Diffusion of innovations

From Citizendium
Jump to: navigation, search
This article is developing and not approved.
Main Article
Related Articles  [?]
Bibliography  [?]
External Links  [?]
Citable Version  [?]
This editable Main Article is under development and subject to a disclaimer.


Diffusion Theory

Diffusion theory is a model that explains how innovation is spread and adopted throughout a society. Adopters of innovation are categorized as innovators, early adopters, early majority, late majority and laggards. The Diffusion of Innovation model was developed by Everett M. Rogers,[1] and uses an S-curve to graph the adoption of an innovation. A good example of diffusion theory is represented in technology lifecycles. Diffusion theory is used in many disciplines to explain trends, economic patterns, health and medical concerns and technology innovations. This model is an important part of change management and contains four key elements:
  1. What is the innovation?
  2. How is it communicated?
  3. The idea is transmitted over time
  4. The idea is diffused to members of the society

Adopters of innovation experience five stages of diffusion:

1. Knowledge - awareness of the idea and perceived benefit
Awareness Knowledge
What is the innovation? This represents information that an innovation may exist. May motivate an individual to seek a second and third type of knowledge. Most change agents focus on the awareness-knowledge. Benefits could be seen by focusing energy on the how-to knowledge. Change agents consider principles-knowledge to be outside their responsibilities and is more a task for formal education.
How-to knowledge
How does it work? Mainly information necessary to use the innovation properly. The level of complexity will determine how much how-to knowledge is necessary. Inadequate levels of how-to knowledge prior to trial or adoption, rejection and discontinuance are likely to occur.
Deals with the functioning principles underlying how an innovation works. It is possible to adopt an innovation without a deep understanding of principles-knowledge, but the danger of misusing a new idea is greater and discontinuance may result.
2. Persuasion – convinced of the value of the innovation
At this stage one's peers can have a great deal of influence. During this phase a person might attempt to envision themselves using the innovation.
3. Decision – judgment to adopt the innovation
Samples or trials at this phase can have a positive effect on the chances of the innovation being adopted. Demonstrations or use by peers can also weigh heavily in this stage.
Rejecting an innovation after deciding to use it is called discontinuance. There are two types of discontinuance:
Active: Considering, but never actually using the innovation
Passive: Never really considering the innovation
4. Implementation – enacting the innovation
Normally this will happen soon after the decision process. Organizational decisions can slow this stage down considerably. An established organization might be resistant to the innovation rather than supportive.
5. Confirmation - acceptance or rejection of the innovation
The process is not completed and the user of the innovation may search for confirmation for implementation of the innovation. If there is dissonance at this stage, people will seek to eliminate or reduce it.
Discontinuance can also occur when a previously accepted innovation is:
Replacement:The innovation is dropped for a different innovation. Such as replacing steam engine locomotives with diesel powered engines.
Disenchantment:The innovation is not performing to expectations or is performing poorly.

The Innovation

An innovation is an object or process perceived as new by one who is considering adoption. Over time, information about the innovation is created and shared in a social system through a process called diffusion. There are five main attributes which influence an innovation's success in a social system.

Innovation Attributes

Relative Advantage – level of improvement upon previous object or process

Compatibility – degree to which an innovation aligns with current standards, previous experience, and qualities required of potential adopters

Complexity – level of comprehension required to use the innovation

Trialability – degree to which an adopter can experiment with the innovation prior to adoption

Observability – extent to which an innovation’s results are visible to others

New innovations follow a pattern of development. The steps include identifying a need or problem, researching and developing a solution, commercialization, and diffusion of the innovation. The consequences that follow the adoption of an innovation are discussed at the end of this article.

Communication Channels

Communication Channels by Stages of the Innovation-Decision Process

Communication channels are described as:

  1. Interpersonal vs. Mass Media
  2. Localite vs. Cosmopolite

Mass Media vs. Interpersonal Channels

Mass media channels are key during the knowledge stage. Interpersonal channels are crucial during the persuasion stage. The level of advertising can have an impact on the importance of the mass media channel. Interpersonal channels can be very powerful in specific cases. Viral marketing is now a sought after phenomenon in advertising where interpersonal or social channels are used to communicate an innovation.[2] These can be highly profitable or destructive depending on the perception.

Cosmopolite vs. Localite Channels

Cosmopolite channels are essential during the knowledge stage while localite channels tend to be important at the persuasion stage in the innovation-decision process. Cosmopolite channels are links to outside the social structure. Mass Media for example is generally always cosmopilite. Interpersonal channels can be either cosmopilite or local.

Bass Forecasting Model

A model that is used to predict how a product will be received by consumers. This model has been in use since the late 1960's and was developed by Frank M. Bass to evaluate how quickly a product will be adopted by the public. It is also useful for predicting demand.

Communication Channels by Adopter Categories

Earlier adopters find mass media channels more important than late adopters do. They depend more on mass media than they do on interpersonal channels to learn about innovations. Earl adopters also depend more on cosmopolite channels than localite channels.

How the Internet is Changing the Innovation-Decision Process

For certain innovations, diffusion via the Internet greatly speeds up an innovation’s rate of adoption. YouTube, blogging, podcasting, and VOIP are several examples. This can lead to overwhelming the capabilities of some companies to deal with a great infusion of interest. There is a phenomenon known as the DIGG effect when a web site is featured, the site becomes so popular that its servers crash. The upside is when a company is prepared, there are immediate and enormous opportunities.


In medicine, 18 years is the typical translation lag between first description and earliest highly cited article.[3]

The S-Shaped Curve of Adoption and Normality

Time is the key to the diffusion process. The S-Shaped Curve of Adoption and Normality A normal adoption typically follows a bell-shaped curve. If the cumulative number of adaptors is plotted, the result will be an S-Shaped curve. The S-Shaped curve will start slowly, accelerate until half of the individuals in system have adopted, and then slow down again.

The Innovation-Decision Period

The length of time taken for an organization or individual to complete the innovation-decision process

Rate of Awareness-Knowledge and Rate of Adoption

Creating knowledge at an early date through communication can hasten the adoption period.

Length of Period by Adopter Category

Innovators see new ideas positively, thus they are quicker to adopt them. The shorter innovation-decision period because

  • they have more accurate sources
  • regard their sources as being very important and value them highly

Innovativeness and Adopter Categories

1. Innovators – venturesome (2.5%)

Innovators are often members of a clique of innovators and are often geographically distanced from other innovators. They often have substantial financial resources and possess complex technical knowledge.

2. Early Adopters – respect (13.5%)

Early adopters typically possess the highest degree of opinion leadership. They are more of an integrated part of a local system and assist in triggering the critical mass.

3. Early Majority – deliberate (34%)

Members of the early majority take longer amounts of time to adopt new innovations. They seldom hold high positions of opinion leadership and often need the pressure of peers to adopt.

4. Late Majority – skeptical (34%)

The late majority usually adopts when the innovation becomes an economic necessity. They approach innovations with skepticism and caution and need a great deal of peer pressure to adopt. Typically, scarce resources means the late majority must have little uncertainty about an innovation before adoption.

5. Laggards – traditional (16%)

Laggards take a traditional point of view about innovations and are suspicious of changes, instead looking to the past. Laggards are not socially active, possess nearly no opinion leadership, and have few resources to make it easy to embrace change.

Characteristics of Adopter Categories

Socioeconomic Characteristics

Early adopters are typically better educated, have a higher social status, have a higher degree of upward social mobility, and have disposable income. Age is not usually a factor. Personality Variables Early adopters generally have more influence on others, may be less dogmatic, have a greater ability to deal with abstractions, have great rationality, more intelligent, posses a favorable attitude about science and change, are less fatalistic, and have higher aspirations. Communication Behavior Early adopters usually are more socially engaged, are more connected through interpersonal networks, more cosmopolite, posses more interactions with change agents, have greater exposure to mass media and interpersonal communication channels, seek out information about innovations, and have a higher degree of opinion leadership.

Rate of Adoption

Calculation of the number of individuals who adopt an innovation in a certain amount of time is known as rate of adoption. Variation in this rate is highly reliant on the five attributes discussed in Section 2. Other factors which can contribute to this rate are the type of innovation-decision, type of communication channel, nature of the social system, and the change agent’s efforts during the diffusion.

The Social System

Opinion Leaders

Opinion Leaders are individuals who lead in influencing the opinions of others. Their behavior is important in determining the rate of adoption of an innovation in a system.

The Hypodermic Needle Model

The Hypodermic Needle Model suggested that mass media had direct, immediate, and powerful effects on changing the behavior of large numbers of people. Examples included were the Spanish-American War, Joseph Goebbels, and Madison Avenue. The model was ultimately viewed as too simplistic, mechanistic, and too gross to accurately measure the effects of mass media. It also ignored the role of opinion leaders.

The Two-Step Flow Model

The Two-Step Flow Model suggests that communication flows from a source, via mass media channels, to opinion leaders, who in turn pass it on to followers. A study of the 1940 presidential election in Erie County, Ohio discounted the effects of mass media in voting decisions. The first step in this model is a transfer of information from the media to an opinion leader. This is followed by a spread of interpersonal influence. This model does not tell us enough. There are more than just two steps in the flow of communication, but the model focused communication study upon the study of opinion leadership.

Homophily and Heterophily

Homophily is the degree to which pairs of individuals who communicate are similar. The similarities can be things like beliefs, education, socioeconomic status, etc. Communication is generally more effective when source and receiver are homophilous. Heterophily is the degree to which pairs of individuals who interact are different in certain attributes. Heterophilous networks often connect two cliques, thus spanning two sets of socially dissimilar individuals in a system. Homophily accelerates the diffusion process, but limits the spread to those who are closely connected. Ultimately, the diffusion process can occur only through communication links that are at least somewhat heterophilous.

Measuring Opinion Leadership and Network Links

  1. The sociometric method consists of asking people whom they sought for information or advice about a given topic. This is usually a highly valid measure of opinion leadership. A large number of respondents is needed to ensure validity.
  2. An alternative to using sociometry to identify opinion leaders is to ask key informants who are especially knowledgeable about the networks in a system. This technique is almost as accurate as the sociometric method, but requires fewer respondents.
  3. The self-designating technique allows respondents to indicate the degree to which others in the system regard them as influential. This method obviously depends on the accuracy with which respondents can identify and report their images.
  4. Opinion leadership can be measured by observation, in which an investigator identifies and records the communication behavior in a system. This method is typically very accurate, but it can be obtrusive. Also, the members of a system who are being observed might act differently because they know they are being observed. This method has not been used often.

When two or three types of opinion leadership measurement have be utilized with the same respondents, positive correlations among the measures have been obtained, although they are not all in total agreement. All four methods have shown to be about equally valid.

Monomorphic and Polymorphic Opinion Leadership

Monomorphism is the degree to which an individual acts as an opinion leader for a single topic.

Polymorphism is the degree to which an individual acts as an opinion leader for a variety of topics. The degree of polymorphic opinion leadership seems to vary with such factors as the diversity of topics which are measured.

Most studies report, more or less, polymorphism to be more prevalent.

Communication Network Analysis

A communication network consists of interconnected individuals who are linked by patterned flows of information. Networks have a certain degree of structure or stability. This patterned aspect of networks provides predictability to human behavior. The communication structure consists of the differentiated elements that can be recognized in the patterned communication flows in a system. This communication structure is so complex that, except in very small systems, most members are not even aware of the overall structure of which they are a part.

In order to calculate the number of possible network links, we use the formula N(N-1)/2 where N is the number of individuals in a system.

  • Social distance refers to the degree to which an individual perceives a lack of intimacy with individuals who differ in socioeconomic status, ethnicity, and other variables.
  • Communication proximity is the degree to which two linked individuals in a network have personal communication networks that overlap.
  • A personal communication network consists of the individuals who are linked by patterned communication flows to a given individual.
  • Some personal networks consist of a set of individuals, all of whom interact with each other; these are interlocking personal networks.
  • In contrast, a radial personal network consists of a set of individuals who are linked to a focal individual but do not interact with each other.

The Strength-of-Weak-Ties Theory

An individual’s close friends or colleagues seldom know much than the individual does not also know. Such an ingrown system is not effective for catching new information. Weak ties are often referred to as bridge links, which are defined as an individual who connects two or more cliques.

Social Learning Theory

The central idea of social learning theory is that one individual learns from another by means of observational modeling. The observer’s behavior is not exactly the same as the model’s, which would be simple imitation of blind mimicry. Rather, in social modeling the observer extracts the essential elements from an observed behavior pattern in order to perform a similar behavior.

Critical Mass

The notion of critical mass originated in physics, where it was defined as the amount of radioactive material necessary to produce a nuclear reaction. For us, we will define critical mass as the point after which further diffusion becomes self-sustaining. Rogers presents telephones, fax machines, and the Internet as good examples (2003, pp. 343-348).

Notes for a history of the critical mass model

Another concept related to critical mass is network externalities, which is defined as the quality of certain goods and services such that they become available to a user as the number of users increases. The growth of the Internet was slowed in the early stages of diffusion by this.

Cell phones did not suffer this problem since they were connected to the large existing number of land line phones.

Critical mass can also affect the discontinuance of an interactive innovation. Discontinuance of an interactive innovation by one individual may lead eventually to a critical mass of discontinuers, and then perhaps to complete rejection. For example, if an individual stops using E-mail and others follow suit.

All adopters are not equal in their influence in a critical mass situation. A small number of highly influential individuals who adopt a new idea may represent a much stronger critical mass than an equally sized number of individual adopters who have little influence.

Individual Thresholds for Adoption

A threshold is the number of individuals who must be engaged in an activity before a given individual will join that activity. Note that threshold occurs at the individual level, whereas critical mass operates at the system level.

Why Do Individuals Adopt Prior to Critical Mass?

Innovators who rush to adopt an innovation first have a very low threshold for adoption, due to their high degree of venturesomeness. Later adopters have higher thresholds. Threshold levels are normally distributed, thus creating the S-curve of diffusion.

Networks and the Turbocharger Effect

This is the additional variance in a dependent variable well beyond the direct effects of the individual-level variables. Adding network variables to individual-level variables such as education and socioeconomic status can help to explain an increase in diffusion.

Change Agents

A change agent attempts to influence others in a desirable direction as deemed by a change agency. Often, a change agent attempts to have others adopt a new idea. A change agent may also attempt to slow down or prevent the adoption of certain ideas.


Targeting is the practice of customizing the design and delivery of a communication program based on an intended audience.


Tailoring occurs when communication is directed to individuals of a very homogenous audience.


The degree to which an intervention program fulfills its objectives is effectiveness.


Uniqueness is the degree to which an audience of relatively homophilous individuals differs from larger population groups.

Change Agents as Linkers

Change agents main role is to facilitate the flow of an innovation from a change agency to an audience of clients. Often, change agents are teachers, consultants, public health workers, agricultural extension agents, development workers, or salespeople.

The Sequence of Change Agent Roles

1. To develop a need for change.
2. To establish an information exchange relationship.
3. To diagnose problems.
4. To create an intent to change in the client.
5. To translate an intent into action.
6. To stabilize adoption and prevent discontinuance.
7. To achieve a terminal relationship.

Factors in Change Agent Success

Change agents’ success is positively related to change agent effort until opinion leaders are able to meet critical mass. Change agents are half way in between the change agency and their client. Their success depends on client orientation and not on change agency orientation. Change agents must be compatible with their clients’ needs. However, change agents must put clients’ needs before their role in the change process. Change agents must have empathy towards their clients.

Communication Campaigns

In order to be effective, communication campaigns must: 1. :Reach the intended audience. 2. :Set specific but reasonable goals. 3. :Use audience segmentation by dividing a heterogeneous audience into a mostly homogenous segment. 4. :Trigger an interpersonal network by using mass media.

Homophily and Change Agent Contact

Change agents typically differ from clients in many ways, but usually have the most contact with clients that are similar, or homophilous, to them.

Para-Professional Aides

Para-professional Aides are change agents who frequently have contact with clients but are not full professional change agents. Benefits of using aides include a much lower cost, more acceptability by clients, fewer social differences with clients, and can be gender specific (especially important with medical change agents).

Change Agent Credibility

Para-professional aides may have more safety credibility (more trustworthy). Professional change agents have more competence credibility (better educated, more experience).

The Use of Opinion Leaders

The degree to which an individual is able to influence other individuals’ attitudes is opinion leadership. These individuals greatly increase the rate of diffusion.

The Role of Demonstrations

Observing an innovation can greatly help clients to evaluate the effectiveness of an innovation. An innovation can be conducted under field conditions (experimental demonstration), or can be conducted to help an innovation to diffuse quicker (exemplary demonstration).

Centralized and Decentralized Diffusion Systems

Centralized diffusion systems occur when an innovation originates from an expert source, such as from a research and development organization. Decentralized diffusion systems occur when an innovation comes from an operational level. In this system, the innovation is widely shared and adopters may serve as their own change agent.

Types of Innovation-Decisions

The choice to adopt or reject an innovation can be made in the following ways:

Optional Innovation-Decision – made on an individual basis, not dependent on decisions of other system members

Collective Innovation-Decision – made in agreement with other system members

Authority Innovation-Decision – made by a relatively small number of system members who are in power, have high social status, or offer technical expertise

*Contingent Innovation-Decision – made after, but in relation to, another innovation-decision (For example, a teacher deciding to use school laptops in the classroom, which is an optional innovation-decision, can only happen after the school has purchased the equipment for use, which is an authority innovation-decision.)

Consequences of Diffusion

Adoption of an innovation usually produces both desirable and undesirable changes. If these changes are a direct reaction of the innovation, they are considered direct consequences. Indirect consequences are changes that occur due to the direct consequences. While some of these consequences can be predicted, or anticipated, there are other unanticipated consequences that members of a system can not foresee, and might not understand. The main goal of change agents is to create a state of dynamic equilibrium, which is the ability of a social system to cope with the amount of change that is occurring.

Measuring diffusion

In healthcare, diffusion has been measured by monitoring published recommendations by experts[4][5]

Diffusion by consumers and patients have been measured by monitoring sales[6] and claims[7] for prescriptions. Surveys about usage are available as well.[8]

References & Links

  1. Rudd, Rima E. (2004). “Diffusion of Innovation”, Norman B. Anderson: Encyclopedia of Health and Behavior. Thousand Oaks, CA: Sage, 248-251. ISBN 978-0-7619-2360-2. OCLC 370099072. 
  2. Berger, Jonah; Raghuram Iyengar (October 2013). "Communication Channels and Word of Mouth: How the Medium Shapes the Message". Journal of Consumer Research 40. DOI:10.1086/671345. ISSN 0093-5301. Research Blogging.
  3. Contopoulos-Ioannidis DG, Alexiou GA, Gouvias TC, Ioannidis JP. Medicine. Life cycle of translational research for medical interventions. Science. 2008 Sep 5;321(5894):1298-9. PMID 18772421
  4. Tatsioni A, Bonitsis NG, Ioannidis JP (December 2007). "Persistence of contradicted claims in the literature". JAMA 298 (21): 2517–26. DOI:10.1001/jama.298.21.2517. PMID 18056905. Research Blogging.
  5. Antman EM, Lau J, Kupelnick B, Mosteller F, Chalmers TC (July 1992). "A comparison of results of meta-analyses of randomized control trials and recommendations of clinical experts. Treatments for myocardial infarction". JAMA 268 (2): 240–8. PMID 1535110[e]
  6. Tilburt JC, Emanuel EJ, Miller FG (September 2008). "Does the evidence make a difference in consumer behavior? Sales of supplements before and after publication of negative research results". J Gen Intern Med 23 (9): 1495–8. DOI:10.1007/s11606-008-0704-z. PMID 18618194. PMC 2518024. Research Blogging.
  7. Brunt ME, Murray MD, Hui SL, Kesterson J, Perkins AJ, Tierney WM (February 2003). "Mass media release of medical research results: an analysis of antihypertensive drug prescribing in the aftermath of the calcium channel blocker scare of March 1995". J Gen Intern Med 18 (2): 84–94. PMID 12542582. PMC 1494819[e]
  8. CDC. NHANES http://www.cdc.gov/nchs/nhanes.htm