Valente (1996) Social network thresholds in the diffusion of innovations

Valente TW (1996) Social network thresholds in the diffusion of innovations. Social Networks 18:69-89. 

This is a very nice empirical analysis of patterns of innovation adoption.

Personal networks are the sets of direct ties that an individual has within a social system.

Adopter categories are based on innovativeness relative to personal networks.

The present model deviates from past diffusion models by (1) explicitly including the influence of non-adopters on adopter decisions, (2) linking micro- and macro-level influences in one model, and (3) testing the results against data rather than relying on computer simulation.

Network thresholds

• Threshold models of collective behavior argue that individuals have varying thresholds, which are postulated as one cause for varying times-of-adoption and thus as a cause for the S-shaped rate of adoption.
• Granovetter’s (1978) threshold model is based on the premise that thresholds are the proportion of adopters in the social system needed for an individual to adopt an innovation.
• This means that the threshold is a property of the network, not of an individual.  One issue is that individuals may not monitor everyone else in the network, particularly for innovations that are not directly observable. Another difficulty is that some innovations are uncertain, ambiguous, and risky.
• “Thus, adoption thresholds should be measured in terms of direct communication network links with others.”
• Exposure is the proportion of adopters in an individual’s network at a given time.
• The threshold is the exposure at the time of adoption.

Adopter categories

• Early adopters are individuals whose time-of-adoption is greater than one standard deviation earlier than the average time-of-adoption.
• The early and late majorities are individuals whose time-of-adoption is bounded by one standard deviation earlier and later than the average.
• Laggards are those individuals who adopted later than one standard deviation from the mean.
• These are average adoption times relative to the whole society.

Personal network threshold categories may be created by partitioning the network threshold distribution in the same manner described for time-of-adoption adopter categories:

• Very low network threshold individuals have personal network thresholds one standard deviation lower than the average threshold.
• Low and high network threshold individuals have personal network thresholds bounded by one standard deviation less than and greater than average.
• Very high network threshold individuals have personal network thresholds one standard deviation greater than average.
• The average threshold being the mean threshold for the community.

External Influence

In addition to personal networks, people are also influenced by external sources, two of which are (1) cosmopolitan actions (“cosmopolitan” people are attuned to the larger world outside their social system and relate their local social systems to the larger environment) and (2) communication media.

These provide actors with an earlier awareness of an innovation and freedom from system norms, enabling them to be early adopters.

Empirical analysis:

The theory basically states that individuals with more network ties will be the early adopters, and those with fewer network ties will be the later adopters. The personal network threshold categories were assessed based on individual’s network degrees in professional networks in three domains: (1) medical drug adoption among Illinois physicians, (2) hybrid corn adoption among Brazilian farmers, and (3) “family planning methods” adoption among married Korean women.  

External influence was measured by things like self-reported media exposure, visits to a nearby large city, and number of medical journals subscribed to. .

The results show that the categories of time-of-adoption and personal network degree are highly correlated (Table 2).

External influence scores vary for individuals who are innovative relative to the two dimensions considered above.

Table 3 shows that external influence scores were usually highest for individuals who are most innovative relative to the system and their personal network. These are the earliest adopters (innovators), who are the first to adopt the innovation. Their early adoption is associated with high external influence.