Pesticide risk perceptions and the differences between farmers and extensionists: Towards a knowledge-in-context model

Full text

Pesticide risk perceptions and the differences between farmers
and extensionists: Towards a knowledge-in-context model
$
Adriana Ríos-González
a,b
, Kees Jansen
c,
n
, Héctor Javier Sánchez-Pérez
a,b
a
Society, Culture and Health Academic Area, El Colegio de la Frontera Sur. Carretera Panamericana y Periférico Sur s/n, Barrio de María Auxiliadora cp. 29290
San Cristóbal de Las Casas, Chiapas, Mexico
b
The Africa and Latin America Research Groups Network (GRAAL), Faculty of Medicine, Biostatistics Unit, Barcelona Autonomous University, Edificio M,
08193 Bellaterra, Spain
c
Knowledge, Technology and Innovation Group, Wageningen University, Hollandseweg 1, 6706 KN Wageningen, The Netherlands
article info
Article history:
Received 31 October 2012
Received in revised form
11 February 2013
Accepted 25 March 2013
Available online 21 April 2013
Keywords:
Risk perception
Knowledge-in-context
Blaming
Pesticides
Schooling
Social perspective
abstract
A growing body of literature analyzes farmer perceptions of pesticide risk, but much less attention has
been given to differences in risk perception between farmers and technical experts. Furthermore,
inconsistencies in knowledge have too easily been explained in terms of lack of knowledge rather than
exploring the underlying reasons for particular forms of thinking about pesticide risks. By doing this, the
division between expert and lay knowledge has been deepened rather than transcended.
Objective: This study aims to understand differences and similarities among the perceptions of pesticide
risks of farmers, farm workers, and technical experts such as extensionists, by applying a social science
approach towards knowledge and risk attitudes.
Methods: Semi-structured interviews and field observations were conducted to smallholders, farm workers,
extensionists, health professionals and scientists involved in the use and handling of pesticides. Subse-
quently, a survey was carried out to quantify the farmers and extensionists'acceptance or rejection of typical
assertions expressed previously in the semi-structured interviews.
Results: Smallholders showed to gain knowledge from their own experiences and to adapt pesticides
practices, which is a potential basis for transforming notions of pesticide safety and risk reduction strategies.
Though extensionists have received formal education, they sometimes develop ideas deviating from the
technical perspective. The risk perception of the studied actors appearedto vary according to their role in the
agricultural labor process; they varied much less than expected according to their schooling level.
Conclusions: Commitment to the technical perspective is not dramatically different for extensionists on the
one hand and farmers as well as farm workers on the other hand. Ideas about a supposed lack of knowledge
by farmers and the need of formal training are too much driven by a deficit model of knowledge. Further
research on risk perceptions of pesticides and training of rural people will benefitfromthedevelopmentofa
knowledge-in-context model.
&2013 Elsevier Inc. All rights reserved.
1. Introduction
The unsafe handling of pesticides in Latin American countries has
been widely documented, revealing practices such as users (farmers
and farm-workers) not protecting themselves, use of restricted
pesticides, and children playing in the farm (Barraza et al., 2011;
Polidoro et al., 2008;Blanco-Muñoz and Lacasaña, 2011). The
literature on farmers'knowledge of pesticide risks is steadily growing
(for example, Hashemi et al., 2012;Ibitayo, 2006;Mokhele, 2011;
Palis et al., 2006;Salameh et al., 2004;Sam et al., 2008). Many of
these studies attribute mishandling of pesticides to a supposed lack
of knowledge of farmers, who may be qualified as ‘under trained’
and ‘illiterate’(Sam et al. 2008). Solutions are formulated as the need
for more ‘formal education’or ‘better education’(Ibitayo, 2006;
Salameh et al., 2004;Sam et al., 2008). However, in many of these
studies farmers'perceptions are mainly interpreted from an expert
position (Ibitayo, 2006;Salameh et al., 2004;Sam et al., 2008)which
may downplay the value of the lay people's conceptualization of
risks. Science studies have drawn attention to social processes that
disqualify lay peoples'knowledge (Blok et al., 2008) and reinforce the
dichotomy between lay and expert (Wynne, 1996). The differences
between lay people and expert conceptualization of risk may hinder
the implementation of effective risk communication.
Contents lists available at SciVerse ScienceDirect
journal homepage: www.elsevier.com/locate/envres
Environmental Research
0013-9351/$ - see front matter &2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.envres.2013.03.006
☆
This research was supported by the National Council of Science and Technology
of Mexico (Consejo Nacional de Ciencia y Tecnología CONACyT) and the National
Institute of Public Health of Mexico (Instituto Nacional de Salud Pública INSP).
n
Corresponding author.
E-mail addresses: adrianariosg@hotmail.com (A. Ríos-González),
Kees.Jansen@wur.nl (K. Jansen), hsanchez@ecosur.mx (H. Javier Sánchez-Pérez).
Environmental Research 124 (2013) 43–53

In this study we attempt to examine how farmers and farm
workers (considered as lay people) differ in their pesticide risk
perceptions from technicians who are supposed to be experts in
these matters. The case study was carried out in Chiapas in
Southern Mexico among people involved in pesticide use in
tomato and banana cultivation. Like elsewhere in Mexico, policies
do not promote effective safety measures (Idrovo, 2005). Most of
the production of this region goes to the national market and is
produced under weak regulations for pesticide application, as is
more often the case in developing countries (Galt, 2007;Jansen
and Vellema, 2004).
Specifically, this study addresses the questions of how pesticide
risk perceptions of farmers, farm workers, and extensionists and
other experts are different or similar, how these perceptions are
related to practices in the field, and how different groups think
about how other actors view and handle pesticides. Our research
approach to address these questions has several key elements.
First, it broadens the focus from farmers (the single focus of most
literature) to other actors by comparing farmers'perceptions with
those of extensionists who provide advice about how to handle
pesticides. Second, it develops an approach in which risk percep-
tions are first recorded without using an a priori assessment
whether they are correct or not. This differs from most literature
on farmers'risk perception which often uses one-sided notions to
qualify farmers as ‘aware/unaware’,‘act incorrectly’, and so on. A
sociological perspective is taken that assumes that different forms
of knowledge can bring useful insights to a particular situation;
people's knowledge should be appraised without giving different
hierarchy to one or another type of knowledge (Blok et al., 2008).
This does not mean that comparison with a technical perspective
(to be defined in the next section) is impossible. But such a
comparison should be made explicit and is only carried out in a
second instance. Third, we investigate how different actors frame
pesticide problems (Jansen, 2003,2008) and blame other actors
while talking about pesticide issues. Fourth, we develop a com-
bined qualitative and quantitative approach in order to quantify
how representative the various views are and how they relate to
other variables (Ton, 2012). This may help to bridge the gap
between more interpretative anthropological approaches of local
narratives (e.g. Guivant, 2003;Jansen, 2008) and the many
quantitative studies on risk perceptions.
2. Methodology
The term ‘technical perspective’is defined in this study as a
way of thinking in line with scientific knowledge about how to
deal with pesticide risks. It is for example reflected in the good
practices as defined in International Code of Conduct on the
Distribution and Use of Pesticides managed by the Food and
Agriculture Organization of the United Nations. It concerns a set
of ‘good practices’about how to label the package, store, trade,
mix, use, adjust spraying equipment, and use personal protective
equipment, among other practices. The technical perspective also
includes notions used by health specialists on what to do in case of
poisonings.
3. Local situation
Chiapas is one of the states with the highest production of
fruits and vegetables in Mexico. Soconusco and Frailesca, the
locations of our study, are the most important agricultural regions
of Chiapas. In Soconusco, with its tropical wet climate, coffee and
tropical fruits such as papaya, mango, and banana are grown. In
Frailesca, with its humid subtropical climate, farmers cultivate
vegetables such as tomato (Instituto Nacional de Estadística
Geografía e Informática (INEGI), 2010). In both regions, large-
scale farmers and smallholders compete from an unequal starting
position (Solis, 2005). Large-scale farmers own large plantations,
have been trained in technical innovation, have enough resources
to hire farm-workers to apply pesticides, and are integrated into a
large commercial-network to market their produce. In contrast,
smallholders hardly receive training, have small plots, are less
mechanized, lack commercial networks, and thus face unfavorable
conditions in the market. They mostly use family labor, which
implies that they themselves apply pesticides. Many smallholders
depend on government support in order to be able to stay active in
agriculture. Another relevant group for this analysis is the farm-
workers with no access to land. They work in the plantations in
exchange of low wages. Commonly they have not received any
type of training in agriculture or in any other job, thus agricultural
work is crucial for their survival. The mentioned differences result
in high incomes for large-scale farmers, low incomes for small-
holders, and very low incomes for farm-workers.
Different types of pesticides are used in the study sites: the
highest number of applications is with fungicides in banana and
tomato (tridemorph, mancozeb and bitertanol: between 35 and 52
applications per year). The most used pesticides are the herbicides
glyphosate and paraquat, and the insecticides bifenthrin, carbaryl
and carbofuran. Some of them (e.g. carbofuran, carbaryl, paraquat
and tridemorph) have been banned in European countries due to
their potential environmental hazard (European Union (EU),
2003). All pesticides are sprayed, except for carbaryl, which is
applied by granulates. In many cases, conditions of applications in
Mexico lead to unsafe use. For example, all farmers eat in the field,
up to 47% use the pesticides without any personal protective
equipment (Tinoco-Ojanguren and Halperin, 1998), only 28% take
a shower after having applied pesticides, and 42% of the farmers
store pesticides in their homes (Blanco-Muñoz and Lacasaña,
2011).
4. Methods
Unlike many other studies on pesticide risk perceptions that use either
qualitative (e.g. Guivant, 2003) or quantitative methods (Isin and Yildirim, 2007)
we developed a combination of both methods. First, data on perceptions of
pesticide risks and pesticide practices were collected using semi-structured inter-
views and field observations. Subsequently, we formulated a survey containing a
set of typical assertions based on local expressions related to pesticide risks. In the
second step we quantified peoples'acceptance or rejection of these assertions
through a survey. In this way the survey questions were formulated very much in
the local discourse. This not only increased comprehension by respondents. It also
allowed for asking about assertions that a scientist would normally not include in a
survey (see for example assertion 1–3inTable 2).
The field observations of pesticide use practices were carried out during
November and December of 2009 in four study sites: one large-scale banana farm
and one banana co-operative (collective farm) located in the Soconusco region, and
one large-scale tomato farm and one community with smallholder tomato farmers
in the Frailesca region. We selected sites that would be instructive for observing
different kinds of pesticide use according to researchers knowing the region and
key informants from farmer associations.
The semi-structured interviews (n¼27) with large-scale farmers (n¼8), small-
holders (n¼4) and farm-workers (n¼4), health professionals (physicians, health
promoters and employees of the ministry of health) (n¼4), scientists (researchers
working in agriculture) (n¼3) and extensionists (employees of the ministry of
agriculture and pesticide companies) (n¼4) took place in November 2009 and
January 2010. Within each group, we selected those respondents who were
involved in pesticide use and management; e.g. farm workers who apply pesticides
were selected over those who do other tasks. Interviews with farm-workers were
conducted in their house and not in the plantation in order to avoid that
interviewees would feel intimidated and reluctant to talk about the different actors
around the farm. Except for two tomato smallholders and two farm-workers who
stated feeling ashamed of being recorded, all the other interviewees gave their
consent to be recorded during the whole conversation. Two large scale farmers and
one extensionist asked to stop the recording at one point of the interview. All
interviews included similar questions; the questions were formulated in different
A. Ríos-González et al. / Environmental Research 124 (2013) 43–5344

variations depending on the role of the particular actor interviewed. First, we asked
about their work career, e.g. we asked farmers to describe the history of their farms
and to the rest of the actors how they started their current job. Subsequently we
asked farmers about their methods of pest management, extensionists and
scientists about the problems of pests control and health professionals about the
health problems in the area. Then we inquired about the advantages and
disadvantages of their work in relation to pest management and the use of
pesticides. In this part of the interview we asked to actors who do not manipulate
pesticides; to health professionals about the principal health risks in the area and
to scientists about the main problems in the agricultural production. In many cases
interviewees then themselves started to talk about pesticides (not yet introduced
as topic by the interviewer). Only in the fourth part of the interview, we asked
farmers/farm workers, extensionists, scientists and health professionals about their
views on pesticide risks, for example, if they think that pesticides are harmful or
not for environmental and human health. If the interviewee had not previously
mentioned any disadvantage of pesticides and health risk of these substances, but
now stated that pesticides are harmful, then we asked why they had not mentioned
pesticide risks earlier on. Based on this question it was possible to deepen the
conversation about pesticides risk. Finally, we asked all interviewees to give their
view on what other actors do regarding pesticides in order to study any form of
blaming.
The survey used a Likert scale and was conducted face to face to 99
smallholders and 96 extensionists during February and March 2010. Respondents
specified their level of agreement or disagreement with 30 items according to a
five-point scale; 1¼strongly disagree, 2 ¼disagree. 3 ¼neither agree nor disagree,
4¼agree, 5¼strongly agree. Each item corresponded to an assertion. The assertions
that in the analysis were most remarkable and could be further interpreted by
triangulation with data from the semi-structured interviews and observations are
listed in Table 2. To reduce possible gender influence during interviewing the
female interviewer was mostly accompanied by a male assistant.
5. Type of analysis
5.1. Analysis of the interviews
The semi-structured interviews and behavior were transcribed
and coded (using NVivo 8.0, software for qualitative analysis),
interesting assertions were identified, and differences between the
various actor groups were explored. We did a preliminary analysis
to design the survey (see a longer discussion in the results
section). In this phase we did not use the technical perspective
as the standard from which to formulate the assertions and
instead of this, we took a more neutral position, not giving more
authority to either the farmer or the extensionist'sview(we
hypothesized that the extensions'view would more or less follow
the technical perspective). The technical perspective already prior-
itizes the expert view of what is risky (Wynne, 1996). We consider
that our approach gives more insight into the origins and dynamics
of the different types of knowledge. However, we do not think that
a fully neutral or agnostic position regarding whether a particular
view on risk is more or less correct and can be maintained
throughout the analysis (drinking a pesticide is likely to cause
more harm than keeping it in the bottle). Hence, in the later stage of
the analysis we used the technical perspective as a yardstick when
necessary. For the discussion below, we classified first the respon-
dents'answers in three groups. The first two are expressions
meaning “there is risk”and expressions meaning “there is no risk”
or “there is little risk”. Then we found a category of people who in
one interview expressed “there is risk”in one moment while in
other moments “there is little risk”. Secondly, these expressions
were checked with field observations, in particular the practices of
respondents who had asserted that pesticide use involves risks, for
example a pesticide user who expressed that there is risk but does
not protect himself. In the case of actors who do not handle
pesticides (scientists and health professionals) we observed beha-
vior and interviewed them during field visits about their own
possible exposure and the exposure of others (e.g. regarding any
kind of protection they used when going to farmer fields).
The interviewees who only used expressions of ‘there is risk’in
the interview and showed practices confirming this, were
considered to have a risk aware attitude. Those who only used
expressions downplaying the risk or did not find them important
and showed practices confirming this were considered to have a
risk disregarding attitude. Then we found people who expressed
‘there is risk’but whose practices did not coincide. We also found
people who in a single interview expressed different perceptions:
in one moment as risk aware while in other moments disregarding
risks. These last two categories we call the risk dissonant attitude.
We looked how these three categories of attitudes were
distributed over the different actors, grouped in various ways:
type of actor (large-scale farmer, smallholder, farm worker, exten-
sionists, health-professionals and scientists), exposure to pesti-
cides (exposed¼to have contact with the substance, non-
exposed¼not having or little contact with the substance) and
school level of the actors. The latter ranged from farm-workers
without elementary school and smallholders with elementary
school and in few cases with high school to large-scale farmers,
extensionists with a university degree and scientists with a
postgraduate studies (‘low-schooling’¼high school or less, and
‘high-schooling’¼university degree).
5.2. Analysis of the survey
The level of agreement of the respondents with the various
assertions in the survey was analyzed for different groups using
crosstab and chi-square tests (using SPSS v. 15). Relevant variables
were type of actor (farmer or extensionists), exposure and the
schooling level of farmers. Considering that having completed
elementary school is a minimum requirement for more remuner-
ated jobs (as alternative to agricultural work), the schooling level
of farmers were classified into ‘elementary school not completed’
and ‘elementary school completed’(the latter includes higher
education).
6. Results
This section presents separately the results of the semi-
structured interviews and the observations (qualitative, interpre-
tive study), and of the survey (quantitative study). In the discus-
sion section we will make cross-links. In Each sub-section we will
successively discuss perceptions that people have about the risk
attributed to the pesticides itself (risk nature of pesticides), the
risk perceptions that relate to the role of pesticides in agriculture,
and the perceptions of people about who is to blame for possible
problems with pesticides (blaming).
7. Results of semi-structured interviews and observations (the
interpretive study)
From 28 actors approached for this study, 27 accepted to be
interviewed. The smallholders and farm-workers (n¼8) were
exposed and had low-schooling, while other actors (n¼19) we re
non-exposed and had high-schooling, with the exception of three
large-scale farmers who have low-schooling (Table 1).
7.1. Perceptions of the risk nature of pesticides
The interview data reveal that smallholders gained knowledge
about the health risks of pesticides from different sources. One
important source was their own experiences, which often led to
adaptations in pesticides practices, risk prevention or locally-
rooted views on how to treat pesticide contaminations. For
example, we discovered that smallholder tomato farmers had
stopped using carbofuran (an insecticide/nematicide), which had
A. Ríos-González et al. / Environmental Research 124 (2013) 43–53 45

been used for years in those communities. Asked about why they
had stopped, a smallholder in a group conversation stated: “After
many years of applying this pesticide, we realized that many of us
feel nausea and dizziness after mixing or applying this pesticide,
thus we know that pesticides are not good for our health”(Inter-
view 1, [21–12–2009] all translations of quotes and interview
questions from Spanish to English by the authors). They link these
bodily experiences in the field to their decision to reduce the use
of carbofuran. Likewise idiosyncratic treatments may also be based
on local experiences. For example, the oldest smallholder tomato
farmer stated: “Once I became sick due to the pesticides, I came
home with dizziness and nausea, my wife gave me water with
lemon and that relieved me. That is a good remedy to relieve the
effects of pesticide, thus I advise my friends to relieve the
symptoms with water and lemon”and later he stated “drinking
a glass of milk before applying [pesticides] is good to prevent
intoxication”(Interview 2, [22–12–2009]). During the field work,
we encountered many smallholders who drink a glass of milk
before applying and stated to drink lemon with water when they
feel nausea. We observed women soaking clothes impregnated
with pesticides after applications in water with lemon before they
wash it. They explained that the lemon removes the toxicity of
pesticides (short personal communication in the field, [10–12–
2009]). These practices evidence that for smallholders one source
of knowledge is their own experiences, often, as the second quote
suggests, shared with friends and thus subject to social interaction
at the local level.
This does not mean that all perceptions about health risks that
differ from the technical perspective originate among smallholders
themselves. Some come from people with higher schooling levels.
One of the interviewed farm-workers, when asked if pesticides
imply risks to human health, answered: “Some of us have already
been adapted; I have been growing tomato for a long time, thus I
do not cry and I am not afraid, but there are people who are
quickly affected. (..) Pesticides do not affect equally the people. It
depends on how we are made; those who are adapted are
stronger. All pesticides are toxic but there are people who resist
and others who do not resist”. When asked how he knew that
pesticides do not affect people equally, he replied: “I have heard it
from an extensionist, we were talking about it”(Interview 3, [21–
12–2009]). Although this quote is not by itself sufficient evidence
for considering extensionists as the source of such ideas, it raises
the question about how different or similar the views are between
farm workers and smallholders on the one hand, and extensionists
on the other hand (to be discussed later).
Another issue revealed by the data concerns on how actors read
the label on the package. Interview data suggest that users
interpret the labels differently from the official message that they
contain. For example, some people believe that the shift from red-
labeled pesticides to pesticides with a green or blue label (indicat-
ing a lower acute human toxicity) has reduced the health pro-
blems. “I wonder it is because of the liquids they spray; my mother
in law died of cancer, my father in law too, [like] most of the
people here. (..) The current [pesticides] arrive diluted for the
human: the labels indicate it and tell ‘no toxic’or ‘slightly toxic’
and the letters of the product are green or blue: formerly the
labels were red and with a big skull”(Interview 4, [15–11–2009]).
The farm-worker related the red label to higher levels of cancer.
However, the colors on the labels refer to acute toxicity and not to
cancer. Although the simplicity of the color system has had its
function in pesticide risk management, it has introduced new
meanings not foreseen in the design of the labeling system. In the
same way we found that farmers sometimes considered the colors
to be an indication of effectiveness, whereby red labeled pesticides
were considered more effective (see the discussion below about
the survey). Again this goes against the technical perspective: the
color is not a measure of effectiveness.
7.2. Risk perceptions related to pesticides in agricultural practice
By means of classifying the expressions and observed practices
of actors we identified the type of risk attitude (Table 1). A small
proportion of the interviewees (6 out of 27) showed a ‘risk aware’
attitude. An example of this attitude is the interview and observa-
tions of one smallholder tomato farmer who, even when we have
not yet introduced pesticide risk as topic, stated: “Formerly there
were no such terrible diseases as nowadays. I think that it
[sickness such as cancer] is due to the amount of pesticides we
use in tomato. (..) I even do not want to eat my own tomato. (..) All
of us want someone to teach us how to produce organic tomato; I
think that every tomato farmer wants to stop using pesticides,
because these [pesticides] mean danger for our health”(Interview
5, [28–12–2009]). Besides his wish to stop using pesticides,
expressed during the interview, we also observed him performing
Table 1
Risk perception and attitude according to type of actor, exposure and schooling level.
Expressions observed during the interviews Respondents whose daily
activities confirmed their
expression “there is risk”
Risk attitude
n“There is
risk”
“There is no
or little risk”
Both
expressions
Confirming No
confirming
Aware Disregarding Dissonant
Total actors 27
n
16276106219
Type of actor
Health-professionals 4 –13–––13
Extensionists 4 3 1 ––3–13
Scientists 3 3 ––3–3––
Large-scale farmers 8
n
6–––6––8
Smallholders 4 3 –13–3–1
Farm-workers 4 1 –3–1––4
Exposure
Exposed 8 4 –4313–5
Non-exposed 19 12 23393214
Schooling level
Low-schooling 12 7 –5343–9
High-schooling 15 922363210
n
Two large scale farmers avoided talking about pesticide use, thus we could not determine their expressions regarding “there is risk”or “there is no or little risk”.
In these cases we could through field observations determine their risk attitude as dissonant.
A. Ríos-González et al. / Environmental Research 124 (2013) 43–5346

adapted practices to prevent pesticide risk. He used a specific set
of clothes for applying pesticides composed of a long-sleeved shirt,
pants, gloves and a handkerchief, which he normally left in the
field. Occasionally he took the clothes home where his wife
washed it separately and hung it on a high clothesline to keep
them away from the children. Though these and similar practices
may not be an adequate protection, the mere intention is an
expression of a risk aware attitude.
On the other hand, we also found actors who disregarded
pesticide risks, although this was a minority (2 out of 27). One of
these two was a physician who seemed to give priority to other
types of risk, thereby neglecting what happens in agriculture.
When asked if she thinks that the use of pesticides in the farming
community presents health risks for the population, she answered
“no”. Several of her statements can be interpreted as a narrative to
minimize the importance of pesticide risk: “the cases of cancer
that have been registered are of women, but they do not even
apply pesticides. (..) In Comitán the production of tomato is higher
than here, but the population is not sick. (..) If they [the farmers]
eat tomato, they wash it but they do not remove the bottom,
where the cholera could be present. That is even worse”(Interview
6, [19–12–2009]). In this answer she referred to the idea that
pesticide contamination causes cancer. The shift in her words to
cholera as a much higher risk of tomato consumption indicates her
prioritization of risk. The second risk disregarder in the sample
was an official of the Ministry of Agriculture who strongly stated
that pesticides are crucial for agricultural production “There is no
another way to manage pest control; (…) you will not find any
alternative [different to pesticides] to manage banana pests
because that is not possible; it is like that!”(Interview 7, [14–
12–2009]). This interviewee did not want to talk about pesticide
risks. Both cases are examples of people who prioritize other risks
(infectious diseases, yield reduction) and deny or neglect
pesticide risks.
A larger group of respondents (19 out of 27) combined risk
awareness and disregarding risks; we call this the ‘dissonant risk’
perception. From these respondents 12 out of 19 expressed to be
aware of risk during the interviews or did not want to talk about
pesticides but exposed practices that reject or neglect that there is
risk. For example, a large-scale banana farmer asked us to stop
recording the interview and then in many moments showed risk
aware attitudes in the interview, such as the following statement:
“They [pesticides] are so toxic! It is necessary to use protection
while applying pesticides; therefore in our farm we provide
protective equipment to the workers”(Interview 8, [16–12–
2009]). During the field work, employees of this farm assured
that they comply with the rules of safe use of pesticides estab-
lished in the ‘agricultural good practices protocol’(Secretaria de
Agricultura Ganadería, Desarrollo Rural, Pesca y Alimentación
(SAGRPA), 2002) because they want to obtain a certification that
would make market expansion possible. However, we observed
that the personal protective equipment and showers are not
sufficient for the number of farm-workers who apply pesticides
in this farm. This coincided with what one farm-worker of this
farm stated: “No, [in the farm] they have never given protective
equipment to me (..) no…they have neither said how I must apply
the pesticide”(Interview 9, [13–11–2009]). This expression can be
seen as a rejection of what his employer, the large-scale farmer
quoted above, had stated. Likewise another large-scale farmer
stated: “The pesticides cause harm, therefore we are trying to
reduce the applications in the farm”(Interview 10, [14–12–2009]).
However, during the fieldwork in his farm, the production man-
ager explained that they try to reduce the number of pesticide
applications because pesticide costs put a pressure on profit
margins; they had already shifted to using cheaper pesticides
(communication during the field work, [17–12–2009]). Hence, the
large-scale farmer presented the reduction of pesticide applica-
tions as part of a risk aware discourse, while it was basically a
strategy to diminish the production costs. We consider this as a
dissonance between expressed perceptions in the interview and
practical reasons in the farm.
We also found other forms of dissonant risk attitudes. 7 out of
19 interviewees showed contrasting perceptions within an inter-
view (independently of their practices in the field). These state-
ments in particular informed the formulation of the assertions
included in the survey for quantitative analysis (Table 2). An
example of this form of dissonance is the case of a farm-worker
who shifted his expressions about pesticides from negative while
talking about possible health consequences to positive when
talking about work load. “I think that sooner or later the pesticides
will affect us”. Asked about disadvantages and advantages of
working in the farm he stated: “I do not like it because, for
example, today I dropped everything and it [the pesticide]
splashed on me. (..) The advantage [of applying pesticides in the
farm] is that it is only three hours [from 6:00 to 9:00 am]; that is
why I like it. Well, it is not that I like it. I mean, I prefer that. Even
while I do not like it, I prefer it because it is only three hours and
they [the farm owners] pay the whole day [9.2 dollars]. In this way
I can do other jobs thus I earn more. (..) My son tells me to stop
with this work because it can affect to my health but in this way I
am early at home [to do other jobs]. (..) I have sprayed for many
years and nothing has happened to me”(Interview 4, [15–11–
2009]). Initially, this interviewee expressed that pesticides can
cause harm, which is an indication of a risk aware attitude. When
we moved to talk about pesticide use in his work, he indicated to
prefer applying pesticides. From this moment in the interview he
shifted to giving statements trying to downplay the risk by
alleging that he has not got sick. Such a shift in meaning attributed
to pesticides within a single interview suggests that when people
consider pesticides within the context of agricultural practices and
their socioeconomic situation, this leads to the expression of
dissonant attitudes of risk.
The collected data on risk perceptions showed clear differences
for the type of actor and exposure and less clear differences for
schooling level (Table 1). The risk aware attitude was identified for
the major proportion of smallholders (3 out of 4) and for all
scientists (n¼3), whereas dissonant attitude were identified for all
large-scale farmers (n¼8), farm-workers (n¼4), the major propor-
tion of extensionists (3 out of 4), and the health professionals (3
out of 4). Likewise the risk aware attitude was identified for a
larger proportion of directly exposed actors (3 out of 8) than for
non-exposed actors (3 out of 19). In contrast, interviewee attitudes
according to the schooling level differed slightly. One quarter of
the low-schooling actors showed risk aware attitude (3 out of 12)
whereas only a fifth of the high-schooling actors showed this
attitude (3 out of 15).
7.3. Blaming
We also collected data on how interviewees perceive the
attitude of other actors towards pesticides (Table 1). The data
revealed that 11 out of 19 non-exposed interviewees put all the
blame on pesticide users (smallholders and farm workers). Most of
them disregard pesticide risks and consider themselves blameless.
Expressions to blame users were variations on one idea: “pesti-
cides are not harmful; the problem is not the pesticides but how
the people handle the pesticides”. For example an extensionist
working in a pesticide manufacturer stated “My product [pesti-
cides] aims and is manufactured to control plagues. (..) [But] who
misuses it? [He used an ironical style of expressing] (…) the long
term problems that they [smallholders] have in their intestines,
genitals and nervous central system could be attributed to
A. Ríos-González et al. / Environmental Research 124 (2013) 43–53 47

pesticides but the problem is not the product, it is rather the doses
they use. (..) I have not found any evidence of cancer in our
products. I am an environmental engineer, thus if our products are
dubious, I would be the first to raise the hand”(Interview 11, [25–
01–2010]). First, this extensionist referred to the idea that if users
would apply the recommended doses of pesticides they would not
be at risk. This suggests that the misuse of pesticides is perceived
as much more risky than the substance itself. Second, although he
attributed problems with intestines and some other health pro-
blems to pesticides, he prioritizes carcinogenic pesticides as the
only serious health problem, creating a frame in which pesticides
that are not carcinogenic are not really harmful. Third, on the top
of this, he hinted to be blameless because he is an environmental
engineer, presuming and underlining the superiority of expertise.
One other regularly displayed form of blaming users involved
one or another reference to the “Mexican idiosyncrasy”, a term
which is often used by locals to allude to the stereotype of the
Mexican lounging around a cactus, traditionalist, defiant and
reckless. Some people attribute this stereotype only to poor and
low-schooled people who live in rural areas. Extensionists, large-
scale farmers and health professionals regularly use this frame
when blaming smallholders and farm-workers. For example a
large-scale farmer stated: “We have to face our idiosyncrasy. (..)
People contaminate by throwing away the wrapping of processed
food [he referred to plastic in the streets] because formerly they
did the same with the traditional products [he referred to the
countryside]. (..) Thus, this problem originates in our education”
(Interview 12, [17–12–2009]). In this view coming from rural areas
means to be uncivilized and environmentally unconscious.
Another example was an extensionist working for a pesticide
manufacturer and responsible for risk communication, who sta-
ted: “They [smallholders] do not read the label, they prefer to ask
to their friends. (..) We [he and his colleagues] were in a meeting
to define the design of labels and we coincided that the problem is
cultural. The idea of Mother Nature has taken root among the
peasants very strongly. Therefore, like the Spanish who managed
to conquer us with the virgin Guadalupe [Maria], we will
search for a symbol with which the peasant can feel identified”
(Interview 11, [25–01–2010]). In his analogy between the strate-
gies for communication with smallholders and the establishing of
Spanish dominance during the conquest of Mexico, he presumed
to be the dominant part that has the moral right and power to
reshape peasant's culture. From this perspective, it is culture
rather than the nature of the technology or social relations that
is the underlying prime cause of problems related with pesticide
use (Jansen, 2008).
Smallholders and farm-workers in their turn blamed the
extensionists (3 out of 4), but unlike the latter, pesticide users
also considered themselves as blameworthy. One smallholder
typically blamed extensionists as follows: “In the Ministry of
Table 2
Assertions included in the survey and percentage of extensionists and farmers agreeing with a particular assertion
Type of actor Schooling of farmers
Extensionists Smallholders pElementary
school
completed
Elementary
school not
completed
p
%(n)%(n)%(n)%(n)
Perceptions contrasting with technical perspective
1 Current pesticides are not as harmful as in the
past; nowadays less people die from poisoning
37 (92) 43 (91) .253 38 (45) 50 (44) .172
2 People exposed to pesticides become resistant
to its effects
23 (90) 56 (90) .0 00 51 (23) 58 (25) .327
3 If a label shows a skull, the pesticide is
dangerous. If it does not show a skull it is no so
dangerous
39 (91) 58 (93) .0 06 60 (47) 59 (44) .566
Perceptions regarding place of pesticides within the agricultural context
5 Agriculture is more profitable with the use of
pesticides than without.
66 (92) 83 (97) .0 08 83 (48) 81 (47) .481
6 The effect of pesticides on the environment and
human health are justified by the benefits that
these substances bring to agriculture.
46 (89) 65 (88) .0 09 52 (46) 83 (40) .003
Blaming
7 In case of an accident with pesticides, the user is
to blame.
82 (73) 83 (78) .4 46 83 (48) 84 (43) .593
8 Pesticides are not harmful if they are applied
correctly.
63 (91) 68 (98) .250 61 (49) 77 (47) .080
9 Pesticides harm the health of the farmers,
because they do not protect themselves.
88 (95) _ _ _
10 Extensionists think that we do not use
protection because we do not want to and we
do not know the harm it can cause, but the truth
is that nobody gives us protective equipment
*
64 (89) –––
11 The producers do not apply the pesticides
correctly because they do not know; they lack
training
78 (96) –––
12 Extensionists think they know better than us
how to control pests because they have studied,
but they do not have the experience in the field
as we have.
73 (92) –––
n
The expression “anybody gives us protection”is used by smallholders to refer to the idea that institutions do not provide economical support to buy protective
equipment nor advices them on how to protect themselves.
A. Ríos-González et al. / Environmental Research 124 (2013) 43–5348

Agriculture, they [the engineers] sit cross-legged on their desk,
smoking their cigarette. I feel they should come here. (..) We do
[apply pesticides] with our understanding, but sometimes we do
so wrongly. Once there were farmers who used to spray nitrate
mixed with herbicide, some engineers saw that and said to the
farmers no to do that because they can cause an explosion, but
[mostly] there is nobody who tells us anything (..) We would like
that the engineers teach us organic methods. There is a farmer in
Agronomos [another farming community] who grows organic
maize and has good results”(Interview 5, [28–12–2009]). This
smallholder complained about the lack of attention from exten-
sionists but would also welcome guidance about organic agricul-
ture by an extensionist. A difference between smallholders and
farm workers on the one hand and extensionists on the other hand
is that the former also blame themselves when considering
pesticide-related problems. In the interviews, 3 out of 4 farm-
workers and one smallholder expressed such self-blaming. Inter-
estingly, this self-blaming overlapped with the idea expressed by
non-exposed actors that it is not the pesticides that are harmful
but that the user has to be blamed for any harm caused by
pesticide handling. An example of self-blaming was given by a
farm-worker: “Sometimes at the end of the journey my neck and
my chest turn red and itch, just like this [he showed us the rash on
his chest]. I do not know what it is. I believe that it is because I
touch the bags [those impregnated with pesticides] and all that
having contact with substances [pesticides] and that kind of stuff.
However, it must also be because I do not always use protection
[he uses a tone showing shame]. I only use a pair of gloves
sometimes, but it hampers my work and makes me sweat even
more. (Interview 13, [13–11–2009]). Such forms of little self-
blaming occur often in interviews.
The frame of self-blaming can also be partly incorporated but at
the same time ironized. One farm-worker who described an
accident in the farm stated: “A colleague farm-worker became
sick, he used to pick up the bags [those impregnated with
pesticide], we told him to ask for a mask…[silence, expressing
hhhummm, and continuing with a cynical tone in his voice]: in the
farm they give us boots. Everybody there uses boots, thus if we get
sick it is our fault. I have been years working there and I have
never got sick”(Interview 14, [27–11 –2009]). This farm worker
suggested that it is your own fault when you get sick because you
do not wear protective equipment. At the same time he talked
ironically about the notion that with the availability of some
personal protective equipment the worker will be blamed for
accidents and damage, rather than the farm owner.
8. Results of the survey
We did a preliminary analysis of interviews in which we
identified the following interesting issues: perceptions that con-
trast with the technical perspective, the reference to the role of
pesticides in agricultural production and the importance of
blaming people for negative effects of pesticide use. From the
items of the survey we selected the 12 assertions with the most
remarkable results (Table 2). Some of these assertions contrast
with the typical technical perspective. The schooling level and the
exposure t pesticides were very different between the two groups:
farmers and extensionists. All extensionists had university degree
whereas most smallholders barely finished the elementary
school (63% had not finished elementary school and 36% had
entered, though not necessarily finished high school). Whereas
extensionists are occasionally exposed to pesticides (once per
month), all smallholders are exposed almost daily during the
growing season of tomatoes. In an initial analysis we noticed that
the distribution of responses showed clear tendencies. For many
assertions the majority of the respondents chose for “disagree”
(option 2) or “agree”(option 4), a minor proportion chose
“strongly disagree”(option 1) and “strongly agree”(option 5)
and only few respondents chose “neither agree nor disagree”
(option 3). To show the contrast we eliminated the responses of
option 3 and we categorized options 4 and 5 as agreement and
options 1 and 2 as disagreement.
8.1. Perceptions contrasting with the technical perspective
We hypothesized that extensionists would have ideas that
approximate the technical view considering that they are trained
in pesticide toxicity and thus would disagree with assertions that
go against the technical perspective. However, an important
percentage of extensionists expressed ideas that deviate from
the technical perspective (Table 2, assertions 1–4). Agreement
with the idea of “Current pesticides are not as harmful as in the
past; nowadays less people die from poisoning”were almost
similar for both groups (43% and 37%, respectively, p¼.253).
Although more smallholders than extensionists agreed with “Peo-
ple exposed to pesticides become resistant to its effects”(56% and
23%, respectively, p¼.000), the fact that almost a quarter of the
extensionists agreed is remarkable. This result coincides with the
finding of the semi-structured interviews suggesting that high-
schooled actors may support ideas that differ from the technical
perspective (e.g. Interview 3).
Just as it was found in the semi-structured interviews (e.g.
Interview 4), both extensionists and smallholders appeared to
interpret the labeling differently from the original intentions of
the designers of the labeling system (Table 2, assertions 3–4).
Smallholders agreed more than extensionists with two ideas: “If a
label shows a skull, the pesticide is dangerous. If it does not show a
skull it is no so dangerous”(58% and 39%, respectively, p¼.006)
and “Pesticides with a red band are the most hazardous but also
the most effective against pests”(80% and 64%, respectively,
p¼.013). Assuming that the lower percentages for the extensio-
nists is an expression of better technical knowledge (red band and
skull only inform about acute toxicity, not overall hazardousness;
band color is no indication of effectiveness) it is remarkable that
still a high percentage of the extensionists agree with these
assertions.
8.2. Perceptions regarding the role of pesticides in agricultural
production
In the semi-structured interviews the interviewees often
expressed dissonant risk perception when they started to talk
about the role of pesticides in agricultural production (e.g. Inter-
view 4). Especially smallholders agreed with assertions that justify
the risks they take when applying pesticides (assertions 5–6in
Table 2), such as “Agriculture is more profitable with the use of
pesticides than without”(83% and 66%, respectively, p¼.008) and
“The effect of pesticides on the environment and human health are
justified by the benefits that these substances bring to agriculture”
(65% and 46%, respectively, p¼.009). It is noteworthy that both
assertions may have a particular meaning for each group of actors.
Agriculture is the major source of livelihood for smallholders and
increasing yields is an important issue for them. Therefore
smallholders may be more eager to seek justification for the use
of pesticides than extensionists, who do not depend economically
on agricultural production. The position of the actor in the
agricultural process seems to be important in shaping risk
perception.
A. Ríos-González et al. / Environmental Research 124 (2013) 43–53 49

8.3. Blaming the user and the other
The notion that the problem is not the pesticide but how
people handle pesticides (blaming the user for misuse), as identi-
fied in the semi-structured interviews, is actually shared by
smallholders and extensionists (Table 2, assertions 7 and 8). While
this idea means that extensionists put the blame on the other, for
smallholders it means self-blaming. Interestingly, the results of the
survey showed that a similar percentage of smallholders and
extensionists agreed with the idea that “In case of an accident
with pesticides the user is to blame”(83% and 82%, respectively,
p¼.446). Likewise, similar percentages of smallholders and exten-
sionists agreed with the idea “Pesticides are not harmful if they are
applied correctly”(68% and 63%, respectively, p¼.250). This
supports the idea that if harm occurs, the reason is incorrect
handling (misuse).
The survey results confirm the finding of the semi-structured
interviews that smallholders and extensionists blame each other
for the lack of personal protection (Table 2, assertions 9–12).
Whereas the majority of extensionists (88%) agreed with the idea
“Pesticides harm the health of farmers, because they do not
protect themselves”, the majority of smallholders (64%) agreed
with “Extensionists think that we do not use protection because
we do not want to and we do not know the harm it can cause, but
the truth is that anybody gives us protective equipment”. Likewise,
we found that both groups questioned the knowledge of the other.
A majority of extensionists (78%) agreed with the “The producers
do not apply pesticides correctly because they do not know; they
lack training”. A majority of the smallholders (73%) agreed with
“Extensionists think they know better than us how to control pests
because they have studied, but they do not have the experience in
the field as we have”. These results not only show that extensio-
nists and smallholders blame each other, but also indicate that
they have different views on who has better knowledge and no
confidence in what the other actor knows.
The smallholders and extensionists in this study have very
different schooling levels and levels of exposure to pesticides, so
the differences between both groups corresponded also to the
differences according to schooling level and exposure. To deepen
the analysis of schooling we also compared the schooling level of
smallholders. In almost all assertions smallholders who did not
complete elementary school and smallholders with elementary
school completed showed similar levels of agreement, with
exception of assertions 6 and 8. Smallholders with elementary
school not completed agreed more with “The effect of pesticides
on the environment and human health are justified by the benefits
that these substances bring to agriculture”(83% and 52%, respec-
tively, p¼.003) and with “Pesticides are not harmful if they are
applied correctly”than smallholders with elementary school
completed (77% and 61%, respectively, p¼.080). A low-schooling
level coincides with smallholders justifying the use of pesticides
and expressing self-blame.
9. Discussion
Risk communication is seen as a key element to promote the
safe handling of pesticides, whereby much emphasis is given to
labeling and education as instruments for improving such com-
munication (Hashemi et al., 2012;Lichtenberg and Zimmerman,
1999). The results of our research provide new insights about such
communicative activities. In this section we first discuss the issue
of labeling and then the role of formal education in understanding
risk information. After that we discuss the relationship between
experts and lay people, focusing on the constraining role of
blaming as discursive strategies. Finally, this discussion helps us
to formulate an alternative model for the classical model of
thinking about the knowledge deficit of pesticide users.
This study offers insight into the discussion of how to compose
pesticide labels. Several studies have documented that users often
do not read the information; one reason for this is that informa-
tion is often too technical and not understandable (Damalas et al.,
2006;Waichman et al., 2007). Particularly in conditions of devel-
oping countries with illiterate users or users with less formal
education, one has tried to solve this problem by using simplifying
symbols, e.g. as pictograms or the color of the label. The use of a
red band on the label of highly hazardous pesticide with a high
acute toxicity is well known. Our study confirms Rother's (2008)
finding that also the simple pictograms may be misunderstood.
The results of our study suggest that simplification may lead to
new misunderstandings and false images of safety and risks (e.g.
that red labels are more effective for pest control). There is no
simple solution to this dilemma. Some studies attribute farmers'
lack of understanding of the label to a low schooling level or lack
of training (Damalas et al., 2006;Mokhele, 2011). In contrast, this
study found that label interpreters with a higher schooling level
such as extensionists who have received training in pesticide use,
also misunderstand the pesticide labeling in unexpected ways.
This study also provides data that reflects the relationship
between the level of formal education (schooling level) and a
correct interpretation of pesticide information. The results show
that an unexpected high number of extensionists express attitudes
that are dissonant with the mainstream technical perspective. Like
smallholders, many extensionist share views that can be consid-
ered incorrect from a technical perspective (although the percen-
tage that deviates from the technical perspective is different for
smallholders and extensionists). This study demonstrated that the
schooling level is not necessarily related with a more correct risk
perception of pesticides, reflected in the little difference in risk
perception of high-schooled and low-schooled actors (Table 1) and
the finding that not only farmers but also extensionists have ideas
that differ from the technical perspective (assertions 1–4, Table 2).
Moreover, there were no significant differences in the opinions of
smallholders with elementary school completed and smallholders
with elementary school not completed (with the exception of
assertion 6). These results differ with other studies that suggest a
more direct, linear relationship between schooling level and risk
perception (Yassin et al., 2002;Hashemi et al., 2012;Blanco-
Muñoz and Lacasaña, 2011;Ibitayo, 2006;Salameh et al., 2004;
Sam et al., 2008;Mokhele, 2011). For example, one study found
that in Mexico workers with an education level above elementary
school are more aware about pesticide risks than those with a
lower level (85.7% versus 52.9%, p¼.04) (Blanco-Muñoz and
Lacasaña, 2011). One reason for the difference in findings may be
that we developed a different methodological approach. Although
we compared farmer's expressions with a mainstream technical
perspective we did not a priori categorize farmers'expressions
from a normative perspective that classifies them as technically
‘correct’or ‘not correct’(Yassin et al. 2002., Hashemi et al., 2012,
Blanco-Muñoz and Lacasaña, 2011). In our study we used a
sociological approach whereby we first just recorded pesticide
perceptions from a more neutral position, i.e. formulating asser-
tions based on people's own ways to express the matters rather
than following the ‘technical correct’approach. More formalized,
technical correct questioning in other studies could, instead,
privilege the higher-schooled farmers, whereas drawing more
‘incorrect’answers from the lower-schooled farmers. Added to
this, we also interviewed extensionists in the same way, taking a
symmetrical perspective as much as possible (most other studies
only interviewed farmers or farm workers). Only at a later stage in
the analysis we compared the given expressions with a main-
stream technical perspective. If our argument is valid that a more
A. Ríos-González et al. / Environmental Research 124 (2013) 43–5350

formal education does not lead in a linear way to more ‘correct’
pesticide perceptions (i.e. in line with the technical perspective), a
more dynamic view on pesticide knowledge will have to be
developed (in which formal education may just be one element).
If the level of formal knowledge is not the single explaining
factor how much of the technical perspective is followed by actors,
we have to look at other factors. One major factor that appears to
be important in our study is the position of the actor in the labor
process. By this we mean how people make a living and the role
that pesticides play in this process. Justifying pesticides by
smallholders has been attributed to the poor conditions of life
and dependence of smallholders on agricultural production (Kishi
et al., 1995,Hunt et al., 1999,Recena and Caldas, 2008) and their
situation of vulnerability (Fonseca et al., 2007). Our study suggests
that many users do not just deny pesticides risks even if their
short term income is dependent on working with pesticides. But
we also observed that the latter experiences influence the
expressed perceptions. In the qualitative interviews we observed
shifts in expressions when the interview evolved from talking
about the properties and effects of pesticides to the issue why
people apply pesticides. In the first moment, many respondents
expressed to be aware about possible health effects, using negative
qualifications for pesticides, whereas in the second moment they
used less negative qualifications of pesticides or only positive ones,
shifting to language expressing the need to apply, either to have
the spraying job as farm-worker or to save the harvest as
smallholder. Therefore we argue that neglection of pesticide risks
in a dissonant risk attitude is less driven by a lack of knowledge as
by strategies to make a living in the given conditions. Both
smallholders and larger farmer may be locked into the use of
pesticides to save the harvest and thus their investments. The role
of actors in the labor process also helps to understand the
apparently rejection of pesticide risks by the health professionals
who displayed a dissonant attitude. Their epidemiological per-
spective informs them that diarrhea or tract infections are a much
larger risk than pesticides as diarrheas frequently reach epidemic
proportions.
The research results indicate that attributing a lack of knowl-
edge to a particular group is a discursive strategy to blame the
other (humans rather than the technology) for harm resulting
from pesticide use. The results have revealed that extensionists
and scientists expressed themselves in a more consistent way
about pesticide risks and use than the ‘non-experts’(Table 1).
Large-scale farmers also expressed a consistent narrative. We
hypothesize that scientist, large-scale farmers and extensionists
are used to defend their position in a discussion about pesticides,
meanwhile smallholders and farm-workers are not used to
express their insights on pesticides linguistically and, thus, may
express dissonances. Therefore, smallholders and farm-workers
may appear as more inconsistent. This could lead to the suggestion
that they need to require more (formal) knowledge. As we have
argued above, blaming users by referring to a lack of knowledge is
rooted in a reductionist explanation of the nature of pesticide
knowledge of smallholders and farm-workers (Blok et al., 2008,
refer in this context to the ‘imagined’lay-person) and a more
comprehensive model is needed. The process of boundary setting
between those who know, or are supposed to know, and those
who lack knowledge relates to the discussion on the public
understanding of science and the division and relationships
between ‘experts’and ‘lay people’(Blok et al., 2008;Jansen,
2008). Our results point at boundary setting processes, in
particular by extensionists who blame the users for problems.
By doing so, they reproduce the model of a large expert-lay
distance. Smallholders also reproduce this distance by showing
mistrust in their own knowledge. But smallholders also tend to
point at some typical faults of extensionists, thus developing a
double position: extensionist should know and deliver (advice but
also protective equipment) but in real life they often fail to do so.
Despite farmers'questioning of the role of extensionist, they
demanded training from the last actors. It is worth to notice that
the concept ‘training’can differ from both groups (experts-lay)
(Binder and Schöll, 2009). It is likely that existing training concepts
will fail as they are based on the classical expert-lay division and
the model in which pesticide users are just seen as recipients of
formal knowledge and have no knowledge, skills or experiences to
tap into.
This discussion brings us to the point where we have to
develop a broader and more refined model of knowledge about
pesticide risks that goes beyond a simple knowledge deficit model
(Jansen et al., 2004)(Fig. 1). We will call this refined model the
knowledge-in-context model. The first element is that formal
education and the congruence between the expressed perception
and the technical perspective are not (or not only) the standards to
assess people's knowledge about pesticide risks. Rather an analysis
is needed of knowledge in its context in order to understand why
people, for example smallholders and farm-workers, express
dissonant attitudes. Both individual conditions (e.g. livelihood
situation) as well as more systemic factors, such as the organiza-
tion of the labor process or the dominance of particular risk
narratives, shape the knowledge that people express (formal
education is just one element of this context). The second element
is that experiences of people inform their views. Farmers who
have suffered accidents with pesticides perform more safety
practices than those who have not suffered accidents (Feola and
Binder, 2010). Our study shows that personal experiences, often in
combination with social interaction (talking among farmers), led
farmers to adapt their pesticide safety practices. Experiences
mentioned were not only practical experiences with pesticides
(e.g. accidents in the field), but also experiences with ‘institutions’,
such as the distrust of extensionists discussed above. The third
element of our knowledge-in-context model is the important role
of larger knowledge networks that strongly shape pesticide use
perspectives (Toleubayev et al., 2010), and may induce changes in
behavior (Barraza et al., 2013;Toleubayev et al., 2011). In our
research the current social discourse on organic farming appeared
to be a point of reference for many farmers and has produced
shifts in their thinking about pesticides (not necessarily in their
practices of pesticide use). This element also links individual
oriented research on risk perception to approaches that identify
and analyze the more collective social processes that shapes
peoples thinking. For example, former banana plantation workers
in Central America have filed collectively law suits against the
manufacturers of dibromochloropropane (Bohme, 2011;
Rosenthal, 2004). Galt (2007,2008) has analyzed how farmer
perspectives of risk emerge and adapt to the concrete political
economic situation.
A knowledge-in-context model will have consequences for
ideas about training in pesticide safety measures that is now
predominantly based in the knowledge deficit/ignorant user
model. Calls for more training about risks and safety measures
that are uncoupled from further analysis of peoples view on the
context of pesticide use, including the social relationships that
shape how people practice agriculture, may not change attitudes.
Methods based on guiding farmers to share their own experiences
and learn from it could be more effective in pesticide risk
communication. Our study shows that (similar to the finding of
Damalas et al., 2006) other farmers are a principal source of
information about safety measures, rather than extensionists or
information on the label. It confirms the notion that to bring about
changes in pesticide practices requires multiple actions at different
levels, ranging from the intra- and interpersonal level to market
and political structures (Cole et al. 2011;Orozco et al., 2011).
A. Ríos-González et al. / Environmental Research 124 (2013) 43–53 51

10. Conclusions
In this study we analyze from a social science perspective the
knowledge and risk perception of pesticides of different actors.
Smallholders often derive their risk perception from direct experi-
ences with handling pesticides, whereas extensionists draw to a
larger extent on formal education. Notwithstanding these differ-
ences, risk perceptions of extensionist not always coincide with
the technical perspective as taught in formal education. On the
other hand, smallholders'knowledge acquired through practical
experiences has often resulted in attempts to take safety measures
while handling pesticides. Our study shows that blaming is a social
practice that creates a difference between experts and lay people
regarding who follows good practices. However, despite the social
construction of a contrast between these two groups, we found
that commitment to the technical perspective is not dramatically
different for extensionists on the one hand and farmers and farm
workers on the other hand. Finally, we conclude that ideas about a
supposed lack of knowledge by farmers and the need for formal
training are too much driven by a deficit model of knowledge. In
contrast we propose a knowledge-in-context model for analyzing
people's risk perception and consider that it will have conse-
quences for training in pesticide risk and risk reduction strategies,
for example that training in pesticide risks and safety practices
should focus more on real life conditions, people's own experi-
ences, and local forms of knowledge transmission.
Acknowledgments
We express our gratitude to the National Council on Science
and Technology and the National Institute of Public Health of
Mexico for their financial support as well as to the farmers,
farming communities and the persons who accepted to be
interviewed.
References
Barraza, D., Jansen, K., van Wendel de Joode, B., Wesseling, C., 2011. Pesticide use in
banana and plantain production and risk perception among local actors in
Talamanca, Costa Rica. Environ. Res. 111 (5), 708–717.
Barraza, D., Jansen, K., vanWendeldeJoode, B., Wesseling, C., 2013. Social move-
ments and risk perception: unions, churches, pesticides and bananas in Costa
Rica. Int, J. Occup. Med. Environ. Health 19 (1), 11–21.
Binder, C.R., Schöll, R., 2009. Structured mental model approach for analyzing
perception of risks to rural livelihood in developing countries. Sustainability 2,
1–29.
Blanco-Muñoz, J., Lacasaña, M., 2011. Practices in pesticide handling and the use of
personal protective equipment in Mexican agricultural workers. J. Agromed. 16
(2), 117–126.
Blok, A., Jensen, M., Kaltoft, P., 2008. Social identities and risk: expert and lay
imaginations on pesticide use. Public Underst. Sci. 17 (2), 189–209.
Bohme, S.R., 2011. Preserving a forum for transnational litigation is a public health
imperative. Int. J. Occup. Med. Environ. Health 17 (1), 95–97.
Cole, D., Orozco, F., Pradel, W., Suquillo, J., Mera, X., Chacon, A., Prain, G.,
Wanigaratne, S., Leah, J., 2011. An agriculture and health inter-sectorial research
process to reduce hazardous pesticide health impacts among smallholder
farmers in the Andes. BMC Int. Health Hum. Rights 11 (Suppl. 2), S6.
Damalas, C.A., Theodorou, M.G., Georgiou, E.B., 2006. Attitudes towards
pesticide labelling among Greek tobacco farmers. Int. J. Pest Manage. 52 (4),
269–274.
European Union (EU) (Ed.), 2003. Regulation of the European Parliament and of the
Council no. 304/200. OJEU. European Union Parliament. Brussels, Belgium.
Feola, G., Binder, C.R., 2010. Why don't pesticide applicators protect themselves?
Exploring the use of personal protective equipment among Colombian small-
holders. Int. J. Occup. Med. Environ. Health 16 (1), 11–23.
Fonseca, M.G.U., Peres, F., Oliveira-Araújo, F.J., Uchoa, E., 2007. Percepção de risco:
maneiras de pensar e agir no manejo de agrotóxicos. Ciência en Saúde Coletiva
12 (1), 39–50.
Galt, R.E., 2007. Regulatory risk and farmers'caution with pesticides in Costa Rica.
Trans. Inst. Br. Geogr. 32 (3), 377–394.
Galt, R.E., 2008. Beyond the circle of poison: significant shifts in the global pesticide
context. Global Environ. Change 18 (4), 786–799.
Guivant, J., 2003. Pesticide use, risk perception and hybrid local knowledge: a case
study from Southern Brazil. Int. J. Sociol. Agric. Food 11 (1), 41–51.
Hashemi, S.M., Hosseini, S.M., Hashemi, M.K., 2012. Farmer's perceptions of safe use
of pesticides: determinants and training needs. Int. Arch. Occup. Environ.
Health 85 (1), 57–66.
Hunt, L.M., Ojanguren, R.T., Schwartz, N., Halperin, D., 1999. Balancing risks and
resources: applying pesticides without using protective equipment in Southern
Mexico. In: Hahn, R.A. (Ed.), Anthropology in Public Health: Bridging Differ-
ences in Culture and Society. Oxford University Press, New York, USA,
pp. 235–254.
Ibitayo, O.O., 2006. Egyptian farmer's attitudes and behaviors regarding agricultural
pesticides: implications for pesticide risk communication. Risk Anal. 26 (4),
989–995.
Idrovo, J.A., 2005. Hacia una Salud Pública Pluralista: El Caso de los Plaguicidas y la
Salud Humana. Revista de Salud Pública 7 (3), 349–359.
Instituto Nacional de Estadística Geografía e Informática (INEGI). (INEGI) (Ed.),
2010. Anuario Estadístico 2010. Aguascalientes, Mexico.
Isin, S., Yildirim, I., 2007. Fruit-growers'perceptions on the harmful effects of
pesticides and their reflection on practices: the case of Kemalpasa, Turkey. Crop
Prot. 26 (7), 917–922.
Jansen, K., 2003. Crisis discourses and technology regulation in a weak state:
Responses to a pesticide disaster in Honduras. Dev. Change 34 (1), 45–66.
Jansen, K., Vellema, S., 2004. Agribusiness and Society: Corporate Responses to
Environmentalism Market Opportunities and Public Regulation. Zed Books,
London.
Jansen, K., 2008. The unspeakable ban: the translation of global pesticide govern-
ance into Honduran national regulation. World Dev. 36 (4), 575–589.
Jansen, K., Leeuwis, C., Mani, S., Roquas, E., Skutsch, M., Verschoor, G., 2004.
Unpacking and re-packing knowledge in development. In: Kalb, D., Pansters,
W., Siebers, H. (Eds.), Globalization and Development. Themes and Concepts in
Current Research. Kluwer, Dordrecht, pp. 163–190.
Kishi, M., Hirschhorn, N., Djajadisastra, M., Satterlee, L.N., Strowman, S., Dilts, R.,
1995. Relationship of pesticide spraying to signs and symptoms in Indonesian
farmers. Scand. J. Work Environ. Health 21 (2), 124–133.
Knowledge
about pesticide
risks
Formal education
Position in the labour
process/livelihood issues
Bodily experiences of
pesticide effects
Knowledge acquired
through social interaction
Trust in institutions
(including trust in extensionists)
regulation (e.g. new labels, prohibition of
Indirect messages embedded in
certain pesticides)
Fig. 1. Knowledge-in-context model.
A. Ríos-González et al. / Environmental Research 124 (2013) 43–5352

Lichtenberg, E., Zimmerman, R., 1999. Information and farmers'attitudes about
pesticides, water quality, and related environmental effects. Agricult. Ecosys.
Environ. 73 (3), 227–236.
Mokhele, T.A., 2011. Potential health effects of pesticide use on farmworkers in
Lesotho. S. Afr. J. Sci. 107, 7–8.
Orozco, F.A., Cole, D.C., Ibrahim, S., Wanigaratne, S., 2011. Health promotion
outcomes associated with a community-based program to reduce pesticide
related risks among small farm households. Health Promot. Int. 26 (4),
432–446.
Palis, F.G., Flor, R.J., Warburton, H., Hossain, M., 2006. Our Farmers at risk:
behaviour and belief system in pesticide safety. J. Pub. Health 28 (1), 43–48.
Polidoro, B.A., Dahlquist, R.M., Castillo, L.E., Morra, M.J., Somarriba, E., Bosque-Pérez,
N.A., 2008. Pesticide application practices, pest knowledge, and cost-benefits of
plantain production in the Bribri-Cabécar Indigenous Territories, Costa Rica.
Environ. Res. 108 (1), 98–106.
Recena, M.C., Caldas, E.D., 2008. Risk perception, attitudes and practices on
pesticide use among farmers of a city in Midwestern Brazil. Rev. Saude Publica
42 (2), 294–301.
Rosenthal, E., 2004. The DBCP pesticide cases: seeking access to justice to make
agribusiness accountable in the global economy. In: Jansen, K., Vellema, S.
(Eds.), Agribusiness and Society: Corporate Responses to Environmentalism,
Market Opportunities and Public Regulation. Zed, London, pp. 176–199.
Rother, M.A., 2008. South African farm worker's interpretation of risk assessment
data expressed as pictograms on pesticide labels. Environ. Res. 108 (3),
419–427.
Salameh, P.R., Baldi, I., Brochard, P., Saleh, B.A., 2004. Pesticides in Lebanon: a
knowledge, attitude, and practice study. Environ. Res. 94 (1), 1–6.
Sam, K.G., Andrade, H.H., Pradhan, L., Pradhan, A., Sones, S.J., Rao, P.G., Sudhakar, C.,
2008. Effectiveness of an educational program to promote pesticide safety
among pesticide handlers of South India. Int. Arch. Occup. Environ. Health 81
(6), 787–795.
Secretaria de Agricultura Ganadería, Desarrollo Rural, Pesca y Alimentación
(SAGRPA) (Ed.), 2002. Manual de Buenas Prácticas Agrícolas. México DF.
Solís, D.V., 2005. Rural Chiapas ten years after the armed uprising of 1994: an
economic overview. J. Peasant Stud. 32 (3-4), 461–483.
Tinoco-Ojanguren, R., Halperin, D.C., 1998. Poverty, production, and health: inhibi-
tion of erythrocyte cholinesterase via occupational exposure to organopho-
sphate insecticides in Chiapas, Mexico. Arch. Environ. Health 53 (1), 29–35.
Toleubayev, K., Jansen, K., Van Huis, A., 2010. Knowledge and agrarian de-
collectivisation in Kazakhstan. J. Peasant Stud. 37 (2), 353–377.
Toleubayev, K., Jansen, K., Van Huis, A., 2011. From integrated pest management to
indiscriminate pesticide use in Kazakhstan. J. Sustain. Agric. 35 (4), 350–375.
Ton, G., 2012. The mixing of methods: a three-step process for improving rigour in
impact evaluations. Evaluation 18 (1), 5–25.
Waichman, A.V., Eve, E., Nina, N.S.C., 2007. Do farmers understand the information
displayed on pesticide product labels? A key question to reduce pesticides
exposure and risk of poisoning in the Brazilian Amazon. Crop Prot. 26 (4),
576–583.
Wynne, B., 1996. The sheep safely graze? A reflexive view of the expert-lay
knowledge divide. In: Lash, S., Szerszynski, B., Wynne, B. (Eds.), Risk Environ-
ment and Modernity Towards a New Ecology. Sage, London, pp. 44–81, May.
Yassin, M.M, Abu Mourad, T.A., Safi, J.M., 2002. Knowledge, attitude, practice, and
toxicity symptoms associated with pesticide use among farm workers in the
Gaza Strip. Occup. Environ. Med. 59 (6), 387–393.
A. Ríos-González et al. / Environmental Research 124 (2013) 43–53 53