sustainability

Article
Marketing Green Fertilizers: Insights into
Consumer Preferences
Johannes Dahlin 1,2, *, Verena Halbherr 1 , Peter Kurz 3 , Michael Nelles 2,4 and Carsten Herbes 1
1 Institute for International Research on Sustainable Management and Renewable Energy (ISR),
Nuertingen-Geislingen University, Neckarsteige 6-10, 72622 Nuertingen, Germany;
[email protected] (V.H.); [email protected] (C.H.)
2 Faculty of Agricultural and Environmental Sciences, University of Rostock, Justus-von-Liebig-Weg 6,
18059 Rostock, Germany; [email protected] or [email protected]
3 Department of Applied Marketing Science, TNS Deutschland GmbH, Landsberger Str. 284, 80687 Munich,
Germany; [email protected]
4 DBFZ Deutsches Biomasseforschungszentrum Gemeinnützige GmbH, Torgauer Str. 116,
04347 Leipzig, Germany
* Correspondence: [email protected]; Tel.: +49-7022-201-277

Academic Editor: Richard Henry Moore

Received: 21 July 2016; Accepted: 9 November 2016; Published: 11 November 2016

Abstract: In an effort to support the long-term viability of the bioenergy industry through an
end market for digestate, we investigated purchasing preferences for fertilizer product features
in the home gardening market. We conducted a discrete choice experiment (DCE), presenting
504 respondents with a total of 6048 product attribute choices in a simulated context that replicated
the tradeoff decisions made in the real marketplace. We analyzed the choice data using a hierarchical
Bayes estimate to generate part-worth utilities for fertilizer product attributes. We then conducted a
latent class analysis to identify market segments that could be expected to respond to differentiated
product design strategies. We were able to quantify both purchasing preferences for fertilizer product
attributes as well as the importance of each attribute to the perceived utility of a product. We were
further able to identify five distinct market segments that make clear the potential for differentiated
strategies in the home gardening market. We found both negative and positive price sensitivities, with
sociodemographically distinct subgroups that favored low-, mid-, and high-priced products. We also
found purchasing preferences for brand status, product labeling and nutrient values. Our results
provide insights that should help product managers in the biogas industry develop marketing
strategies to integrate digestate into a sustainable energy production system.

Keywords: green fertilizer marketing; biogas digestate; private gardeners; discrete choice experiment

1. Introduction
Few would question that progress towards making societies more sustainable depends on
consumer choices. While green energy and transportation choices plainly contribute to a more
resource-efficient society, less obvious may be how choices in another area—the household garden—can
contribute to an ecologically sustainable future. Yet the soil amendment and fertilizer choices made
by private gardeners form key links in a chain of production processes, and these processes carry
significant environmental consequences.
Many gardens are over-fertilized, and this contributes both to greenhouse gas emissions and
phreatic water contamination [1–6]. The most important of these are nitrogen (N), phosphorus (P) and
potassium (K) [7–9], and the production of each consumes energy and contributes to climate change.
Nitrogen production relies on ammonia as a raw material; the production of ammonia generally
proceeds through the Haber-Bosch manufacturing process, an energy-intensive nitrogen-fixation

Sustainability 2016, 8, 1169; doi:10.3390/su8111169 www.mdpi.com/journal/sustainability

Sustainability 2016, 8, 1169 2 of 15

process that accounts for 1.2% of global energy consumption [10,11]. Phosphorus and potassium, on
the other hand, are obtained through mining [12,13]. Phosphorous production relies heavily on limited
and nonrenewable mineral resources, making it the subject of extensive scrutiny by environmental
scientists [8,11,14,15]. The phosphorous production chain is all the more fragile because 77% of
the world’s phosphate reserves are concentrated in just one country: Morocco [9,16,17]. That is
decidedly not the case with potassium fertilizers, which are mined and produced in many parts of the
world. However, though potassium itself may pose limited environmental risks, the mining activities
involved in the extraction of phosphate rock and potash can cause extensive and long-lasting damage
to the environment.
Seeking to minimize the environmental impacts of mineral fertilizers, researchers have proposed
the use of organic alternatives [2]. One of the well-known alternatives is biogas digestate [18–22]—the
term “digestate” refers to the nutrient-rich residue from the anaerobic fermentation process used in
biogas production. Though every biogas plant (currently 17,240 in Europe, with 10,786 of those in
Germany) produces digestate, it often cannot be applied locally due to high regional nutrient surpluses
and legal restrictions.
An obvious approach to maximizing the value of digestate to a cleaner production system
would be to transport the digestate to arable farming regions with pressing nutrient demand.
Numerous researchers have examined means of facilitating this approach [23–25], including upgrading
solid digestate so it can be transported at reasonable cost and utilized in both agricultural and
non-agricultural sectors [26,27]. But implementing cleaner digestate-based fertilizer production
systems faces obstacles, meaning the economic and ecological benefits of using digestate remain
largely unrealized [28]. Obstacles include the legal status of digestate-based products [28,29], as well as
socioeconomic challenges [30]. The private gardening sector, however, shows considerable potential for
integrating biogas digestate-based products in soil amendments [31]. Still, such integration demands
that digestate producers apply innovation and creativity to overcoming current obstacles; it demands
that they create fertilizer products able to compete successfully in the consumer market. To do so,
producers need to understand the purchasing behavior of their target consumers.
To that end, this study examines the fertilizer purchasing preferences of private gardeners. We use
a discrete choice experiment to analyze how different product attributes play into consumer decisions;
we further evaluate the importance of these attributes to different segments of the non-agricultural
consumer market. Through such analyses emerge strategies for integrating digestate-based products
into the already mature market for soil amendments. We expect these strategies to contribute to the
design decisions of all soil amendment marketers, but in particular to those product managers in the
biogas and livestock sector. The more these managers can create a profitable market for digestate-based
products, the better they will be able to recycle their nutrient surpluses and the sounder will be the
footing on which the biogas industry can stand [31–34].

2. Material and Methods

2.1. Consumer Research and Discrete Choice Experiments

We elected to research consumer fertilizer purchasing preferences through a discrete choice
experiment (DCE) administered by an online questionnaire. DCEs have gained widespread acceptance
in recent years as a sophisticated tool for eliciting consumer preferences for products or services.
In such experiments, products are characterized by a set of attributes (e.g., color and cost) and
attribute levels (e.g., green, red, blue and prices); the combination of attributes at specific levels
comprises the perceived utility of a good [35]. In contrast to simpler survey methods, a DCE approach
presents consumers with decisions that more closely reflect those made in realistic tradeoff situations.
That allows researchers to gain insight into real purchasing preferences.
This approach was originally developed by [36,37] for marketing and transportation research,
but its ability to more faithfully capture realistic decision-making scenarios has made it applicable
Sustainability 2016, 8, 1169 3 of 15

throughout social science. Whether used to evaluate customer preferences for electricity or food
products [38–44], or contracts in the agricultural sector [44–47], or even conservation services and
natural resources [48–52], the discrete choice approach has led to valuable insights into how decisions
are made in the real world.
The validity of discrete choice experiments rests on well-tested random utility theory (RUT) [53].
First introduced by [54] and further developed in by [55–57], random utility theory models consumer
preferences in an intuitively elegant manner: it assumes consumers make choices based on the utility
benefits they gain from the various product attributes. These choices emerge under the influence of
two factors: one rationally explicable and a second inexplicable. That is [53,58,59]:

Uin = Vin + εin

where Uin is the unobservable (latent) utility U, that an individual i associates with a given choice n.
This utility consists of an explainable component, Vin , as well as a random and unexplainable
component εin associated with the consumer choice [53].

2.2. Experimental Design

In selecting relevant product attributes and attribute levels for our experimental design, we sought
to replicate the real-world choices consumers face in making purchasing decisions. We started with a
systematic qualitative process, which is crucial for a valid DCE design [60]. We conducted 20 interviews
with private gardeners, recording each for later transcription. Transcripts were analyzed using the
qualitative data analysis software MAXQDA. From this analysis, we identified the product attributes of
relevance to most gardeners. We then created questions for an online survey based on these attributes
and those levels readily available in German garden stores. This led to the configuration options listed
in Table 1.

Table 1. Fertilizer attributes and their respective levels.

Attribute Attribute Levels

• Universal (general-purpose) fertilizer
Fertilizer type
• Flower (specific purpose) fertilizer
• Premium brands (e.g., Compo, Neudorff)
Brand name • Middle class brands (e.g., Gartenkrone, Flora Self)
• Private labels of garden stores (e.g., Obi, Toom)
• Not labeled
Label
• Labeled as “Organic”
• 4-2-3
Nutrient values (NPK level) • 8-6-7
• 10-12-11
• Organic
Resource (raw material)
• Mineral
• €6
Price (2.5 kg package) • €9
• €12

We choose to associate the resource attribute, i.e., the raw material from which the fertilizer is
made, with the values “organic” or “mineral”, refraining from explicit use of the term “digestate”.
As we know of no product in the market explicitly labeled as digestate, this choice mirrors current
offerings. It further allows our results to be transferred to products derived from other organic
materials such as manure or bone meal. In our fertilizer DCE, an orthogonal fractional factorial design
was applied and was part of a larger survey. It further encompassed a DCE on potting soil, aspects of
Sustainability 2016, 8, 1169 4 of 15

the gardeners purchasing behavior, their risk perception as well as their attitude towards renewable
and resources. In this paper we focus on the results of the fertilizer DCE. The Sawtooth Software is
using a complete enumerated algorithm and resulted in 100 different blocks of which 12 were shown
to each respondent in the survey. Each block contains 3 product options as well as an opt-out option to
choose from.

2.3. Data Collection and Analysis

The survey was conducted in collaboration with the market research institute TNS Deutschland
GmbH. The respondents were recruited from an online panel that contains 160,000 members in total.
All participants were rewarded in cash or through vouchers.
Data collection took place from March through April 2016. This period was chosen for two reasons.
First, it represented the start of the gardening season, when suppliers launch their advertising
campaigns for soil amendment products. Second, extending over the Easter holidays, this period
was thought to offer respondents more time to complete the survey, meaning we could expect a
good response rate and a high degree of engagement in the survey questions. Acceptance criteria for
respondents included:

• must live in a house or apartment;

• must be at least 18 years old;
• must have purchased a garden fertilizer within the last three years.

Only gardeners who met all three criteria were allowed to complete the survey.
In total 1456 gardeners were asked to complete the survey ether containing the fertilizer or the
potting soil DCE. 507 gardeners completed potting soil DCE and 445 did not buy any soil amendments
such as fertilizers and potting soil or did not complete the survey.
In total, 504 respondents qualified and completed the fertilizer DCE. Respondents were each
asked to complete 12 conjoint discrete choice questions. Hence, a total of 6048 choices were available
for analysis.
By independently varying the attributes shown to the respondents and collecting their responses
to the different product profiles, we could statistically deduce what product features were most
desired and which attributes made the most impact on choice. This gave us a full set of preference
scores (often called part-worth utilities) for each attribute level included in the study. Our statistical
approach used a two-level hierarchical Bayesian model to estimate part-worth utilities at the individual
level [61,62]. In this model, the upper level operated under the assumption that part-worth utilities
were characterized by a single multivariate normal distribution. At the lower level, the multinomial
logit model governed the probability of choosing one alternative over another.
The individual part-worth utilities so obtained were further used to segment the respondent
sample into groups with similar purchasing preferences. Segmentation proceeded using the latent
class (LC) method. This approach can reveal details about the market’s structure and so improve the
results derived from a discrete choice experiment [61].

2.4. Selection of Sociodemographic Variables

To analyze the characteristics of different consumer groups, we collected data on a number
of sociodemographic variables. We also queried the respondents about their gardening practices.
The parameters of interest are listed in Table 2.
Sustainability 2016, 8, 1169 5 of 15

Table 2. Sociodemographic Variables and Gardening Questions (standard deviations in parentheses).

Mean Minimum Maximum

Sociodemographic Characteristics
• Age in years 52.32 (13.06) 18 81
• Gender (female in %) 44%
• Number of individuals living in the household 2.63 (1.01) 1 5
• Level of income (total monthly net income):
1 = under €1000,
2 = €1000–€1999,
3 = €2000–€2999, 3.88 (1.52) 1 6
4 = €3000–€3999,
5 = €4000–€4999,
6 = €5000 and over
• Level of education:
1 = no formal education
2 = still attending school/vocational training,
3 = finished vocational training 3.88 (1.21) 1 6
4 = master craftsman training/vocational school
5 = Degree in applied sciences
6 = University degree, PhD
Gardening Questions
• Kind of fertilizer for outdoor use (not potted);
from 1 to 5 with
2.13 (0.87) 1 5
1 = exclusively solid
5 = exclusively liquid
• Kind of fertilizer for potted plants; from 1 to
5 with
2.98 (1.03) 1 5
1 = exclusively solid
5 = exclusively liquid
• I am certain the fertilizer I buy fits my plant’s
needs; from 1 to 6 with
4.28 (1.13) 1 6
1 = strongly disagree
6 = strongly agree
• Fertilizer is needed to promote plant growth;
from 1 to 6 with
4.96 (0.88) 1 6
1 = strongly disagree
6 = strongly agree
• Gardening is an important hobby for me; from 1
to 6 with
4.59 (1.29) 1 6
1 = strongly disagree
6 = strongly agree

3. Results
Overall, the majority of our respondents (92.7%) purchase fertilizers from brick-and-mortar
businesses; only 13.9% purchase products online, while 1.6% have products delivered to their homes
(respondents could select more than one answer, hence percentages sum to values greater than 100).
Of the brick-and-mortar purchases, 80.8% were made in do-it-yourself (DIY) stores, 18.5% in plant
nurseries, and 17.3% in supermarkets. Asked how they made their decisions, 71% of the respondents
said independently; 28% were supported by their partner, children, sales consultants or others.

3.1. Hierarchical Bayes Estimation

Table 3 lists the estimated mean utility values for the hierarchical Bayes (HB) model.
Sustainability 2016, 8, 1169 6 of 15

Table 3. Mean utility values of the hierarchical Bayes (HB) model.

Attribute Level HB Est. Results

Universal fertilizer 4.0
Fertilizer type
Flower fertilizer −4.0
Premium brand 20.9 ***
Brand name Middle class brand −14.5 **
Private label brand −6.4
Not labeled −17.9 ***
Label
Organically labeled 17.9 ***
NPK values (4-2-3) −21.7 ***
Nutrient values NPK values (8-6-7) 8.5 **
NPK values (12-10-11) 13.2 ***
Organic (fertilizer) 10.4 *
Resource
Mineral (fertilizer) −10.4 *
€6 60.0 ***
Price €9 39.7 ***
€12 −99.7 ***
McFadden pseudo R-squared (Percent certainty) 0.63
Root Likelihood (RLH) 0.60
Average Variance 3.53
Parameter RMS (root mean square) 2.35
***, **, * indicate significance at the 0.1%, 1%, and 5% levels respectively.

The utility values convey the importance of the particular attribute level to the overall utility of
the product. These values were available for each respondent, but are presented here as mean values
to capture significant trends. Values are expressed as zero-centered part-worth differences, meaning
the utiles for levels within one attribute are rescaled so they sum to zero. Values close to zero denote
uncertainty in respondents’ evaluation of that particular attribute level. Larger values reveal stronger
coherence between perceived utility and the attribute level [63]. Positive values contribute to greater
utility of the product and negative values to lesser utility in the eyes of the consumer. The willingness
to pay (wtp) is often calculated with this regard; however, it creates inflated values and distorts the
results and even new features in DCE have been proposed to improve the wtp estimates [64–69].
As shown in Table 1, the findings for almost all attribute levels are statistically significant at the
p-values indicated. For example, these results demonstrate that labeling a fertilizer as organic adds to
the perceived utility of the product, so contributes to its purchase; the absence of a label detracts from
a product’s perceived utility, making its purchase less attractive. Both statements are significant at
the p = 0.001 level. At p = 0.05, results demonstrate that respondents favored a fertilizer made from
organic resources over a mineral-based fertilizer. At p = 0.01 or better, the higher the NPK levels in
the fertilizer (at least across the three levels studied), the greater the perceived utility of the product.
Statistical significance, however, is not undisputed and should not be confused with the economic
significance [70–72].
Table 4 lists the attribute importance scores for the whole sample.
These attribute values are expressed as percent of the whole and so sum to 100%. They capture
the degree to which each attribute factors into the consumer’s purchasing decision. These can be, and
should be, compared across attributes, so it is fair to say that considering the respondents as a whole,
price plays the most important role in their purchasing decisions. The quality connoted by the brand
name ranks second, while the raw material used to produce the fertilizer plays the least important role.
Sustainability 2016, 8, 1169 7 of 15

Table 4. Attribute importance scores.

Attributes Importance Scores

Fertilizer type 13%
Brand name 19%
Label 9%
Nutrient values 11%
Resource 8%
Price 40%

3.2. Latent Class Analysis

As discussed in Section 2.3, individual part-worth utilities permit segmentation of the respondent
sample into subgroups with similar purchasing preferences. Identifying subgroups means discovering
statistically significant heterogeneity in respondent choice data. Should such exist, segment-level
part-worths can provide product designers with the insight they need to tailor their offerings to the
preferences of specific customer groups.
To determine the extent of heterogeneity in the choice data, we applied latent class analysis [64],
evaluating the following goodness-of-fit indicators:

(1) McFadden pseudo R-squared (percent certainty)

(2) Consistent Akaike information criterion (CAIC)
(3) Chi-square (chi sq)
(4) Relative chi-square (rel chi sq)

A summary of these indicators is displayed in Table 5, for group sizes varying from two to six.

Table 5. Best Replications for latent class analysis.

Groups Percent Certainty CAIC Chi Sq Rel Chi Sq

2 21.66 13,340.24 3632.24 172.96
3 26.26 12,675.01 4404.25 137.63
4 29.14 12,300.00 4886.04 113.63
5 30.82 12,124.63 5168.19 95.71
6 32.32 11,980.13 5627.26 74.04

In general, the higher the values for percent certainty, consistent Akaike information criterion,
and chi-square, the better the model fit. The lower the values of the relative chi-square, the better
the fit. We decided, however, to choose the model with five groups rather than six. We judged six to
represent excessive compartmentalization, inasmuch as five sufficed to capture distinct purchasing
preference profiles. Mean utility values for these five profiles were then generated using the method
described in Section 3.1, producing the values in Table 6.
The results clearly show heterogeneity in the respondents and their differentiated emphases on
attributes and their levels; we have sought to capture this in the naming of each group. The universal
product buyer emerges as the consumer who strongly prefers a general purpose fertilizer over one
designated for a particular plant. The price sensitive and price sensitive green buyers show strong
preference for low priced products, and so favor private label, i.e., inexpensive, brands. The price
sensitive buyers are differentiated by one group’s strong preference for an organic label (hence “green”)
and the other’s near indifference to the labeling.
Sustainability 2016, 8, 1169 8 of 15

Table 6. HB mean utility values for five segments.

Group 1 Group 2 Group 3 Group 4 Group 5

Universal Price Sensitive Price Sensitive Multi Criteria Premium
Product Buyer Green Buyer Buyer Buyer Product Buyer
n = 53 n = 36 n = 96 n = 273 n = 46
Fertilizer Type
Universal fertilizer 101.9 2.0 −6.2 −37.1 10.0
Flower fertilizer −101.9 −2.0 6.2 37.1 −10.0
Brand Name
Premium brand 20.1 −15.4 −21.5 22.6 176.9
Middle class brand −20.9 −4.2 5.0 −28.9 −76.4
Private label brand 0.8 19.6 16.4 6.2 −100.6
Label
Not labeled −35.0 −57.3 2.8 −54.0 38.5
Organically labeled 35.0 57.3 −2.8 54.0 −38.5
Nutrient Values
NPK values (4-2-3) −33.5 −19.5 −14.9 −45.9 −6.9
NPK values (8-6-7) 16.1 15.2 28.1 0.2 18.1
NPK values (12-10-11) 17.4 4.3 −13.2 45.7 −11.2
Resource
Organic (fertilizer) −10.9 −3.6 11.2 19.4 41.2
Mineral (fertilizer) 10.9 3.6 −11.2 −19.4 −41.2
Price
€6 98.7 196.1 230.6 14.8 −69.5
€9 15.1 12.5 17.7 110.6 25.2
€12 −113.8 −208.6 −248.3 −125.3 44.3

The multi-criteria buyer makes up the largest group and represents the majority of the sample.
This purchasing profile shows preferences in all the attributes tested. Mid-priced but carrying a
premium brand, special-purpose, labeled organic and from organic sources (and of special note,
offering higher nutrient (NPK) values): that is the product profile preferred by the multi-criteria buyer.
Premium product buyers show their strongest preference for the premium brand and
correspondingly for the higher-priced product. Of all the groups, they show the strongest preference for
organic-based fertilizers, but curiously respond negatively to organic labeling. Their NPK preference
is for a medium level nutrient content. These buyers’ strong preference for organic-based fertilizers
and willingness to pay a premium price bodes well for digestate-based and other renewable fertilizer
producers, but their curious response to organic labeling suggests they may be skeptical towards some
certification schemes [73,74]. Hence, validating and overcoming such skepticism may represent a
distinct opportunity for marketing to the premium buyer.
We also computed the importance of each attribute for the purchasing decisions made by buyers
in the five groups; these are shown in Table 7.
Please note that these importance scores only show the weighting given to each attribute in a
purchasing decision; they do not convey information about which level is preferred for a given attribute.
So, for example, although Group 1 and Group 4 both show sensitivity to fertilizer type, Group 1’s
preference is for a general-purpose fertilizer, while Group 4’s is for a special purpose product.
This table illustrates that the premium product buyer—making up almost 10% of the respondents
(46 out of 504)—weighs brand name heavily in making a purchasing decision, while putting roughly
the same utility value on price as on the resource used to make the product. All the other groups rank
price as the most important attribute in making a purchasing decision. For Group 1, fertilizer type
factors into the decision almost as much as price.
Sustainability 2016, 8, 1169 9 of 15

Table 7. Attribute importance scores for five segments.

Group 1 Group 2 Group 3 Group 4 Group 5

Universal Price Sensitive Price Sensitive Multi Criteria Premium
Product Buyer Green Buyer Buyer Buyer Product Buyer
n = 53 n = 36 n = 96 n = 273 n = 46
Fertilizer type 34% 1% 2% 12% 3%
Brand name 7% 6% 6% 9% 46%
Label 12% 19% 1% 18% 13%
Nutrient values 8% 6% 7% 15% 5%
Resource 4% 1% 4% 6% 14%
Price 35% 67% 80% 39% 19%

3.3. Sociodemographic Characteristics of Groups

Our final set of results focuses on the sociodemographic characteristics and gardening practices
of the subgroups identified in Section 3.2. Table 8 presents the mean values and standard deviations
of these descriptive statistics. Please remember that this data represents only gardeners who have
previously purchased soil amendments.

Table 8. Sociodemographic characteristics of the sample (standard deviations in parentheses).

Group 1 Group 2 Group 3 Group 4 Group 5

Universal Price Sensitive Price Sensitive Multi Criteria Premium
Product Buyer Green Buyer Buyer Buyer Product Buyer
n = 53 n = 36 n = 96 n = 273 n = 46
Sociodemographic Characteristics
Average Age of respondents (years) 52.06 (12.42) 57.25 (10.34) 53.14 (13.08) 51.47 (13.78) 52.11 (10.59)
Gender (female in %) 49% 36% 44% 44% 48%
Household size 2.77 (1.15) 2.42 (0.91) 2.51 (1.14) 2.68 (1.07) 2.57 (1.09)
Level of income 4.02 (1.50) 3.78 (1.55) 3.67 (1.61) 3.92 (1.53) 4.02 (1.26)
Level of education 3.98 (1.34) 3.64 (1.15) 3.89 (1.35) 3.86 (1.15) 4.09 (1.15)
Gardening Questions
Kind of fertilizer for outdoor use (not potted) 1.94 (0.79) 2.08 (0.97) 1.99 (0.96) 2.21 (0.85) 2.17 (0.74)
Kind of fertilizer for potted plants 2.68 (1.06) 2.94 (1.16) 3.11 (1.17) 3.00 (0.97) 2.98 (0.95)
Certainty about product fitting plant needs 3.79 (1.23) 4.03 (1.18) 3.89 ( 1.08) 4.43 (1.09) 4.96 (1.01)
Fertilizer is needed to promote plant growth 4.94 (0.85) 4.68 (0.88) 4.79 (0.94) 5.02 (0.82) 5.37 (0.68)
Gardening is an important hobby for me 4.19 (1.34) 4.43 (1.44) 4.25 (1.29) 4.72 (1.22) 5.11 (1.22)

Heterogeneity is again evident in this data, and a number of differences emerge that may help
explain observed preference profiles. For example, premium product buyers have both the highest
level of income and the highest level of education. Being willing to pay the highest price on average,
these buyers also show the greatest certainty in having bought the product that best suits their needs.
They also score highest in ranking gardening as an important leisure activity. This may explain why
they, more than other buyers, perceive the need for fertilizers to promote plant growth.
On the other hand, buyers in the two price sensitive groups have on average the lowest level
of income, so their preference for inexpensive products reflects their income status. Purchasing less
expensive products, these buyers also face greater uncertainty about having purchased the right
product to meet their garden’s needs.

4. Discussion
We have sought to provide insights that will help product managers in the biogas industry create
digestate-based fertilizer products for the home gardener market, and in so doing bolster both the
ecological and economic viability of this promising industry. We designed a discrete choice experiment
that presented respondents with varying arrays of fertilizer product attributes and attribute levels,
then asked respondents to make purchasing choices where they had to evaluate tradeoffs like those
found in the real marketplace. We collected choice data from 504 respondents who made a total of
Sustainability 2016, 8, 1169 10 of 15

6048 purchasing decisions. We subjected the choice data to conjoint analysis, and this allowed us to
quantify both purchasing preferences for fertilizer product attributes as well as the importance of each
attribute to the overall utility of a product.
How best to apply our results to the design of strategies for soil amendment markets depends
on the degree of integration, both vertical and horizontal, biogas facility operators and the biogas
industry achieves. Absent any integration, a single biogas operator can individually create and market
a garden fertilizer. This involves not only production of the digestate, which the operator currently
does but typically only to a raw level, but also further digestate processing as well as developing the
products, the product packaging, the communication strategies and the distribution channels needed
to successfully market the product. The operator would not likely enjoy economies of scale, since
the facility would be marketing a relatively small amount of organic fertilizer. However, an operator
following this course would have the advantage of being able to build a loyal, perhaps predominantly
local, customer base and so obtain relatively high prices.
Taking steps toward horizontal integration in the industry, digestate producers could jointly
create a brand and establish distribution channels for their products. This would enable economies
of scale and facilitate the marketing of larger amounts of organic fertilizer and thus allow for greater
product differentiation. This option could also support distribution into DIY stores that demand
large quantities and a range of products. This option would give biogas operators greater marketing
power while allowing them to remain independent from intermediaries. Clearly, however, plant
operators with individual interests would have to work cooperatively to develop a market for their
digestate products.
A third option entails vertical integration into the existing production system operated by larger
fertilizer manufacturers and distributors. This would allow for the classic division of labor whereby the
biogas plant operator focuses on producing high quality digestate for an integrated production system
that distributes processing and marketing across multiple levels of a supply chain. This approach
has the advantage that all participants can focus on their core competencies and already established
fertilizer marketers can integrate digestate into their existing product offerings. The drawback of this
option is that plant operators and their products would be replaceable. A single biogas plant would be
merely one of many digestate suppliers, so the bulk buyer would have great leverage on the setting
of price. This option, however, does allow the plant operator to manage and market the occurring
digestate in a lean and resource efficient way [35].
Each level of integration offers a different opportunity for leveraging the purchasing preference
results found in this study. Single biogas operators with limited resources and smaller product
quantities are unlikely to have the capacity to satisfy the entire spectrum of customer preferences found
in our latent class analysis. Assuming they could design and market only one product, they could target
either the most common consumer group—the multi-criteria buyer—or one of the more specialized
groups. The mean utility values for the hierarchical Bayes model could then guide their design
decisions. Aiming for the multi-criteria buyer, a producer would create a mid-priced, special-purpose
product, labeled organic and from organic sources and offering higher NPK values. However, the
operator would also need to establish premium status for the facility’s brand to fully accommodate this
buyer. Should the operator instead focus on a smaller, more specialized customer segment, targeting
the premium product buyer would seem a sensible decision, as they exercise the greatest purchasing
power and willingness to pay. However, to reach these buyers, the operator would have to work much
harder to establish premium status for the facility’s brand, since this is the predominant attribute that
attracts the affluent buyers for whom gardening is an essential leisure activity. Attaining such status
on a national scale would likely exceed the means of a single biogas plant; however, the strategy could
also work were the brand perceived as premium only in the mindsets of local buyers. A savvy operator
could exploit local sensibilities to appeal to a localized concept of “premium”.
Of course, these, considerations can just as easily motivate a group of integrated operators.
But with larger digestate quantities and greater marketing resources, managers could design products
Sustainability 2016, 8, 1169 11 of 15

to focus on all five preference profiles found in this study. The greater the degree of integration, the
greater the opportunity to fine-tune offerings to address specific market opportunities.
For all but perhaps the single operator, access to the distribution channels provided by DIY
stores is key to mass market success. This does pose a challenge, however, since the large companies
operating DIY stores have requirements that smaller operators may find difficult to meet. Operators
not eager to do so could develop alternative channels through plant nurseries or the like. This might
be a particularly promising approach to targeting the premium buyer, who would likely respond
favorably to a more refined sales experience. Marketing to the premium buyer as though to a gardening
connoisseur entails greater ingenuity and effort, but the rewards could be substantial.
From an ecological point of view, the fact that the majority of respondents display a
“more-is-better” preference for high NPK values represents an opportunity for improved educational
initiatives. Most home gardeners over-fertilize and more potted plants drown from excessive watering
than die from too little. Over-fertilization leads to eutrophication of water bodies [75,76], and no home
gardener wants to see his or her plants die. An educational concept like “What You Can Learn from a
Green Thumb” could help home gardeners develop practices that were both ecologically more sound
and personally more satisfying. Positioning digestate-based fertilizers as “green thumb friendly” fits
this concept—because, with digestate as the sole raw material, mid-level NPK values are what can be
expected. Otherwise, to achieve the double digit nutrient values that the multi-criteria buyer currently
prefers, supplements would have to be added, which would further burden the production system.

5. Limitations and Further Research

We focused on the purchasing behavior of German gardeners towards fertilizer attributes.
Other country studies might reveal different national preferences, just as studies have demonstrated
nationally differentiated preferences for wood and food products [77,78]. This study also selected those
product attributes common in the German market. Since many market players exist offering a wide
range of products, only a limited range of these products and product attributes could be captured.
Gardens vary extensively and their owners may focus on particular plants such as roses, rhododendron
or box trees. Given the extremely wide variety of potential garden types, a differentiation of all potential
alternatives would have been too complex for this experiment.
Further, although the discrete choice experiment (DCE) represents the most refined way of eliciting
customer preferences, gaps concerning endogeneity have been outlined by [79,80]. Moreover, choices
in a DCE depend heavily on the way a product is presented [81–83]. Product images of fertilizers
should ideally be redesigned to better reflect the attribute options, but this would have entailed a much
greater effort than we were able to expend. It can be expected that nicely designed product packaging
(mostly premium brands) would have achieved higher scores.
Further research would do well to focus on the labeling and certification of soil amendments, for
example on different quality labels. There are several labels in use; however, what they communicate
is not as widely understood as for, say, food labels. A further research area to consider might be
packaging design. Different sizes, materials and forms are available. Since sustainability continues
to gain importance in many sectors, recyclable versus non-recyclable materials could be significant.
A third target for future research would be to look closely at the final product itself. Although it often
remains concealed until its application, a fertilizer can be marketed in different product forms (e.g.,
pellets, powder or granules), structures (liquid versus solid) and colors (e.g., blue, brown, grey, etc.),
all of which are likely to elicit different preference responses. Other avenues of research remain open
as well; the better consumer choices can be aligned with ecologically sound practices, the better our
prospects for an environmentally cleaner future.

Acknowledgments: This research was conducted within the framework of the research project GAERWERT
(No. 22402312). The project is supported by Fachagentur Nachwachsende Rohstoffe e.V. (FNR) on behalf of
the German Federal Ministry of Food and Agriculture. We would also like to thank Charles Duquette for the
English revisions.
Sustainability 2016, 8, 1169 12 of 15

Author Contributions: Johannes Dahlin conceived the experiment, elaborated the survey, analyzed the DCE
and LC data and wrote the article; Verena Halbherr elaborated Table 8 and assisted in methodological question;
Peter Kurz fielded the survey, collected the data and assisted in the methodological setup; Michael Nelles and
Carsten Herbes have both been supportive throughout the process by providing general supervision and guidance.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Livesley, S.J.; Dougherty, B.J.; Smith, A.J.; Navaud, D.; Wylie, L.J.; Arndt, S.K. Soil-atmosphere exchange
of carbon dioxide, methane and nitrous oxide in urban garden systems: Impact of irrigation, fertiliser and
mulch. Urban Ecosyst. 2010, 13, 273–293. [CrossRef]
2. Cameira, M.R.; Tedesco, S.; Leitão, T.E. Water and nitrogen budgets under different production systems in
Lisbon urban farming. Biosyst. Eng. 2014, 125, 65–79. [CrossRef]
3. Taylor, J.R.; Lovell, S.T. Urban home gardens in the Global North: A mixed methods study of ethnic and
migrant home gardens in Chicago, IL. Renew. Agric. Food Syst. 2015, 30, 22–32. [CrossRef]
4. Carey, R.O.; Hochmuth, G.J.; Martinez, C.J.; Boyer, T.H.; Nair, V.D.; Dukes, M.D.; Toor, G.S.; Shober, A.L.;
Cisar, J.L.; Trenholm, L.E.; et al. A Review of Turfgrass Fertilizer Management: Practices: Implications for
Urban Water Quality. HortTechnol. 2012, 22, 280–291.
5. Carrico, A.R.; Fraser, J.; Bazuin, J.T. Green with Envy: Psychological and Social Predictors of Lawn Fertilizer
Application. Environ. Behav. 2013, 45, 427–454. [CrossRef]
6. Dewaelheyns, V.; Elsen, A.; Vandendriessche, H.; Gulinck, H. Garden management and soil fertility in
Flemish domestic gardens. Landsc. Urban Plan. 2013, 116, 25–35. [CrossRef]
7. Ciceri, D.; Manning, D.A.; Allanore, A. Historical and technical developments of potassium resources.
Sci. Total Environ. 2015, 502, 590–601. [CrossRef] [PubMed]
8. Dawson, C.J.; Hilton, J. Fertiliser availability in a resource-limited world: Production and recycling of
nitrogen and phosphorus. Food Policy 2011, 36, S14–S22. [CrossRef]
9. Wellmer, F.-W.; Scholz, R.W. The Right to Know the Geopotential of Minerals for Ensuring Food Supply
Security: The Case of Phosphorus. J. Ind. Ecol. 2015, 19, 3–6. [CrossRef]
10. International Fertilizer Industry Association (IFA). Fertilizer Facts: Ammonia Production: Moving towards
Maximum Efficiency and Lower GHG Emissions; International Fertilizer Organization: Paris, France, 2014.
11. Hasler, K.; Bröring, S.; Omta, S.; Olfs, H.-W. Life cycle assessment (LCA) of different fertilizer product types.
Eur. J. Agron. 2015, 69, 41–51. [CrossRef]
12. Manning, D.A. Mineral sources of potassium for plant nutrition: A review. Agron. Sustain. Dev. 2010, 30,
281–294. [CrossRef]
13. Geissler, B.; Mew, M.C.; Weber, O.; Steiner, G. Efficiency performance of the world’s leading corporations in
phosphate rock mining. Resour. Conserv. Recycl. 2015, 105, 246–258. [CrossRef]
14. Skowrońska, M.; Filipek, T. Life cycle assessment of fertilizers: A review. Int. Agrophys. 2014, 28, 101–110.
[CrossRef]
15. Mew, M.C. Phosphate rock costs, prices and resources interaction. Sci. Total Environ. 2016, 542, 1008–1012.
[CrossRef] [PubMed]
16. Cooper, J.; Lombardi, R.; Boardman, D.; Carliell-Marquet, C. The future distribution and production of
global phosphate rock reserves. Resour. Conserv. Recycl. 2011, 57, 78–86. [CrossRef]
17. Walan, P.; Davidsson, S.; Johansson, S.; Höök, M. Phosphate rock production and depletion: Regional
disaggregated modeling and global implications. Resour. Conserv. Recycl. 2014, 93, 178–187. [CrossRef]
18. Huttunen, S.; Manninen, K.; Leskinen, P. Combining biogas LCA reviews with stakeholder interviews to
analyse life cycle impacts at a practical level. J. Clean. Prod. 2014, 80, 5–16. [CrossRef]
19. Egle, L.; Rechberger, H.; Zessner, M. Overview and description of technologies for recovering phosphorus
from municipal wastewater. Resour. Conserv. Recycl. 2015, 105, 325–346. [CrossRef]
20. Sigurnjak, I.; Michels, E.; Crappé, S.; Buysens, S.; Tack, F.M.; Meers, E. Utilization of derivatives from nutrient
recovery processes as alternatives for fossil-based mineral fertilizers in commercial greenhouse production
of Lactuca sativa L. Sci. Hortic. 2016, 198, 267–276. [CrossRef]
21. Tampio, E.; Marttinen, S.; Rintala, J. Liquid fertilizer products from anaerobic digestion of food waste: Mass,
nutrient and energy balance of four digestate liquid treatment systems. J. Clean. Prod. 2016, 125, 22–32.
[CrossRef]
Sustainability 2016, 8, 1169 13 of 15

22. Möller, K. Effects of anaerobic digestion on soil carbon and nitrogen turnover, N emissions, and soil biological
activity. A review. Agron. Sustain. Dev. 2015, 35, 1021–1041. [CrossRef]
23. Vázquez-Rowe, I.; Golkowska, K.; Lebuf, V.; Vaneeckhaute, C.; Michels, E.; Meers, E.; Benetto, E.; Koster, D.
Environmental assessment of digestate treatment technologies using LCA methodology. Waste Manag. 2015,
43, 442–459. [CrossRef] [PubMed]
24. Golkowska, K.; Vázquez-Rowe, I.; Lebuf, V.; Accoe, F.; Koster, D. Assessing the treatment costs and the
fertilizing value of the output products in digestate treatment systems. Water Sci. Technol. 2014, 69, 656–662.
[CrossRef] [PubMed]
25. Fuchs, W.; Drosg, B. Assessment of the state of the art of technologies for the processing of digestate residue
from anaerobic digesters. Water Sci. Technol. 2013, 67, 1984–1993. [CrossRef] [PubMed]
26. Wang, S.; Fan, J.; Zhao, D.; Yang, S.; Fu, Y. Predicting consumers’ intention to adopt hybrid electric vehicles:
Using an extended version of the theory of planned behavior model. Transportation 2016, 43, 123–143.
[CrossRef]
27. Sheets, J.P.; Yang, L.; Ge, X.; Wang, Z.; Li, Y. Beyond land application: Emerging technologies for the
treatment and reuse of anaerobically digested agricultural and food waste. Waste Manag. 2015, 44, 94–115.
[CrossRef] [PubMed]
28. Saveyn, H.; Eder, P. (Eds.) End-of-Waste Criteria for Biodegradable Waste Subjected to Biological Treatment (Compost
& Digestate): Technical Proposals; European Union: Sevilla, Spain, 2014.
29. Cesaro, A.; Belgiorno, V.; Guida, M. Compost from organic solid waste: Quality assessment and European
regulations for its sustainable use. Resour. Conserv. Recycl. 2015, 94, 72–79. [CrossRef]
30. Riding, M.J.; Herbert, B.M.J.; Ricketts, L.; Dodd, I.; Ostle, N.; Semple, K.T. Harmonising conflicts between
science, regulation, perception and environmental impact: The case of soil conditioners from bioenergy.
Environ. Int. 2015, 75C, 52–67. [CrossRef] [PubMed]
31. Dahlin, J.; Herbes, C.; Nelles, M. Biogas digestate marketing: Qualitative insights into the supply side.
Resour. Conserv. Recycl. 2015, 104, 152–161. [CrossRef]
32. Guenther-Lübbers, W.; Bergmann, H.; Theuvsen, L. Potential analysis of the biogas production—As
measured by effects of added value and employment. J. Clean. Prod. 2016, 129, 556–564. [CrossRef]
33. Kröger, R.; Reckermann, M.; Schaper, C.; Theuvesen, L. Gärreste als Gartendünger vermarkten?
Ber. Landwirtsch. 2016. [CrossRef]
34. Dahlin, J.; Nelles, M.; Herbes, C. Vermarktung von Gärprodukten an Privathaushalte. In Biogas in
der Landwirtschaft—Stand und Perspektiven; KTBL: Darmstadt, Germany, 2015; pp. 307–315. ISBN:
978-3-945088-07-4.
35. Lancaster, K.J. A New Approach to Consumer Theory. J. Political Econ. 1966, 74, 132–157. [CrossRef]
36. Louviere, J.J.; Hensher, D.A. On the design and analysis of simulated or allocation experiments in travel
choice modelling. Transp. Res. Rec. 1982, 890, 11–17.
37. Louviere, J.J.; Woodworth, G. Design and Analysis of Simulated Consumer Choice or Allocation Experiments:
An Approach Based on Aggregate Data. J. Mark. Res. 1983, 350–367. [CrossRef]
38. Thøgersen, J.; Nielsen, K.S. A better carbon footprint label. J. Clean. Prod. 2016, 125, 86–94. [CrossRef]
39. Tabi, A.; Hille, S.L.; Wüstenhagen, R. What makes people seal the green power deal?—Customer
segmentation based on choice experiment in Germany. Ecol. Econ. 2014, 107, 206–215. [CrossRef]
40. Rommel, J.; Sagebiel, J.; Müller, J.R. Quality uncertainty and the market for renewable energy: Evidence from
German consumers. Renew. Energy 2016, 94, 106–113. [CrossRef]
41. Vecchiato, D.; Tempesta, T. Public preferences for electricity contracts including renewable energy:
A marketing analysis with choice experiments. Energy 2015, 88, 168–179. [CrossRef]
42. Meyerding, S.G. Consumer preferences for food labels on tomatoes in Germany—A comparison of a
quasi-experiment and two stated preference approaches. Appetite 2016, 103, 105–112. [CrossRef] [PubMed]
43. Ma, C.; Burton, M. Warm glow from green power: Evidence from Australian electricity consumers. J. Environ.
Econ. Manag. 2016, 78, 106–120. [CrossRef]
44. Lombardi, G.V.; Berni, R.; Rocchi, B. Environmental friendly food. Choice experiment to assess consumer’s
attitude toward “climate neutral” milk: The role of communication. J. Clean. Prod. 2016. [CrossRef]
45. Sauthoff, S.; Musshoff, O.; Danne, M.; Anastassiadis, F. Sugar beet as a biogas substrate? A discrete choice
experiment for the design of substrate supply contracts for German farmers. Biomass Bioenergy 2016, 90,
163–172. [CrossRef]
Sustainability 2016, 8, 1169 14 of 15

46. Kragt, M.E.; Gibson, F.L.; Maseyk, F.; Wilson, K.A. Public willingness to pay for carbon farming and its
co-benefits. Ecol. Econ. 2016, 126, 125–131. [CrossRef]
47. Lizin, S.; van Passel, S.; Schreurs, E. Farmers’ perceived cost of land use restrictions: A simulated purchasing
decision using discrete choice experiments. Land Use Policy 2015, 46, 115–124. [CrossRef]
48. Hoyos, D.; Mariel, P.; Hess, S. Incorporating environmental attitudes in discrete choice models:
An exploration of the utility of the awareness of consequences scale. Sci. Total Environ. 2015, 505, 1100–1111.
[CrossRef] [PubMed]
49. Narjes, M.E.; Lippert, C. Longan fruit farmers’ demand for policies aimed at conserving native pollinating
bees in Northern Thailand. Ecosyst. Serv. 2016, 18, 58–67. [CrossRef]
50. Hoyos, D. The state of the art of environmental valuation with discrete choice experiments. Ecol. Econ. 2010,
69, 1595–1603. [CrossRef]
51. Justes, A.; Barberán, R.; Farizo, B.A. Economic valuation of domestic water uses. Sci. Total Environ. 2014, 472,
712–718. [CrossRef] [PubMed]
52. Pan, D.; Zhou, G.; Zhang, N.; Zhang, L. Farmers’ preferences for livestock pollution control policy in China:
A choice experiment method. J. Clean. Prod. 2016, 131, 572–582. [CrossRef]
53. Louviere, J.J.; Flynn, T.N.; Carson, R.T. Discrete Choice Experiments Are Not Conjoint Analysis.
J. Choice Model. 2010, 3, 57–72. [CrossRef]
54. Thurstone, L.L. A law of comparative judgment. Psychol. Rev. 1927, 34, 273–286. [CrossRef]
55. McFadden, D. The Choice Theory Approach to Market Research. Mark. Sci. 1986, 5, 275–297. [CrossRef]
56. McFadden, D.; Train, K. Mixed MNL Models for Discrete Response. J. Appl. Econom. 2000, 15, 447–470.
[CrossRef]
57. McFadden, D. Conditional Logit Analysis of Qualitative Choice Behavior; Academic Press: New York, NY, USA,
1974; pp. 105–142.
58. Train, K. Discrete Choice Methods with Simulation; Cambridge University Press: Cambridge, UK, 2009.
59. Rao, V.R. Applied Conjoint Analysis; Springer: Berlin/Heidelberg, Germany, 2014.
60. Kløjgaard, M.E.; Bech, M.; Søgaard, R. Designing a Stated Choice Experiment: The Value of a Qualitative
Process. J. Choice Model. 2012, 5, 1–18. [CrossRef]
61. Orme, B. CBC/HB v5: Software for Hierarchical Bayes Estimation for CBC Data; Sawtooth Software, Inc.: Swquim,
UT, USA, 2009.
62. Rossi, P.E.; Allenby, G.M. Bayesian Statistics and Marketing. Mark. Sci. 2003, 22, 304–328. [CrossRef]
63. Orme, B. Latent Class v4.5: Software for Latent Class Estimation for CBC Data; Sawtooth Software, Inc.: Orem,
UT, USA, 2012.
64. Wertenbroch, K.; Skiera, B. Measuring Consumers’ Willingness to Pay at the Point of Purchase. J. Mark. Res.
2002, 39, 228–241. [CrossRef]
65. Herbes, C.; Friege, C.; Baldo, D.; Mueller, K.-M. Willingness to pay lip service? Applying a
neuroscience-based method to WTP for green electricity. Energy Policy 2015, 87, 562–572. [CrossRef]
66. Matthes, J.; Wonneberger, A.; Schmuck, D. Consumers’ green involvement and the persuasive effects of
emotional versus functional ads. J. Bus. Res. 2014, 67, 1885–1893. [CrossRef]
67. Allenby, G.M.; Brazell, J.P.; Howell, J.R.; Rossi, P.E. Economic Valuation of Product Features. Quant. Mark Econ.
2014, 12, 421–456. [CrossRef]
68. Orme, B. Assessing the Monetary Value of Attribute Levels with Conjoint Analysis: Warnings and Suggestions;
Sawtooth Software, Inc.: Orem, UT, USA, 2001; pp. 1–6.
69. Schlereth, C.; Skiera, B. Two New Features in Discrete Choice Experiments to Improve Willingness-to-Pay
Estimation That Result in SDR and SADR: Separated (Adaptive) Dual Response. Manag. Sci. 2016. [CrossRef]
70. Ziliak, S.T.; McCloskey, D.N. Size matters: The standard error of regressions in the American Economic
Review. J. Socio-Econ. 2004, 33, 527–546. [CrossRef]
71. McCloskey, D.N.; Ziliak, S.T. The Standard Error of Regressions. J. Econ. Lit. 1996, 34, 97–114.
72. Ziliak, S.T.; McCloskey, D.N. The Cult of Statistical Significance: How the Standard Error Costs Us Jobs, Justice,
and Lives; University of Michigan Press: Ann Arbor, MI, USA, 2008.
73. Olson, E.L. The rationalization and persistence of organic food beliefs in the face of contrary evidence.
J. Clean. Prod. 2016. [CrossRef]
74. Janssen, M.; Hamm, U. Governmental and private certification labels for organic food: Consumer attitudes
and preferences in Germany. Food Policy 2014, 49, 437–448. [CrossRef]
Sustainability 2016, 8, 1169 15 of 15

75. Lundberg, C. Eutrophication, risk management and sustainability. The perceptions of different stakeholders
in the northern Baltic Sea. Mar. Pollut. Bull. 2013, 66, 143–150. [CrossRef] [PubMed]
76. Ulrich, A.E.; Malley, D.F.; Watts, P.D. Lake Winnipeg Basin: Advocacy, challenges and progress for sustainable
phosphorus and eutrophication control. Sci. Total Environ. 2016, 542, 1030–1039. [CrossRef] [PubMed]
77. Cai, Z.; Aguilar, F.X. Consumer stated purchasing preferences and corporate social responsibility in the
wood products industry: A conjoint analysis in the U.S. and China. Ecol. Econ. 2013, 95, 118–127. [CrossRef]
78. Tait, P.; Saunders, C.; Guenther, M.; Rutherford, P. Emerging versus developed economy consumer
willingness to pay for environmentally sustainable food production: A choice experiment approach
comparing Indian, Chinese and United Kingdom lamb consumers. J. Clean. Prod. 2016, 124, 65–72. [CrossRef]
79. Guevara, C.A. Critical assessment of five methods to correct for endogeneity in discrete-choice models.
Transp. Res. A Policy Pract. 2015, 82, 240–254. [CrossRef]
80. Vij, A.; Walker, J.L. Preference endogeneity in discrete choice models. Transp. Res. B Methodol. 2014, 64,
90–105. [CrossRef]
81. Pichert, D.; Katsikopoulos, K.V. Green defaults: Information presentation and pro-environmental behaviour.
J. Environ. Psychol. 2008, 28, 63–73. [CrossRef]
82. Yoo, J.; Kim, M. The effects of online product presentation on consumer responses: A mental imagery
perspective. J. Bus. Res. 2014, 67, 2464–2472. [CrossRef]
83. Jurado, J.V.; Ruiz-Madrid, M.N. A Multimodal Approach to Product Presentations. Procedia Soc. Behav. Sci.
2015, 173, 252–258. [CrossRef]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).