

1) Why social networks are important to social capital

The social capital agenda offers new potential for SNA because of its leading theorists all emphasise that networks are central.‘features of social organization such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit’ (Putnam 1996: 67). ‘the core idea of social capital theory is that social networks have value….. social capital refers to connections among individuals – social networks and the norms of reciprocity and trustworthiness that arise from them’ (Putnam 2000: 19),

‘social capital is the aggregate of the actual or potential resources which are linked to possession of a durable network of more or less institutionalized relationships of mutual acquaintance and recognition’

(Bourdieu 1996: 51),

Networks and the ambivalence of social capital

• Network methods are also essential in allowing us to explore the positive and negative aspects of social capital. Consider – connectedness– ‘the old boys club’– Cliques– ‘structural holes’ and the potential to build

bridges

Networks are a fundamental part of social capital

• Consider Putnam’s increasing interest ‘informal’ social capital in his more recent work

• Research increasingly doubtful that formal associational membership necessarily conveys significant social capital

• Freitag• Claiborn and Morris

Recent British research

Using the BHPS, Li, Pickles and Savage (2005) constructed three different measures of social capital

- membership of formal organisations- neighbourhood relations- social networks

The analysis showed that these three factors were independent of each other and had different kinds of effects.

Formal organisations were the least effective in generating social trust.

Logistic regression on social trust by types of social capital (top quartile compared to bottom)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Neighbour Network Membership

Model 1

Model 2

Model 3

Current survey questions on networksThese issues explain why there is so much current interest in finding indicators

for social capital include measures for networks

Notably, the ONS Social Capital Question Bank includes questions on ‘Interaction, social networks, social support’ in large surveys

proximity of friends, relativesperceived barriers to contact with friends, relativesHas someone to rely on outside householdReceived practical helpDepth of socialisation networksperceived norms of social supportSocial relations at work

These questions can be asked easily but have several problems:- - uncertain how far perceptions match the actual structure of social support- confuse social capital with its outcomes- Are ‘ego-centred’, and do not allow us to evaluate the structure of networks in

the community at large

We need to think harder about how to use network approaches

The potential of social network analysis SNA examines the ties (networks) between points, using graph theory and other

mathematical models. It has a number of distinct features• Keen to study whole populations, hence allowing systematic analysis of the structure

of roles and relationships.• Interest less in individual attributes, and more in the extent and nature of ties to

others

It has wide applications and its ideas have become social science orthodoxy: most famously

• The strength of weak ties (Granovetter)• The remaking of urban community (Fischer; Wellman)• Corporate elites organised around constellation of interests (Scott)

By way of background UK researchers were key in the early history of social network analysis, especially the LSE/

Manchester anthropologists (Bott, Barnes, Mitchell).

Although there continue to be distinguished SNA researchers in the UK (e.g Everitt, Scott), there is no school of SNA researchers here.

However in North America and parts of Europe SNA has flourished (with specialist journals, conferences, etc), and has produced important contributions to studies of e.g innovation, elite structures, social movements, health, family and community.

There are still significant limits to applying SNA to study social capital. Why?

• Limits of network questions on sample surveys, and the problem of sample surveys themselves.

• Although it often uses case studies, its’ reliance on formal methods for mapping networks, means it is not a standard ‘qualitative’ method.

• Need for specialist software (e.g. UCINET, PAJEK)• Lack of research exemplars showing its potential for

addressing substantive research questions. • Lack of UK centres of excellence and possibility of

research training.

2) Case study of social movements in NW England

Our aims were to examine the structure of social ties amongst all the members of three local organisations: a branch of Labour Party; conservation group; a branch of an ecological Group, 1999-2000.

Three phase research: (a) postal questionnaire; (b) in-depth interview and © life histories

Obtains information on all the members of these groups, and their ties to others across different dimensions

1

2

3

4

5

6

78

9

10

11

12

13

14

15

16

1718

19

20

21

2223

24

25

26272829303132

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49 50

51

52 5354

5556

57

58

5960

61

62

6364

65

6667

6869

70

7172737475

76

77

78

7980

81 82

83

8485

86

87

88

89

90

91

92

9394

95

96

97

98

99

100

101102

103

104

Pajek

Note Respondents were asked ‘with whom do you discuss things to do with the organisation (for example, activities, issues, strategy)’.

Communication networks within the conservation group

1

2

3

4

5

6

7

8

9

10

11

12

13

1415

16

17

1819

20

212223

24

2526

2728

29

30

31

32

33 34

3536

3738

3940

4142

43

44

45

4647

4849

5051

52 53

5455

56

57

58

59

60

61

6263

64

65

6667

68

69

70717273747576

77

787980

8182

83

84

85

8687

8889

90

91

9293

949596979899

Pajek

Note Respondents were asked ‘with whom do you discuss things to do with the organisation (for example, activities,

issues, strategy)’.

Communication networks within the environmental group

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

1819

20

21

22

23 2425

26

27

28

2930

31

32

33

Pajek

Implications

We can use SNA to question important aspects of the arguments about social capital. SNA allows us to map social capital mechanisms more directly:

1. Cores become less trusting2. Different models of organising cores3. Cores find different methods of reproducing

themselves

Social capital is generated around tensions and fractures, rather than consensus and community

4) The position generator and social networks

We have seen some difficulties in using sample surveys to capture aspects of social networks

1. The extent and structure of ties between whole (or quasi-whole) populations

2. The extent to which relationships are reciprocal.

Yet it will not always be feasible to conduct research on whole populations. In the UK and Canada, but not, hitherto, in the UK, an alternative ‘position generator’ question has been developed to rectify this problem.

Cultural Capital and Social Exclusion Project

• Directed by Tony Bennett, Elizabeth Silva, Mike Savage and Alan Warde

• Seeks to develop new measures for cultural, economic and social capital.

• Contains national survey of 1564 respondents, 28 focus groups, and 60 in-depth interviews.

• Early papers on CRESC website

The position generator question` On this card is a list of jobs. Please tell me whether you happen to

know anyone socially who has any of these jobs? Please include friends and relatives

1. Secretary2. Solicitor3. Clerical officer in national or local government4. Bus or coach driver5. Bank of building society manager6. Factory worker7. University/ college lecturer8. Electrician9. Nurse10. Sales or shop assistant11. Postal worker

01

02

03

04

0P

erc

ent

0 2 4 6Number of memberships in civic organisations

05

10

15

Pe

rcen

t

0 2 4 6 8 10Range of social networks

Figure 2 Number of civic memberships and social contactsSource: Cultural Capital and Social Exclusion Survey (2003) (same below).

Range of contacts vary by social group

Age 18-35: mean of 3.59: 36-60: 4.43, 61+: 2.70

Gender Male 3.92: Female 3.61

Class Professional and managerial 4.34: intermediate 3.87: working class, 3.11

Regression on social trust

• Civic membership has coefficient of .174** with social trust

• Those with contact with have co-efficientsprofessional and managerial .247*

intermediate .120working class - .269*

• Those who only know professionals and managers are the most trusting of all (1.067**)

The nature of social contacts thus makes important differences to trust and reveal how trust is the result of privileged social location

ConclusionsSocial networks offer a different approach to social capital that

questions some of the ‘benevolent’ interpretations

Trust and belonging may arise from tension and conflictSocial capital is not evenly distributed, and middle class networks are most effective in

conveying its advantagesSocial capital does not depend primarily on associational membership. We need to

understand the dynamics of associations to assess if and how they generate social capital

Position generator question can be useful in allowing the conflictual nature of social capital to be registered on sample surveys.

Social capital is not a ‘magic bullet’. It is a resource which is contested and which helps reproduce unequal social relationships, though it can also be a resource for those who contest these inequalities