In continuing our quest to create, vet, and deploy meaningful innovation, this week we focus our efforts on a very specific and very powerful strategy and tool for our effort: Means-Ends Chains and cognitive mapping.
Equal parts innovation strategy, research validation, and foundational marketing in our use case, Means-Ends Chains (MEC) quite literally can reveal the landscape into which your innovation fits and if it is delivering maximum potential.
While MEC is rather intuitive in concept and people tend to feel that they grasp it easily, it is a practice which can yield significant dividends from diligent effort. It can reveal gaps in thinking between intent and delivery of the concept, which is our overall intent at this point, as we need to know if the concept can potentially fail not because the strategy/brief is unsound, but if somehow that strategy was not translated or delivered to the market.
Much like consumer research, if there is a gap in the offering, we need to be both precise and diagnostic about that gap, and equally precise in filling it.
By the end of this lesson, you should be able to:
To Read | Chapters 13, 14, 15 (Keeley, et al.) Documents and assets as noted/linked in the Lesson (optional) |
---|---|
To Do | MEC & Cognitive Mapping
|
If you have any questions, please send them to my axj153@psu.edu [1] Faculty email. I will check daily to respond. If your question is one that is relevant to the entire class, I may respond to the entire class rather than individually.
For the purposes of this discussion, in the beginning, there was Abraham Maslow. Maslow was an American psychologist who, in 1943, would publish one of the most influential psychological journal articles of the last 100 years: "A Theory of Human Motivation [2]." In this piece, he would lay out his case for the "Hierarchy of Needs," which is commonly referred to as "Maslow's Hierarchy" today.
Maslow's Hierarchy is elegant in its simplicity: It delineates universal human needs, beginning with the most fundamental physiological needs at the bottom, and progressing to the most abstract at the top:
The hierarchy is used in many contexts today to explain any number of psychological phenomena, and it has been adapted for use in everything from management to motivation to marketing. Some of the applications of Maslow's Hierarchy may be a bit of a stretch from his original vision, but to provide some feel for the influence Maslow still holds on psychology and other fields, below is a graph of the use and citation of the more important psychologists in recent history. I have also included "Kardashian" for some frame of reference. (Bear in mind, thankfully, that this chart shows incidence of the terms in books and journal articles.)
Nonetheless, in the scope of meaningful reference and use, Maslow's work is in some fairly-rarified territory:
Why do we care about Maslow's Hierarchy to kick off a Lesson centered on insight-driven innovation? Because in the First World, much of our lives are not spent on the literal act of daily subsistence, but on pursuits higher on the pyramid. So, while Maslow's hierarchy does an excellent job of explaining human needs, many of our efforts are centered around wants.
Most interestingly: Consider how sustainability manifests itself in Maslow's Hierarchy. What you may commonly see is that sustainability has this somewhat odd tendency to function at the bottom-most (Physiological) and top-most (Self-actualization) levels. What I refer to is that even the word "sustainability" refers to that bottom level, addressing long-term viability. This can be either from a human perspective (i.e., addressing the continuation of life as we know it at the most basic level) or the organizational perspective (i.e., addressing the continuation of the organization as we know it at the most basic level).
But, then, as we have addressed in the Clouds and Roots Model, and as we will address further in the Lesson, sustainability is very much about self-actualization for many people and organizations. What I refer to there is that the person or organization is extremely aspirational about sustainability, how sustainability refers to their own personal philosophies, and evaluating how we define and evaluate "success."
What we have is this combination of sustainability operating at the two farthest levels of the Hierarchy, which can make it very difficult to craft a coherent strategy for the innovation we are moving through research and development. This is not a trivial concern, as the strategy shapes the research, which refines the strategy, which drives the messaging.
It's a very interconnected web of the offering we are trying to understand and refine here, and far too often, messaging is considered an afterthought. The effort is centered on "what's in the box," which is then passed over to marketers and merchandisers later. We absolutely can not divorce or defer the strategy and messaging from our research, as quite literally, messaging shapes how customers experience the offering, and how they perceive the offering we are researching. Messaging has so much power that it can quite literally change how the brain processes signals from taste receptors, for example. (More on that later.)
Although Means End Chain thinking has been around for quite some time, it began to receive elevated attention and application in academic journals in the 1980s, with significant contributions by Olson [6], Reynolds [7], and Gutman [8]. Unlike some tools and methodologies which fall into and out of favor, the Means-End Chain (MEC) is a classic model, and one that has grown in popularity and refinement over the years.
The thinking behind the Means End Chain is quite simple: That we, as humans, may think that we "hire" a product or service to do a tangible "job" for us, but that there are very important emotional drivers underlying all of it. It may seem a simple rationalization that we choose one product over another only because of "what's in the box"... but this, at best, is only part of the story.
Ultimately, the most important decisions we make about products and services are the decisions we do not realize we are making, as they are emotional and subconscious. To give you some frame as to how important the subconscious** is to the purchase decision, Gerald Zaltman of Harvard is popularly quoted as stating that around 95% of decision making takes place in the subconscious [9]. (**See "A note on terminology" at the bottom of this page.)
When taken in isolation, the connections between the products we buy and the ultimate end benefits we seek from them may seem quite detached and ridiculous.
For example:
While many innovation development cycles tend to take a heavy focus on product attributes and "what's in the box" early in the development cycle, pushing forward our understanding of the deeper and more emotional associations customers have with a prototype offering can bear significant dividends. From Zaltman [emphasis is mine]:
The insights offered by methods that probe the unconscious mind are relevant at all stages of the product life cycle. For instance, when introducing a radically new product, it is necessary to understand how consumers currently frame their experience of the problem addressed by the new offering. That is, no matter how radical a new product is, it will always be perceived initially in terms of some frame of reference. It is essential that this frame be understood, especially if it is an inappropriate one, detrimental to early trial of the product. For a mature product, insights about the category or a specific brand can lead to modifications that will extend its life and sustain its economic value to the firm. One firm with a very "tired" brand explored consumers' hidden thoughts and feelings and discovered a relevant, basic emotion that had been overlooked by all brands in the category. They were able to connect this emotion with their brand giving it a major sales boost. Other firms use the hidden treasures of the unconscious mind to identify new product opportunities.
There are quite a few different versions of the Means End Chain that have been born and refined by various professors and practitioners, and some with quite a few levels of classification and nuance. These granular models may be well suited to applications which require a very high level of resolution, but, in practice, are likely overcomplicated for our purposes. For this reason, the four-level Means End Chain is more than enough for our purposes, and will allow us ample resolution to understand our concepts.
A few notes on the Means End Chain itself:
Pivot your chair to the left (if in a fixed chair, please turn your head as to not strain yourself).
Find the first thing you have purchased nearby. It may be a window, a book, or some souvenir you bought on your last trip.
Try to write a Means End Chain for that product from your own perspective. If you were lucky enough to choose a lower involvement item (i.e., the air vent on the floor), you might find that creating a MEC is a little hard. Perfectly normal, and you're illustrating a point we'll cover later.
If you hit an impasse, or it doesn't feel right, start with a higher involvement item you bought recently and try another MEC.
What is useful to the interviewer is that much of the Means End Chain may be revealed through effective in-depth interviewing, which we covered in the last Lesson. [10] If you have done a good job interviewing, chances are that you will be able to recreate a significant portion of the interview's thoughts as MEC, from which you can then do further analysis.
While Means End Chains may make perfect sense in a static model, in my experience, much of being able to identify MEC during a live interview (i.e., "out in the wild") depends on:
There are some specific interview techniques referred to as "Laddering" which are designed to specifically elicit the most valuable MEC, which are chains that are robust, descriptive, and complete from Attribute to Emotional End. The following article gives some specific examples of these techniques and what they look like in a live interview.
Read "Laddering Theory, Method, Analysis and Interpretation [11]" by Reynolds and Gutman, paying special attention to the "Laddering Methods" section beginning on page 4 of the document. You will see that they refer to Attributes, Consequences, and Values as the levels in the Means Ends Chain, but the laddering methods themselves are unchanged.
**A note on terminology: Let me elaborate on that "subconscious" point for a moment, and help make sure that we understand the difference between subconscious and unconscious thought. When we say that there are subconscious drivers in decision making, we are not lapsing into some far flung decision of "triggers," "buy buttons" and subliminal messaging, but instead referring to those drivers which we do not actively think about, but can be revealed through research and interview. For example, we see the subconscious at play on those commutes home where we have little recall of the trip, and seemingly just "appear" at home. We did not require some intense amount of cognitive processing during the trip, and we can recall the trip itself in hindsight, and so it is just below our active consciousness: it is subconscious.
While one Means End Chain is certainly an interesting view, we want to be able to understand the experience of the offering across multiple test users to see what is working about the offering, what isn't, and areas of potential interest. To do this, we will combine the Means End Chains of test users onto a single view called a Cognitive Map.
Cognitive maps are called this because they quite literally end up looking like maps, and are designed to illuminate pathways of thinking from one concept to another, as well as highlight those areas which are more commonly traveled.
For the sake of providing a picturable example for us to work with, let's say the offering is a new native grass seed your company has identified which is extremely drought tolerant, low growing, and low maintenance.
At this point, you have interviewed ten users whom you sent the seed to last year, and they have now used and lived with the lawn for a full four seasons. You had done some other, shorter user testing with the grass seed that spanned either a single season, and invited testers to a lawn which had already been planted with the seed to understand their thoughts, and the results had been promising, but it is the full-year test cycle which you believe is most interesting–and crucial–to the potential success of the new seed offering.
We will imagine you are relying not only on your notes taken during the interviews, but also audio recordings (or, even better, transcripts from those recordings).
We will start by writing down the MECs used by one participant, and for simplicity and readability, we'll try to keep attributes near the bottom of the page, and emotional ends near the top. We'll use lines to not only connect the chains, but also between concepts the participant associated so we don't have to write the same attribute multiple times.
Here's an example of what a very simple map from one of those participants could look like. Take a few moments to look at the map without the red text, and then take a look at the layer with red text to see what the participant verbatims might look like for each link.
As you work through mapping, remember that it is the participant's MEC and thoughts we are mapping... NOT yours. So if they make associations that are "wrong" or don't mention something, don't make the associations for them. In the above example, even though there could be a link between "native species" and "grass grows slowly," we didn't add it because the participant didn't make that connection. This is important because by adding in your own knowledge and "over-mapping," you would be skewing the overall picture.
After compiling the cognitive maps for each participant, we have a deep, but still somewhat unpolished view of the information. This is because we will have the knowledge from taking a very rigorous approach to the interviews and understanding their content, but yet we do not yet have a fully formed analysis showing the consensus across participants. This view will be provided by the overall map, which is also sometimes referred to as a "hierarchical value map."
Without specialized programs, the easiest way to create the final map is to "standardize" similar concepts and enter each linkage into Excel. The purpose of standardization is to make sure we don't have two different names for the same concept in different maps (e.g., "Native species" in one map and "Species that are native" in another), and to allow us a final chance to refine or collapse our list of concepts.
After we have the final list of concepts, it is a matter of putting each pair into Excel and keeping a tally of the number of times participants connect the two concepts. For example, we would have "Native species" in Column A and "Doesn't die in drought" in Column B, and we would keep count of the times we see that connection in the individual maps in Column C.
From here, it is a matter of setting a threshold for those concepts and linkages that will make it to the overall map. If we sort the linkages from most common to least common, we would probably look to eliminate the bottom 10% or so from a map covering ten participants, or otherwise to try to eliminate those concepts which are outliers. From here, you are left with a list of the concepts and linkages for the final, overall map, and it is a matter or redrawing the map from this list, perhaps also noting the number of times a linkage or concept was mentioned by participants.
Here is what the overall map from our native turfgrass research could look like, after we have built and formatted it. It can be useful to create two versions of the map, one as a "pure" view with just concepts and links, and the other with line weighting according to how many participants made each connection. I have included both as examples:
While it does take some effort, the overall map is a valuable tool to provide a deep, rigorous, and yet, simple view of our ten test participants and their thoughts and feelings about the native turfgrass. You will find that the overall map also provides a very valid basis from which to compare future revisions of the offering, definitively showing how your revisions have affected the deep perceptions and associations held by customers.
The value of mapping does not end there, as we may also use the map as a strategic framework to drive further ideation, as we will see in our next topic.
After going through the rigor of interviewing early adopters or a test sample of customers, eliciting Means End Chains, and mapping each interview, it could be easy to consider the Overall Map to be a "summary" to capture what was said in interviews. And the tendency can then be to believe that one "has a handle" on the interviews and analysis from the individual interviews, and to therefore skip this final step. That would be a misstep, as you would be leaving an opportunity for significant strategic learning behind:
The process of creating the map will be illuminating, as those involved most in the offering will start to see "what made it" onto the map, and what fell short. You will see concepts that weren't anywhere on the organization's collective radar appearing en masse, and some associations which you had hoped for nowhere to be seen. You will find the participants seem to have almost premeditated levels of attention to some innocuous details, and other concepts will be so in line with strategy that some in your organization will accuse you of witness tampering. This is perfect. Why?
Importantly, you will see opportunities to add concepts or reframe adjacent thoughts and feelings that users already have as opposed to trying to create new thoughts in less-related spaces. Think of refining strategy and messaging as being similar to remodeling a home. You will be presented with the existing blueprint of the home (the Map), and will find that some of the parts of the structure may be more easily adapted or modified than others. For this reason, it is typically more efficient to maximize spaces in the existing structure or add on than it is to build a completely new room detached from the house.
We see this idea represented in the words of Gerald Zaltman from our last topic: "...no matter how radical a new product is, it will always be perceived initially in terms of some frame of reference." Moving highly embedded frames can take tremendous effort and require years of work, but we may be able to reach the same ends using different means, so to speak.
To illustrate these points, here are examples of a few analyses and strategies you can use on your Map:
Despite marketers' and product developers' best efforts, there will be attributes, benefits, and associations that users just didn't "get." Depending on your cutoff threshold, these concepts may make the Map, but at a far lower level than anticipated.
In this case, perhaps you expected that "Well adapted, hardy" and "Don't need to patch, reseed" would have been much more strongly linked. Maybe users just didn't see that as a concern, or maybe these users haven't had problems with die-off before. From there, you could do some quick survey work to understand if the broad market even considers die-off a problem, or if you're trying to market to a problem that doesn't really exist.
Implication: This is more of a "first pass" question to ask yourself to get started, as it tends to be the most obvious analysis.
Sometimes, concepts or associations make the map that would have been completely unexpected to your team, and sometimes concepts you thought would be minor turn out to quite prevalent.
For example, perhaps "Can enjoy surroundings, weather" could fall into that group, as the link you saw as most important may have been straight from "I mow less" to "Save time." But time and time again users would talk about the fact that 'the times when the grass is growing so fast you have to mow every two days are also some of the most beautiful to be outside, especially late Fall and early Spring.'
There is also a prevalent connection from "Questions from neighbors" to "I'm an early adopter, leader" which you may have expected, but not with that level of potency that makes it one of the three strongest associations on the Map.
Needless to say, these are the types of insights and nuances that can mean the difference between a generic and truly resonant strategy.
At this point, we can go beyond general acknowledgement of concepts which may be underweighted and begin to specifically discuss those concepts we will seek to strengthen in the minds of users. The key at this point is to not concern ourselves with the tactics of how we will strengthen the desired concepts, but instead the pure consideration of what we want the Map to look like in the end. Later, we can use this as the framework by which to refine the offering, tune messaging, change the overall use experience, and more.
In this case, perhaps we really want to own the idea of "Can enjoy surroundings, weather" because it is not only strategically potent for us, but it is emotionally potent for users who are generally "outdoorsy" anyway. But, if we look at the weighted Map, all of the paths to that concept are happening at a low level, and furthermore, the connection to the attribute level (i.e., the seed we are selling) is equally weak. So, we may seek to reinforce the path from "Grass grows slowly" to "Can enjoy surroundings, weather," and add a link from there to the already potent "More time for family/kids."
Reinforcing concepts that are already present on the Map–even at a low level–is significantly more efficient and resonant than trying to create entirely new ideas or regions of thought. For this reason, the moves you take to reinforce links tend to be among the most beneficial.
On the other side of the spectrum, when we start actively talking about strategic goals, there will be concepts and links which we wish to minimize or eliminate altogether.
We have something interesting here, and that is a little bit of a duality around "Light green color." About half of the users associated this with "Not as attractive," which is certainly a problem for us. BUT, "Light green color" is also a prime pathway to "Questions from neighbors" (because the grasses do look so different), which then triggers a tree trunk of a link to "I'm an early adopter, leader." If we seek to use this to our advantage, we want to own the light green color, and reinforce it as a point of pride. (For example, did you know that the native grasses of Europe, such as Poa Trivialis [18], are lighter green by nature, and that Europeans overwhelmingly prefer this color over the darker–water and fertilizer hungry–Kentucky Bluegrasses and Perennial Ryes that Americans are conditioned to?)
At this point, we take a little wider strategic view and create pathways, from Attribute to Emotional End, we think are differentiated, ownable, and important to customers. Remember the Clouds and Roots: We need to limit these paths and leverage them as strongly as we can. If users make other associations naturally, that is excellent... but we need to own one or two paths in the minds of customers.
Take a look yourself for other pathways you think could be of interest.
As a final check, we want to look at the Map for any potential areas of WGB space opportunity.
For example, if I really wanted to take a stand against the lawn industry Black Space, I could use "I am a responsible neighbor" as a wedge issue of sorts. The platform could be to turn the "keeping up with the Joneses" curb appeal competitions from a badge of pride into something not to be proud of. I would cut humorous ads showing people manicuring their lawns while their kids sat on the driveway, and make a point of encouraging people to look past those who have the "greenest lawn" and to those who have the safest lawn for my kids to play on.
If I'm thinking White Space, maybe I do a little bit of research into the environmental impacts of a high maintenance lawn, from the carbon footprint to the impacts on water and air quality. The results could be surprising, and could be used to gain support of NGOs.
We will be doing some mapping and strategy work this week, and my hope is to be able to help you see how Maps can be used to drive and refine innovation in an rigorous way.
There is sometimes the belief that to be "pure" research, it cannot be tainted by the hand of messaging or marketing in any fashion. At best, this notion is somewhat idealistic, and ignores the importance of messaging to the product experience itself.
While it may be facile to say that what is in the box and what is on the box are completely separate in the mind of the consumer and are able to be tidily compartmentalized, this is rarely the case. In fact, it can be quite difficult at times to separate messaging from the customer's perception of the use experience. We will cover a body of research that underscores this in a few moments.
So while we are conducting these interviews to understand perceptions of the offering, it can be exceptionally valuable to include very early phase messaging, or "proto-messaging," if you will. Think of the goal of including a proto-message in your interview as trying to provide a very faint pencil sketch of the message, from which the user can add their interpretations, colors, and extensions. This is very much in keeping with our goals for the interviews in the first place, which is for us to say as little as possible, and gently guide participants through their own thoughts. There is one expert in the interview in regard to the use experience, and it isn't you.
The challenge in creating this "sketch", when including this proto-message in the interview, is that one wants to provide the basest of statements to react to. For this reason, we want to purposely avoid imagery, design, or visual stimulus, instead keeping to black text on a white page. In essence, the message should read as somewhat sterile, but not unfriendly, and should avoid buzzwords, brand names, etc.
Here's how we could include a proto-message in our interviews on the native grass experience:
In the last 15 minutes of the interview, you would give the participant the proto-message, and probe as you have during the rest of the interview. We would want to be especially attentive for any new ideas or associations that didn't arise in the interview of native grass.
A sample statement:
"At [Company], we want to redefine what a "healthy lawn" looks like. We believe that a healthy lawn starts with the seed, one which is adapted and native to your location. After that, nature does the rest. Not "nature and a few hundred pounds of chemicals every year," not "nature and tens of thousands of gallons of irrigation." Just nature.
Lawns are a sanctuary for families, and a place for us to be close to nature... not a science lab. Artificially dark, chemically augmented, nitrogen addicted lawns that require constant attention and bleed chemical runoff into our streams aren't the answer, but neither are lawns that look messy and unkempt.
Our goal is to redefine the American lawn by using American grasses to create healthy, beautiful, natural, and low maintenance lawns. We hope you'll join us."
We would also want to understand the participant's thoughts about anything in the proto-message that didn't "fit" their experiences, or things they didn't understand. Again, we want to use it as a stimulus and understand how they interpret it.
While messaging can meaningfully influence the user experience and their interpretations, there is some evidence that customers will also layer in their interpretations of the company's motivation for making a sustainable or socially beneficial product.
In the following research, we see that customers will make evaluations about whether sustainability benefits were intended or unintended by the company, as well as how the performance of the product may be affected by attempts for a company to make a more "green" product. It offers four well-structured experiments, with findings that can very directly affect our work in creating sustainability-driven offerings.
Here is the Abstract, to give you a preview of the paper itself [emphasis is mine]:
Many companies offer products with social benefits that are orthogonal to performance (e.g., green products). The present studies demonstrate that information about a company’s intentions in designing the product plays an import role in consumers’ evaluations. In particular, consumers are less likely to purchase a green product when they perceive that the company intentionally made the product better for the environment compared to when the same environmental benefit occurred as an unintended side effect. This result is explained by consumers’ lay theories about resource allocation: intended (vs. unintended) green enhancements lead consumers to assume that the company diverted resources away from product quality, which in turn drives a reduction in purchase interest. The present studies also identify an important boundary condition based on the type of enhancement and show that the basic intended (vs. unintended) effect generalizes to other types of perceived tradeoffs, such as healthfulness and taste.
Please read "When Going Green Backfires: How Firm Intentions Shape the Evaluation of Socially Beneficial Product Enhancements [21]" by George E. Newman, Margarita Gorlin, and Ravi Dhar.
Credit: 1889 van Gogh painting Wheatfield with cypresses anagoria
"My view on this is as follows: the result must be an action, not an abstract idea. I think principles are good and worth the effort only when they develop into deeds, and I think it's good to reflect and to try to be conscientious, because that makes a person's will to work more resolute and turns the various actions into a whole. I think that people such as you describe would get more steadiness if they went about what they do more rationally, but otherwise I much prefer them to people who make a great show of their principles without making the slightest effort to put them into practice or even giving that a thought. For the latter have no use for the finest of principles, and the former are precisely the people who, if they ever get round to living with willpower and reflection, will do something great. For the great doesn't happen through impulse alone, and is a succession of little things that are brought together." - Vincent van Gogh to Theo van Gogh, October 22, 1882 [22].
As Lesson 7 comes to a close, I'd like to take a moment to cover a common thread which infuses our time together, and that is that when it comes to innovation, details make the difference. Whether it is finding an opportunity, understanding the space, or creating Means End Chains and Cognitive Maps, many do not realize the amount of time, thought, and rigor it takes to create a new "innovation." This is just as we see in van Gogh's letters: The famously impulsive and eccentric artist best known for cutting off part of his own ear in a fit of frustration was extremely concerned with detail and process.
But, also, just as in van Gogh's passage, "principles are good and worth the effort only when they develop into deeds." We can have the utmost concern about the steps we take in research and rigor, but we must never forget the end goal: This is all in service of creating the most compelling offering possible. We must never find ourselves lost in the process and doing something without knowing exactly and specifically why. Interviewing users and creating MEC and Maps could certainly fall into this category, as one could simply go out and feel that they need to "talk" to some customers, and therefore don't give the interviews or analysis proper attention. Because the desired result of those interviews is muted or lost, interviewing customers simply becomes a minimally engaged "checkoff" in a process of which we have lost sight.
If there is a single methodology or mechanism which time and time again helps to refine and progress the strategy of any organization, is it a well-executed Cognitive Map. It is a simple, yet powerful tool, and one I promise you will find myriad uses and adaptations for over time. I hope to be able to further your experience with both Means End Chains and Cognitive Maps in this week's exercise.
To refresh ourselves, our goals for this week's Lesson are to:
To these ends, this week's effort will have us creating and working hands-on to create a draft Cognitive Map, and then examining strategically interesting spaces within.
Links
[1] mailto:axj153@psu.edu
[2] http://search.proquest.com.ezaccess.libraries.psu.edu/docview/222882410?pq-origsite=summon&accountid=13158
[3] https://zh.wikipedia.org/zh-tw/File:Maslow's_Hierarchy_of_Needs.svg
[4] https://commons.wikimedia.org/w/index.php?title=User:Factoryjoe&action=edit&redlink=1
[5] https://creativecommons.org/licenses/by-nc-nd/2.0/
[6] http://www.acrwebsite.org/volumes/7561/volumes/v21/NA-21
[7] http://www.amazon.com/Understanding-Consumer-Decision-Making-Advertising/dp/0805817301
[8] http://search.proquest.com.ezaccess.libraries.psu.edu/docview/227715870?pq-origsite=summon&accountid=13158
[9] https://www.library.hbs.edu/working-knowledge/the-subconscious-mind-of-the-consumer-and-how-to-reach-it
[10] https://www.e-education.psu.edu/ba850/node/686
[11] https://is.muni.cz/el/1456/jaro2013/MPH_MVPS/39278324/LadderingTheoy_original.pdf
[12] https://www.flickr.com/photos/24736216@N07/6439693009
[13] https://www.flickr.com/photos/24736216@N07/
[14] https://www.flickr.com/photos/auntie/
[15] https://creativecommons.org/licenses/by-sa/2.0/
[16] https://www.flickr.com/photos/manfredmajer/14349371293/
[17] https://www.flickr.com/photos/manfredmajer/
[18] https://en.wikipedia.org/wiki/Poa_trivialis
[19] https://www.flickr.com/photos/bods/16672138476/in/photolist-rpg4Vo-5XUkqh-9JLmNm-b4eNFv-5XdPX7-7M6ieY-soWdLi-5Xa6SB-6TMR7Q-pKk1wz-8teLtb-6Bs1Uw-e65K2b-qpyeKG-hjLx79-pU5ScB-ed2f3H-owMcZC-agdHkW-4NkAdS-9ciahr-dQ7AsE-8ZGXEh-q54YPA-bZPonf-8rvbW5-7J4PkJ-b4fc3x-9JLpPU-gegK55-edr2jf-akkGbv-gegsac-amYe2Y-8hjHcF-dfwKuv-6TJ87X-bsfWGj-jMBvYX-d2G8Ls-JYcPJ-6xPRYQ-6TN5R1-akoqV5-etJ11X-eMK75f-peLNCL-hjNkVC-dnhntr-7R2E7V
[20] https://www.flickr.com/photos/bods/
[21] http://www.jstor.org.ezaccess.libraries.psu.edu/stable/10.1086/677841?pq-origsite=summon
[22] http://vangoghletters.org/vg/letters/let274/letter.html