Diffusion of innovations
Jon Roland

Those who support a cause that they are having difficulty selling to the public would benefit from the study of work in a field called "diffusion of innovations". There is a book with that title by Everett Rogers. See this link Also do a search on that phrase in Google.

The initial work was done in the early 1950s at the University of Chicago. It was funded by corporate sponsors who were considering the large sums they would have to spend on national television advertising, and wanted to know how effective such spending might be, and what kinds of advertising would be most effective.

The researchers found that populations tended to divide into distinct groups of adopters: The primary adopters were quick to try and adopt new things. The secondary adopters tended to defer adopting until after enough of the primary adopters had done so and used the innovations for a while. The tertiary adopters tended to defer adoption until a sufficient number of the secondary adopters had tried and used the innovations for a while. Sometimes there was also a group of quaternary adopters, and sometimes a group of holdouts that would never adopt, or even actively oppose the innovation.

It was found that messages like broadcast advertising could accelerate adoption by the primary adopters, but were not sufficient by themselves to get the secondary and lower level groups to adopt. The primary influence, it was found, was among peers within each group, and downward from one group to the next lower group. Examples of satisfactory use are far more effective in winning converts than the kinds of reasoned argument that might be conveyed in broadcast messages. It was also found that emotional messages are far more effective than reasoned ones.

This research also showed the importance of repetition. Except for the early adopters, people generally do not adopt something new based only on a single message or example, no matter how compelling. They exhibit characteristic herd behavior, in which a member only moves in response to its repeated perceptions of the movements of multiple other members of the herd. The timing of the repetitions is also important. Too frequent repetition, or messages that are too intense, can turn the person against the innovation, but too much delay between repetitions can lose much of the effect of previous repetitions.

What works best is a carefully timed series of repetitive messages or examples, neither too mild nor too intense, that incrementally move each person along a path from where he is to where the promoter wants him to go, avoiding sidetracks. To the extent possible, it is best to conserve resources by focusing not on those who have already adopted, or who are not ready to move forward, but on those who have been prepared by previous efforts and are ready to take the next step.

Promoters of political causes should also be cognizant of competitive diffusion processes. There are likely to be multiple innovations in the field that will tend to compete with one another, and indeed, the introduction of an innovation may stimulate the appearance of a competing or opposing innovation. The receptiveness of the population to the competing innovations may differ greatly, so that we can speak of a characteristic "coefficient of diffusion" for an innovation for that population.

Thus, we can describe what happened in Southeast Asia during the Vietnam war as a competitive diffusion process between the innovations, or memes, of Western liberal republicanism, promoted by the United States, and traditional nationalism, promoted by the North Vietnamese. While the former idea won many converts, the latter had a higher coefficient of diffusion at that stage in the local population's cultural development, with most of the population not understanding the idea and identifying it with foreign intruders. The result was that, despite all the efforts of the U.S. forces to "win the hearts and minds" (WHAM) of the people, nationalism prevailed.

On the world stage we can see the opposition to modernization and globalization that the innovations associated with those mega-diffusions provoke, or on the national stage the similar resistance to demands for constitutional compliance from "reliance interests" threatened by such compliance.

It is useful to examine chains of influence leading to decisionmakers. One who seeks some reform must first describe in detail how it would work, and what resources it would require, then identify who makes the decisions to move forward on implementing it. In most cases the promoter will not have direct access to the decisionmaker, but must work his way up one or more chains of influence, beginning with those with whom he is in direct contact and over whom he has the most influence, and working toward the decisionmaker through those who most influence the next, converting each along the way, or perhaps contriving to replace them with others, including the decisionmaker himself. The key to advancing along the chain is to be able to discern what each wants and will respond to, consistent with what the reformer seeks to do. Such an effort will often be far more cost-effective than attempts to influence the general population in an unfocused way, especially that part of the general population that exerts or is likely to exert little if any influence on the chain of influence to the decisionmakers.

Promoters of a cause must also understand that some causes, however meritorious, will not be adopted by the majority of the population until the conditions for adoption are ripe. The ideas of constitutional republican government adopted by the Thirteen English Colonies in the 18th century were not entirely new. Indeed, they had been developing over the previous 2000 years, adopted by a few leading thinkers, but not adopted by a sufficient number of the general population until the right conditions of a frontier allowed them to flourish. Propagating those ideas beyond that initial frontier environment, even with the compelling example of the United States, is by no means assured within the short to mid term future, and may not even survive in the United States now that the frontier conditions that permitted their emergence have faded. If such ideas succeed, it will take a strong effort by many dedicated people to overcome unfavorable conditions.

It must also be understood that more complex causes are more difficult for the mass of people to understand well enough to adopt, and that a complex cause is likely to have a low coefficient of diffusion, no matter its intrinsic merits. A complex cause or proposal is likely to have to be sold not directly, but by appealing to concern about the problem it seeks to solve, and by promoting the proponents of that solution to decisionmaking positions, where they can carry out the details that the mass of people would never adopt no matter how well it might be explained to them. Most people can be persuaded to have confidence in personalities long before they adopt their proposed solutions.

The coefficient of diffusion is generally not constant throughout a population. It can be high for the highest echelons, the early adopters or pre-adopters, low for the secondary adopters, and high again for the tertiary or quaternary, resulting in an innovation being stalled at a certain level, or perhaps stalled at a level for a time, then skipping a level and being rapidly adopted by a lower level. It may also be high for a segment of the population, but low for other segments, resulting in the emergence of factional behavior. The phenomenon can be seen in the emergence of fandom for celebrities, sports teams, makes and models of motor vehicles, and political parties. Adoption networks can form local clusters, or herd behavior, discussed in terms of "evolving complex networks", such as links to websites, citations to professional papers, or citations of judicial precedents. Sometimes this clustering can penetrate a resistant adoption level to establish a beachhead in a lower level.

Sometimes the only early adopters available are people who are not influential with others, who have few or no secondary adopters who look to them for leadership. That is the definition of a niche market.

Perversities can occur as the result of delays in adoption by each adoptive group. There can be successive bell-curved waves of learning and adoption within a supply chain, a wave for research, acceptance by management, manufacturing, testing and refinement, sales, and then by each adoption echelon of consumers of the product. Because the manufacturer may not be able to afford to continue manufacturing a product nearing the end of adoption by consumers, and may need to shift its tooling to its next product lines, there can remain a large group of prospective consumers who are just learning how to adopt and use something that is no longer being made. Thus there can develop an entrenched population of have-nots for many kinds of innovations that never catch up. This can also be a problem for some kinds of government purchasing, retarded by bureaucratic standards that lag far behind the state of the art in a given field.

Sometimes all that the proponents of a complex cause can hope to do is keep it alive, perhaps for hundreds or thousands of years, until the conditions for it to prevail occur. This may require extraordinary conviction and patience.


Contents | Text Version



Popular Pages