In this Enablers of change episode we’re going to revisit the Diffusion of innovations theory and explore some of the criticisms it’s received over time. We covered what the theory was in an earlier episode, so if you’re not familiar with the theory, it would be a good idea to start there! In that episode we explored some of the research that helped Rogers describe the diffusion of innovations. But we did say that we needed to spend some time reviewing it, and this is the one where we do this!
Roger’s Diffusion of innovations model has been applied in over 5000 studies. It has been well accepted and used in agricultural extension, certainly up to the 1980s when the transfer of technology approach was still prevalent. But even with this success, there have been a number of criticisms. In fact, Rogers himself believed that ‘the progress of a scientific field is helped by realization of its own assumptions, biases, and weaknesses’ (Rogers 2003, p. 106) and so included a section in his books covering criticisms of the theory. It’s good to see this kind of transparency!
The four main categories of criticisms are pro-innovation bias, individual-blame bias, recall problem, and the issue of equality. Let’s now work through each of these.
The pro-innovation bias implies that it’s beneficial for an innovation to be diffused and adopted by all farmers. In reality though, farmers should only adopt innovations that are appropriate to their context. Just look at how land clearing was actively promoted to the early settlers in Australia and New Zealand. In fact, at one time if you did not clear enough land, the government could take your land back from you. But now governments have outlawed excessive land clearing and farmers are being taken to court if they do not comply. So what is perceived as beneficial can change over time.
This pro-innovation bias from researchers and advisers is particularly directed at the laggards—who Frank Vanclay in one of his articles described tongue-in-cheek as those ‘recalcitrant farmers who refuse to adopt the new techniques that scientists consider are appropriate’ (Vanclay 1992, p. 47). While we’re talking about laggards, let’s talk about the use of that term—laggards. It seems an emotionally laden word and quite derogatory. Rogers himself admitted that laggard ‘might sound like a bad name’ (Rogers, 2003, p. 285). He pointed out that the term laggard is not meant to be negative, and suggested that the real issue may not lie with the individual, but with the environment, processes, and system in which the individual operates. We are not sure that this makes the person who has just been labeled a laggard feel any better! We’ve heard others awkwardly try to avoid using the word, instead replacing it with ‘slow coaches’ or other polite synonyms, but at the end of the day, they all mean the same.
A funny story for you… Everett Roger’s own father may have been a laggard! His father loved electro-mechanical farm innovations, but he was highly resistant to biological and chemical innovations. His father resisted adopting the new hybrid corn, even though it yielded 25% more crop and was resistant to drought. It was sadly during the second Iowa drought of 1936, that the crop on his father’s farm wilted. Only then did he adopt the new-fangled innovation—the hybrid corn. Technically speaking though, he was probably a member of the Late majority, and not really a laggard, even though that might make the better punchline.
Now to the second criticism: the individual-blame bias. This occurs when the blame for non-adoption is put on the farmers. Whereas sometimes it is the fault of the extension practitioner for poor or unclear communication of the innovation or just that the innovation is not appropriate. Instead of considering the whole system and the possible influence of external factors, blame is attributed to the individual. This is just not fair!
The third criticism is around the recall problem. This occurs when adopters of an innovation cannot accurately recall the exact time they adopted the innovation, let alone when they may have first heard about it. This then affects the accuracy of the adoption related data and the diffusion curve. However, some research by Mayer et al. 1990 indicated that people can accurately recall details about an event and how they heard about it several weeks afterwards. So we’re not sure whether this criticism is as important as the others, but it’s definitely worth highlighting!
The fourth and final major criticism, as listed by Rogers, is the issue of equality. This relates to the socio-economic benefits of an innovation not being equally distributed through a given population and the consequent widening of the socio-economic gap due to the adoption of the innovation. This is possibly the most difficult of the criticisms. Researchers such as Bordenave (1976) suggested that we should not ask the typical research-style questions. Instead a much broader set of questions could be used to provide a better context to adoption of the innovation, and can highlight some of these issues. The neglect of socio-political factors by extension practitioners, particularly in developing countries, has recently been highlighted by Cook et al. (2021), in their article Humanising agricultural extension.
It is not only Rogers that has criticised the diffusion of innovation theory! Others have criticised the linearity of the model. Some researchers suggest that innovation diffusion is an unstructured, emergent phenomenon, whilst others acknowledge that these days we are operating in an Agricultural Innovation System, which of course is anything but linear!
Geoff Kaine, an Australian agricultural economist now living in New Zealand, cautions against using Roger’s adoption attributes to estimate the number of potential adopters. His studies in 2008 showed this often leads to overestimating the size of the potential population of adopters, and underestimating the actual level of uptake. He also proposed the relative advantage of an innovation is the key factor limiting the adoption of innovations in agricultural contexts (Kaine et al. 2011).
Another criticism is that many people misinterpret the Diffusion of innovations theory and think that innovativeness is a personal characteristic, yet adopters will often fall into different categories for different innovations. So while being a laggard for one innovation, the same adopter may be an early adopter for a different innovation, just like Roger’s father!
There is also an issue with the possible negative impacts from the use of the theory, as raised by Garry Stephenson in 2003. For example, if all farmers grew the same variety of hybrid corn, this mono-culture situation could exacerbate a pest or disease, wiping out all the crops in the entire region, possibly leading to food shortages.
Another challenge to the Diffusion of innovations theory comes from Geoffrey Moore who wrote the book Crossing the chasm (2002). He suggests that the adoption curve is not continuous for discontinuous technologies, where adopters are required to substantially change their behaviour or to modify other products and services they use. Rather, there are gaps between each of the segments, and a rather large one referred to as a ‘chasm’ between Early adopters and the Early majority, as shown in this figure from his book.
This implies there are two separate markets for the product on either side of the chasm. Moore says that leading edge adopters are looking for a competitive advantage (for example, lower product costs and faster time to market). They accept this will involve the pain of changing from the old, established ways to the new, improved ways. They accept that there will be bugs and glitches involved in the change, but they do it to gain the business advantage. Early adopters are willing to suffer the inconvenience and high cost to get that business advantage.
In contrast, the early majority are seeking productivity improvement for existing operations. They want ‘evolution, not revolution’ (Moore 2002, p. 20), and expect the technology to work with their existing systems without any glitches. While the early majority look to other users to validate their purchasing decisions, they do not consider early adopters as being similar enough to them to rely on what they think! This is a catch-22 situation, where the early majority only respect the opinion of other early majority members. It means it is difficult for many innovations to survive past the chasm.
So let’s sum up. The original Diffusion of innovations theory described by Rogers in 1962 has been widely used in agriculture and beyond. We’ve outlined some of the criticisms of it, from highlighting the pro-innovation bias to the discontinuous nature of some innovations. But the bottom line is that the theory hasn’t been disproven. In fact, other models have been built upon the foundation of the diffusion theory, such as the United Theory of Acceptance and Use of Technology ( UTAUT) developed by Venkatesh et al. (2003).
What are the conclusions for enablers of change? We think given that the Diffusion of innovations theory is so widely used that it’s important to understand it! But it is equally important if you’re using it, to read up on these criticisms, acknowledge potential biases and try to make sure you are making an effort to discuss their impact in your context or at the very least acknowledge that they might be affecting how you are applying the theory.
This has been a long episode but you’ve now read our thoughts on the criticisms of the Diffusion of innovations theory! Now we’d like to hear yours. Add a comment below the blog post and tell us what you think. Have you used this theory? How have you explored the biases and criticisms of the theory? Any tips? We don’t want this to be just a one-way conversation—join in by sharing your thoughts and ideas with us!
Thanks folks for reading this Enablers of change post. Remember to subscribe to our newsletter if you’d like to know when new episodes are available. And if you liked what you heard, please tell your friends so they can join the conversation!
Bordenave, J. D. (1976). Communication of Agricultural Innovations in Latin America: The Need for New Models. Communication Research, 3(2), 135-154.
Cook, B. R., Satizábal, P., & Curnow, J. (2021). Humanising agricultural extension: A review. World Development, 140, 105337. doi:https://doi.org/10.1016/j.worlddev.2020.105337
Kaine, G. (2008). The Adoption of Agricultural Innovations. (Doctoral). University of New England, Armidale.
Kaine, G., Wright, V., Cooksey, R., & Bewsell, D. (2011). Identifying potential adopters of an agricultural innovation. In D. Pannell & F. Vanclay (Eds.), Changing Land Management: Adoption of New Practices by Rural Landholders (pp. 69-86). Australia: CSIRO Publishing.
Mayer, M. E., Gudykunst, W. B., Perrill, N. K., & Merrill, B. D. (1990). A comparison of competing models of the news diffusion process. Western Journal of Communication (includes Communication Reports), 54(1), 113-123.
Moore, G. A. (2002). Crossing the chasm: Marketing and selling high-tech products to mainstream customers (Revised edition ed.). New York: Harper Collins.
Rogers, E. (2003). Diffusion of innovations (5th ed.). New York: Free Press.
Stephenson, G. (2003). The somewhat flawed theoretical foundation of the extension service. Journal of Extension 41(4). Retrieved from http://www.joe.org/joe/2003august/a1.php
Vanclay, F. (1992). Barriers to adoption: a general overview of the issues. Rural Society 2(2), 10-12.
Venkatesh, V., Morris, M., Davis, G., & Davis, F. (2003). User acceptance of information technology: Toward a unified view. MIS Quarterly 27(3), 425-478.