Although the distinction between bonding social capital and bridging social capital may immediately seem straightforward, there is an underlying conceptual ambiguity plaguing the current theoretical literature. I have discussed bonding social capital as networks of people who are similar in some important way, and networks of people who typically associate together. This represents two different approaches to conceptualising bonding/bridging:
- Internal – bridging and bonding via socio-economic heterogeneity within organizations
- External – bridging and bonding through interconnections between associations – bridging socio-economic divides might predominantly result from overlapping networks between organizations
Two different ways of conceptualising bonding/bridging
Evaluation of the internal bridging nature of associations involves comparing membership composition of each association to the composition of the overall population on a large number of socio-economic dimensions, for example age, gender, education, income, religion, race, housing status, professional status, occupational classification, marital status, whether one has children.
Assessment of the external bridging potential relies on counting interconnections between associations and correcting this number for the relative size of each association. This gives information about the extent of overlap in networks with a higher degree of overlap representing more bonding social capital.
Let’s explore an example to illustrate the differences. A local Sheffield (UK) cricket team may have players who are socioeconomically similar – they may all work for local steel works, be predominantly white males between 18 and 40 years of age, have similar levels of education and income, and the dominant religion may be Christian. Many of the players may know each other outside of the cricket team, and many may also know other members of their families. They would likely live near each other and have gone to the same schools, attended the same churches, and been members of the same groups or clubs.
This represents a high level of bonding social capital by both approaches.
Contrast this to a local cricket team in Brisbane (Australia) where players would likely come from diverse backgrounds. Some may be university students, others may work in a variety of white and blue collar professions, and some may be unemployed. They may include a variety of ethnic backgrounds and have vastly different socioeconomic characteristics. They would be less likely to know each other outside of the team, and although they may live in the same area higher levels of mobility may mean few of them went to the same school, attended the same church, or were members of the same groups or clubs.
This represents a low level of bonding social capital by both approaches.
In both examples above, regardless of the methodological approach used to define bonding and bridging social capital, we find the same result; high bonding social capital in the first example and low in the second. This is because often heterogeneity of membership and interconnection between associations is related. Likeness of members is often related to the likeness of their associational memberships.
| Bonding social capital | Bridging social capital |
| Within | Between |
| Intra | Inter |
| Exclusive | Inclusive |
| Closed | Open |
| Inward looking | Outward looking |
| “Getting by” | “Getting ahead” |
| Horizontal | Vertical[1] |
| Strong ties | Weak ties |
| People who are alike | People who are different |
| Thick trust | Thin trust |
| Network closure | Structural holes |
| Public-good model | Private-good model |
Binary or continuum?
Generally social capital that is either bonding or bridging describes the nature of a social relationship. If the relationship is with someone like yourself, who moves in similar social circles, then it is described as bonding. If the relationship is with someone different to yourself, who moves in different circles, then it is described as bridging social capital.
Of course, the nature of a relationship is not binary, it is not this or that. Social relationships are far more complicated and typically a relationship will have some characteristics of bonding and some characteristics of bridging. The same is true if we consider social capital at a higher level, as in the cricket team examples. Any network will have some characteristics of bonding and some characteristics of bridging.
The binary nature of the distinction between bonding and bridging risks simplification and reduction in analysis. This is a significant problem if researchers and policy makers assume the two are strictly mutually exclusive.
The distinction is useful in describing social networks and in understanding the function of social capital. We must be careful to clearly define bonding social capital to ensure any empirical analysis is rigorously linked to our theory.
An integrated approach to measurement of bonding social capital
Researchers have developed a method to integrate internal and external approaches to bridging and bonding social capital by combining them into a matrix[2]. This is one way to combine both conceptualisations of bonding/bridging without further simplifying the research context. It may also be possible to create three categorisations rather than a simple binary. This would allow for a ‘middle’ type that is neither distinctly bonding or bridging. This approach would sharpen the distinction between the two extremes of the scale by not unnecessarily, and possibly erroneously, forcing associations in the middle of the ranking into either the bridging or bonding category.
Footnotes
- In practice bridging social capital can be horizontal or vertical. See section on linking social capital for further discussion. ^
- Geys B, Murdoch Z. Measuring the “Bridging” versus “Bonding” Nature of Social Networks: A Proposal for Integrating Existing Measures. Sociology [Internet]. 2010 Jun 18 [cited 2017 Nov 14];44(3):523–40. Available from: http://journals.sagepub.com/doi/10.1177/0038038510362474 ^
