- Email: [email protected]
Danish dairy farmers’ perception of biosecurity
Preventive Veterinary Medicine 99 (2011) 122–129
Contents lists available at ScienceDirect
Preventive Veterinary Medicine journal homepage: www.elsevier.com/locate/prevetmed
Danish dairy farmers’ perception of biosecurity Erling Kristensen a,b,∗ , Esben B. Jakobsen b a b
Knowledge Centre for Agriculture, Cattle, Agro Food Park 15, DK-8200 Aarhus N, Denmark StrateKo Aps, Risagervej 5, DK-6690 Gording, Denmark
a r t i c l e
i n f o
Article history: Received 3 November 2009 Received in revised form 26 January 2011 Accepted 26 January 2011 Keywords: Biosecurity Social dilemma Q-methodology
a b s t r a c t To implement biosecurity measures at farm-level is a motivational challenge to dairy farmers as emerging diseases and their consequences largely are unpredictable. One of the reasons for this challenge is that outcomes are more likely to benefit society than the individual farmer. From the individual farmer’s point of view the impacts of zoonotic risk, international trade and welfare concerns appear less obvious than the direct costs at farm-level. Consequently, a social dilemma may arise where collective interests are at odds with private interests. To improve biosecurity at farm-level farmers must be motivated to change behavior in the ‘right’ direction which could provide selfish farmers with unintended possibilities to exploit the level of biosecurity provided by other dairy farmers’ collective actions. Farmers’ perception of risk of disease introduction into a dairy herd was explored by means of Q-methodology. Participating farmers owned very large dairy herds and were selected for this study because Danish legislation since 2008 has required that larger farms develop and implement a farm specific biosecurity plan. However, a year from introduction of this requirement, none of the participating farmers had developed a biosecurity plan. Farmers’ perception of biosecurity could meaningfully be described by four families of perspectives, labeled: cooperatives; confused; defectors, and introvert. Interestingly, all families of perspectives agreed that sourcing of animals from established dealers represented the highest risk to biosecurity at farm-level. Farmers and policy-makers are faced with important questions about biosecurity at farmlevel related to the sanctioning system within the contextual framework of social dilemmas. To solve these challenges we propose the development of a market-mediated system to (1) reduce the risk of free-riders, and (2) provide farmers with incentives to improve biosecurity at farm-level. © 2011 Elsevier B.V. All rights reserved.
1. Introduction In 2008 it became Danish law that biosecurity plans were mandatory in large dairy herds (>330 cows per year). Supposedly, the biosecurity plans should force farmers to
∗ Corresponding author at: Knowledge Centre for Agriculture, Cattle Health, Welfare and Food Safety, Agro Food Park 15, DK-8200 Aarhus N, Denmark. Tel.: +45 4160 4715; fax: +45 8740 5010. E-mail addresses: ekr@vfl.dk (E. Kristensen), [email protected] (E.B. Jakobsen). 0167-5877/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.prevetmed.2011.01.010
improve on precautionary measures, disease surveillance and controls in order to reduce the risk of introducing animal diseases into the dairy herd and minimize the impact of outbreaks, should they occur. This was in line with the intentions described in the EU initiative, Animal Health Strategy (EU, 2007). However, to motivate farmers into changing their daily routines is a well-known challenge (Burton et al., 2008; Garforth, 2010) and biosecurity measures seem particularly difficult to implement by law as outcomes are more likely to benefit society than the individual dairy farmer, i.e. a social dilemma. Social dilemmas are situations characterized by two factors: (1) at any given
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
decision point, individuals receive higher payoffs for making selfish choices than they do for making cooperative choices regardless of the choices made by those with whom they interact, and (2) everyone involved receives lower payoffs if everyone makes selfish choices than if everyone makes cooperative choices (Mulder et al., 2005). The situation in Denmark brought two social dilemmas into the open: (1) farmers were forced to prioritize between shortterm selfish interests or the long-term interests of dairy farming, i.e. collective interests are at odds with private interests, and (2) farmers were forced to make a decision to defect or accept legislation, i.e. a surveillance and sanctioning dilemma. Interesting to this context is the discussion by Gächter and Fehr (1999) on how subjects’ beliefs about other subjects’ contributions to a social dilemma situation will ultimately determine how much they contribute. Weber et al. (2004) related this dilemma to group mechanisms, involving emotions as fear or greed. Further, being part of a group may define and frame a social dilemma to the group as a situation in which selfish behavior is unlikely to be sanctioned or punished socially (Dawes and Messick, 2000). Farmers’ decision making is not confined to herd health (Stott and Gunn, 2008). Consequently, the underlying assumption behind rational choice models, i.e. the vigilant and calculating decision maker focused on maximizing his expected payoff is unlikely to give a valid picture of farmers’ attitudes and behavior (Valeeva et al., 2007). In fact, rational choice models often downplay social influence processes and overall utility which limit the explanatory power of such models when applied to most social dilemmas (Weber et al., 2004; Rat-Aspert and Fourichon, 2010). It follows, that rational choice models have little predictive effect on farmers’ attitudes and subsequent behavior at farm-level (Burton, 2004; Jansen et al., 2009). Rather, as discussed by Janz and Becker (1984), peoples (preventive) measures are determined by (1) the perceived threat (perceived vulnerability and perceived severity) and (2) the perceived effectiveness of proposed measures (perceived benefits and barriers). Identical findings have been reported from the field of animal health by Heffernan et al. (2008) where farmers’ mutual mistrust (and general mistrust in the authorities) prevented improvements of collective biosecurity measures. Lately, herd health management has been characterized by an integrated, holistic, proactive, data-based and economically framed approach to preventive medicine (LeBlanc et al., 2006) combining disciplines like sociology, psychology, behavior science and communication with veterinary epidemiology (Kristensen and Enevoldsen, 2008) into what we suggest should be labeled ‘social epidemiology’ (Kristensen and Jakobsen, 2010). This approach offers new methodological possibilities to animal and herd health researchers as the social sciences have a longer tradition to account for individual differences, which may be viewed as central to (1) understand the cognitive dissonance between farmers’ biosecurity attitudes and their behavior, and (2) to tailor communication to groups of farmers sharing opinions on biosecurity (Heffernan et al., 2008; Jansen et al., 2010).
123
The objectives of this study were: (1) to explore dairy farmers’ perception of risk of introducing disease into a dairy farm; (2) to contribute to the discussion about social dilemmas in a surveillance and sanctioning system that presently is being imposed on Danish dairy farmers, and (3) to suggest a solution to social dilemmas associated with biosecurity that improves biosecurity at farm- and national level. 2. Materials and methods In this study we explored dairy farmers’ perception of risk of introducing disease into dairy farm by asking the participating farmers to rank a number of statements on a layout guide according to their individual perception of the term ‘biosecurity’. The core research tool of this study was Q-methodology, which was first described by Stephenson (1935) and provides a foundation for the systematic study of subjectivity that is ‘a person’s viewpoints, opinion, beliefs, attitude, and the like’ (Brown, 1993). It follows that Q-methodology does not aim at estimating proportions of different views held by the ‘farmer population’ (Brouwer, 1999). Rather, Q identifies qualitative categories of thought shared by groups of respondents, i.e. dairy farmers. The applied research methodology followed the approach described in Kristensen and Enevoldsen (2008). 2.1. Study population In a stratified design (to make sure that all regions in Denmark was represented) including reflections on sample size, as discussed by Onwuegbuzie and Leech (2007), we selected 27 farmers from a list of 168 farmers with herds larger than 330 cows per year. Within each region, we selected the largest farms and a number of farms that largely reflected cow density in Denmark. To provide the study population with the largest possible comprehensiveness we tried to avoid including more than one farm associated with the same practicing veterinarian simply by selecting the next farmer on the list. Thus, the study population was a sample of dairy farmers, who were likely to have an interest in biosecurity and probably would have clear and interesting viewpoints on the subject, and, because of that quality, could define a factor. The selected farmers were invited to participate in the study by a covering letter and a subsequent phone call by the last author within the following week. Twenty five farmers accepted the invitation (two farmers declined; one farmer had just sold the farm and the other farmer had a large family event coming up at the time data was collected in this region). Farmers did not receive any compensation for their participation. 2.2. Data collection In Q-methodology a ‘concourse’ refers to ‘the flow of communicability surrounding any topic’ (Brown, 1993). A concourse is the technical definition for a contextual structure of all the possible statements that respondents might make about their personal views to answer a single research question. In this study, we constructed the
124
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
-4
-3
-2
-1
0
1
2
3
4 Agree
Disagree mostly
mostly
16
8
10
4
3
1
2
6
5
17
19
14
9
7
13
12
23
27
25
11
15
18
22
20
24
21
26 Fig. 1. The 25 participating farmers ranked the 27 statements (Table 1) from the concourse on a lay-out guide to address the question: from your point of view – which of these issues are most effective in preventing introduction of disease into a dairy herd? The ranking shown on this particular lay-out guide originates from farmer #16.
concourse based on our reflections on viewpoints in the literature, experience, input from the Danish National Board of Health and previous interviews and discussions with veterinarians, researchers, dairy farmers, financial lenders in the agro-business, the dairy industry, consumer and animal welfare organizations etc. Essentially, we included those on the farm; those surrounding the farm and those with a possible interest in dairy farming. This approach was time consuming, but necessary to provide the concourse with enough breath and comprehensiveness to cover the subject under study. Subsequently, we broke the concourse down into as many, yet distinctly different, statements that potentially could provide an answer the research question: from your point of view – which of these issues are most effective in preventing introduction of disease into a dairy herd? Farmers were asked to rank the statements with minimum interference from our side. The statements were sorted on a layout guide along a quasi-normal distribution (mean 0, SD 2.25) ranging from ‘agree mostly’ (+4) to ‘disagree mostly (−4). Each statement was typed on a separate card and marked with a random number for identification (Fig. 1). Following the sorting procedure farmers were asked to elaborate on the contextual structure and thematic saturation of the concourse and we asked all the participating farmers if they had: (1) developed the biosecurity plan, and (2) implemented a system to monitor and improve biosecurity at farm-level. The interviews took place at the individual farms and where administered by the last author, who provided the ‘conditions of instruction’ (Brown, 1993) before the sorting procedure. 2.3. Data analysis, quantitative Farmers’ rankings of statements were analysed by means of ‘PQMethod’ (http://www.lrz-muenchen.de/ ∼schmolck/qmethod/#Q-References) that is tailored to the requirements of Q-methodology. Specifically, the program allows easy entering of data the way it is obtained, i.e. as ‘piles’ of the random statement numbers. ‘PQMethod’ computes correlations among respondents (the variables or columns in the data matrix) that were characterized by the sorting procedure, i.e. each of the 27 statements was represented by one row in the matrix. This is equivalent to reversing the correlation matrix
used in traditional ‘R-factor analysis’, which is based on correlations between variables characterising respondents. Respondents, who were highly correlated with respect to their ranking of statements, are considered to have a ‘familiar’ resemblance, i.e. those statements belonging to one family being less correlated with statements defining other families. A principal component analysis was chosen in ‘PQMethod’ to estimate the total explained variance and the variance attributable to each identified factor (family of perspective). Factors with eigenvalues smaller than 1.00 were disregarded. A factor loading was determined for each respondent as an expression of which respondents were associated with each factor and to what degree. Loadings are correlation coefficients between respondents and factors. The remaining factors were subjected to a varimax (orthogonal) rotation to obtain the necessary rotated factor loadings. Scores and difference scores of each factor were estimated to present the normalized weighted average statement score of respondents defining that factor. The weight (w) was based on the respondent’s factor loading (f) and was calculated as: w = f/(1 − f2 ). The weighted average statement score was then normalized (with a mean of 0.00 and SD = 1.00) to remove the effect of differences in number of defining respondents per factor thereby making the statement’s factor scores comparable across factors. Thus, we took into account that some respondents were closer associated with the factor than others by constructing an idealized Q-sorting for each selected factor. The idealized Q-sorting of a factor may consequently be viewed as how a hypothetical respondent with a 100% loading on that factor would have ranked all the statements (Table 1). The difference score was the magnitude of difference between a statement’s score on any two factors that was required to be statistically significant. ‘PQMethod’ offers the possibility to identify the most distinguishing statements for each family of perspective and consensus statements between families of perspectives (not shown), i.e. those statements that do not distinguish between any pair of families. The limit for statistical significance of a factor loading was calculated as according to van Exel and de Graaf (2005): factor loading/(1 divided by the square root of the number of statements). If this ratio exceeded 1.96 we regarded the loading as statistically significant (P < 0.05).
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
125
Table 1 List of Q-statements and participating farmers’ ranking of biosecurity measures according to their perception of ‘effectiveness’ in preventing introduction of disease into a dairy herd. Statements derived from the concoursea 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 % Variance attributable to each family of farmers’ perspectives (unrotated factors (rotated factors))
Not possible to separate different age groups Staff not adequately educated Not possible to wash and disinfect hands Responsibility for product quality is unclear Reduce international sourcing of animals Lack of facilities to place animals for rendering Geography (e.g. wind direction) Lack of legislation on quarantine when buying animals Lack of an industry codex to regulate biosecurity Pests (e.g. rodents and birds) Visitors’ movements within the herd area Not self-perpetuating with heifers Unclear boundaries surrounding the herd area No use of claw bath Unclear distinction between clean and unclean roads to and from the herd area Sharing or renting machinery Legislation on biosecurity is too weak Grazing areas close to other herds Housing facilities (beddings, floor) Lack of spare clothes and boots for visitors Established dealers’ movement within the herd area Financial lenders are not willing to support investments in biosecurity No specific demands to the health status of purchased animals Lack of knowledge of biosecurity among farmers Unclear work procedures (e.g. manure handling) No clear logistics in feeding management No purchase of animals from established dealers 34/21
Family 1 cooperatives
Family 2 confused
Family 3 defectors
Family 4 selfish
−1 2 1 −1 4 3 −3 1
4 −1 3 2 0 1 −1 −1
0 −1 1 0 0 2 −2 −3
3 0 −1 0 0 1 −2 −4
−3
0
−3
−1
−2 0 2 0 −2 0
−4 −1 3 −2 4 −4
−1 0 3 1 4 −2
1 −3 1 0 3 1
−4 −2 0 −1 3 2
−2 2 −2 1 0 0
−1 −4 2 4 1 −2
−2 0 −4 −1 −3 2
−4
1
2
2
1
0
1
−2
1
−3
−1
−1
0
1
−4
2
−1 4
−3 2
0 3
4 4
12/8
8/11
7/12
a A concourse is a ‘view of the world’ constructed by the researcher from various sources of data. In Q-methodology the concourse is broken down by the researcher into a number of statements that respondents rank according to ‘my point of view’, i.e. how well the individual statement answers the research question: from your point of view – which of these issues are most effective in preventing introduction of disease into a dairy herd?
2.4. Data analysis, qualitative Following the quantitative analysis we did three different qualitative analyses to understand and describe the logic within and differences between the identified families of perspectives. First, we classified all statements into four thematic groups related to the origin of biological risk and the farmer’s possibilities to rapidly improve biosecurity as described by the statements (Table 2). This procedure did not by itself provide us with a clear picture of differences between farmers’ perspectives on biosecurity. Next, we sorted the statements according to their in-built level of abstraction. Statements were classified into three categories: (1) feasible and describing measures/activities that are easily connected to farm-level biosecurity; (2) either feasible or describing measures/activities that are easily connected to farm-level biosecurity, and (3) neither feasible nor describing measures/activities that are easily connected to farm-level biosecurity (Table 3). Last, we grouped the statements related to rules and legislation
(lack of legislation on quarantine when buying animals (8); lack of an industry codex to regulate biosecurity (9); legislation on biosecurity is too weak (17)) and introduced four ‘somebody else is responsible’ indicator statements (pests (10); legislation on biosecurity is too weak (17); established dealers’ movement within the herd area (21); financial lenders are not willing to support investments in biosecurity (22)) to study possible patterns of defection from legislation and expressions of introvert or selfish attitudes.
3. Results None of the participating farmers had developed the mandatory biosecurity plan or implemented any procedures comparable to a systematic biosecurity program. Further, none of the farmers had received any kind of punishment or had been contacted by the responsible veterinary authorities.
126
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
Table 2 Classification of statements (biosecurity measures) according to the origin of the biological risk and the farmer’s possibilities to rapidly improve biosecurity. Thematic group
Statements included in thematic group
External biosecurity
Not possible to wash and disinfect hands Reduce international sourcing of animals Lack of facilities to place animals for rendering Geography (e.g. wind direction)b Pests (e.g. rodents and birds) Visitors’ movements within the herd area Unclear boundaries surrounding the herd area Sharing or renting machinery Grazing areas close to other herds Lack of spare clothes and boots for visitors Established dealers’ movement within the herd area No specific demands to the health status of purchased animals No purchase of animals from established dealers Staff not adequately educated Responsibility for product quality is unclear Not self-perpetuating with heifersc No use of claw bath Unclear distinction between clean and unclean roads to and from the herd area Financial lenders are not willing to support investments in biosecurity Lack of knowledge of biosecurity among farmers Unclear work procedures (e.g. manure handling)d Lack of legislation on quarantine when buying animals Lack of an industry codex to regulate biosecurity Legislation on biosecurity is too weak Not possible to separate different age groups Housing facilities (e.g. beddings and floor) No clear logistics in feeding management
Internal herd managementa
Rules and legislation
The stablea
a In this study we defined ‘internal herd management’ and ‘the stable’ as different thematic groups. Reasons for this decision were time and costs. Likely, the farmer has few possibilities to rapidly change his daily working procedures to improve biosecurity measures within the existing stable. Other measures, e.g. educating staff or increasing focus on reproduction efficiency, may be implemented faster and within the existing buildings. b Geography (e.g. wind direction) was classified as external biosecurity because geography is beyond the farmer’s control. However, it has been argued that geography is in fact within the farmer’s control, as he is free to move his dairy enterprise to a safer location. c Poor reproduction results and subsequent shortage of heifers in the herd to maintain herd size was classified as an internal biosecurity breach because the solution to the problem, improving reproduction management, is a classical management decision within the farmer’s control. d Work procedures may follow a written protocol or a clear procedure communicated to all personnel working at the dairy farm. Equally, work procedures may be unclear, e.g. driving machinery with dirty tires across the feeding alley.
Table 3 Statements (biosecurity measures) classified according to the in-built level of abstraction within each statement. Level of abstraction in statement
Statements in each group of abstraction
Feasible and easy to understand
Not possible to separate different age groups Not possible to wash and disinfect hands Reduce international sourcing of animals Lack of facilities to place animals for rendering No use of claw bath Lack of spare clothes and boots for visitors Established dealers’ movement within the herd area Lack of knowledge of biosecurity among farmers No purchase of animals from established dealers Staff not adequately educated Geography (e.g. wind direction) Pests (e.g. rodents and birds) Not self-perpetuating with heifers Sharing or renting machinery Legislation on biosecurity is too weak Grazing areas close to other herds Housing facilities (e.g. beddings and floor) Financial lenders are not willing to support investments in biosecurity Unclear work procedures (e.g. manure handling) No clear logistics in feeding management Responsibility for product quality is unclear Lack of legislation on quarantine when buying animals Lack of an industry codex to regulate biosecurity Visitors’ movements within the herd area Unclear boundaries surrounding the herd area Unclear distinction between clean and unclean roads to and from the herd area No specific demands to the health status of purchased animals
Either feasible or easy to understand
Neither feasible nor easy to understand
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
The concourse consisted of a number of issues which were separated into 27 statements. A large proportion (61%) of the variation in statements could be explained by 4 factors (families of perspectives) with eigenvalues >1. Family 1 identified risk of introducing disease to the herd as a threat from the outside world (compare Tables 1 and 2). Family 2 and Family 3 were rather unsystematic in their perception of biological risk to the herd. Family 3 ranked statements related to rules and legislation (lack of legislation on quarantine when buying animals (8); lack of an industry codex to regulate biosecurity (9), and; legislation on biosecurity is too weak (17)) much lower (in total −10 point) than Family 2 (in total +1 point). Family 4 perceived internal herd management procedures as the best way to improve biosecurity. It was a general trend for all families of perspectives that statements that were easy to relate to farm-level biosecurity received a higher ranking; however, Family 1 seemed more willing to place a positive value on statements related to external biosecurity even if those statements were classified as ‘neither feasible nor describing measures/activities that are easily connected to farm-level biosecurity. In contrast, Families 2–4, especially Family 2, tended to place a negative value on statements classified to be on a higher level of abstraction. Family 4 ranking the four ‘somebody else is responsible’ indicator statements higher than Families 1–3 (in total −6, −1, −5 vs. +5 point, respectively). Interestingly, all four families of perspectives ranked the purchase of animals from established dealers’ (statement 27) among the statements associated with a high risk of disease introduction (Table 1). We tested this common perception identified among the participating farmers and from the Danish Cattle Database we realized that nine of the participating farmers had purchased animals from more than three established dealers in 2009. Consensus statements (non-significant at P > 0.05) were: lack of facilities to place animals for rendering (6); lack of an industry codex to regulate biosecurity (9); not self-perpetuating with heifers (12). These statements were considered revelatory by virtue of their salience as most farmers agreed upon the importance of statements 6 and 12 and the minor importance of statement 9. The idealized Q-sortings were assigned with informative names (labels) with input from the most distinguishing statements for family of perspective; the consensus statements and the qualitative analyses. The process of giving names to the idealized Q-sortings according to its characteristics serves to facilitate the discussion and communication of the findings (Kiernan and Heinrichs, 1994). To summarize: Family 1: Family 2: Family 3: Family 4:
Cooperatives; oriented towards external biosecurity Confused; unsystematic approach, trouble with more abstract statements Defectors; unsystematic approach, disregarded rules and legislation Introvert; focused on internal herd management, highest score on ‘somebody else is responsible’ indicator statements
127
4. Discussion 4.1. Participating farmers’ perception of biosecurity Farmers’ perception of biosecurity could meaningfully be classified into four families of perspectives. Family 1 (cooperatives) were knowledgeable about preventive measures and perceived the outside world to present the highest biological risk to the herd. None the less, a year after the implementation of biosecurity legislation none of the farmers participating in the study had developed the mandatory biosecurity plan. This leads to the notion that some of the farmers associated with Family 1 (to some extent) could represent a response bias, e.g. social desirability, which may shatter the picture of a cooperative approach. Social desirability represent respondents’ tendency to distort self-reports in a favorable direction, i.e. to deny socially undesirable traits and to claim socially desirable ones and the tendency to say things which place the respondent in a favorable light (Furnham, 1986). Social desirability could potentially explain why these farmers, despite the apparent ‘correct’ perception of biological risk to the herd, defect legislation even though they, by their ranking of statements, expressed a point of view that appeared to be cooperative with the intentions in the current biosecurity legislation. If this was the case, then some of farmers in Family 1 (cooperatives) may actually perceive biosecurity more like Family 3 (defectors; disregarding rules and legislation related to biosecurity). The theory of cognitive dissonance consists of the notion that humans try to establish internal harmony and consistency among opinions, attitudes, knowledge, and values (Festinger, 1957, p. 260). The substantial difference between the perspective on biosecurity expressed by Family 1 (cooperatives) and the fact that none of the farmers had developed a biosecurity plan point to the suggestion that farmers associated with Family 1 could be in a state of cognitive dissonance. To solve such dissonance and motivate the farmer to change his perceptions and attitudes in the ‘right’ direction is, however, no easy task (Heffernan et al., 2008; Jansen et al., 2010). Farmers associated with Family 2 (confused) expressed a point of view on biosecurity that appeared rather confused, according to the groupings used in this study. No obvious pattern was identified by their ranking of statements; however, we realized that biosecurity measures on a higher level of the abstraction-ladder tended to ranked low compared to the other families. Thus, we contend with the observation that Family 2 may have specific needs for tailored and ‘not too-complex’ communication. Family 3 (defectors), which may also include some of Family 1, disregarded statements related to rules and legislation. Improving biosecurity at farm-level may be constrained if such improvements are supported by official rules (Gunn et al., 2008). We interpreted Family 3 as a defector strategy prompted by the farmers’ perception of improvements of biosecurity measures as a business decision (see section on decision frames below) rather than a decision related to the common good. Farmers associated with Family 3 probably experienced little cognitive dissonance from not developing a biosecurity plan because
128
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
their ranking of statements appeared to be in harmony with their (defector) strategy. However, if these farmers allow their negative impression of rules and legislation to have influence on their intent to implement the mandatory biosecurity measures some of these farmers may experience some degree of cognitive dissonance simply because they know that defecting rules and legislation is likely to problematic to any citizen. Family 4 (introvert) was characterized by focusing on internal herd management procedures and the perception that ‘somebody else is responsible’ for the herds external biosecurity. None of the farmers in the study had developed the mandatory biosecurity plan. We speculate if this legislative non-compliance to could be explained by (1) the farmers’ lack trust in other farmers’ motivation or ability to maintain adequate biosecurity, as discussed by Gunn et al. (2008); (2) the farmers were uncertain if their colleagues would contribute to the common good at all (Gächter and Fehr, 1999; Rat-Aspert and Fourichon, 2010); (3) the farmers perceived the risk of disease introduction as too low to demand their attention, or (4) the farmers expected no social punishment from the social dilemma of (non-)compliance, as discussed by Dawes and Messick (2000). To explore the farmers’ reasoning to these decisions, however, is beyond the scope of this study. To summarize: ‘the assumption of complete connectedness between groups of farmers is inappropriate’ (Heffernan et al., 2008) indicating that each group of farmers (or perhaps every farmer) must be addressed individually, when it comes to biosecurity. Social desirability is a response bias that must be taken into account when evaluating biosecurity at farm-level. Participating farmers agreed that the highest risk of introducing disease into a dairy herd is the purchase of animals from established dealers. 4.2. Validity of results This study aims to obtain insight into a phenomenon as experienced by a range of individuals selected for this study because of their ‘information richness’ (Patton, 2002). Therefore, the results from this study are only directly applicable to the particular participants, settings and contexts, as discussed by Onwuegbuzie and Leech (2007). This study could not explain all the variation between statements which indicates that more may be learned about farmers perception of ‘effectiveness’ in preventing introduction of disease into a dairy herd. However, the active participation of end-users, i.e. farmers and other stakeholders, in the modeling–validating process (when constructing the concourse) is emphasized as an important part of the usefulness dimension of validity in operations research (Landry et al., 1983). Morse (1995) defined the concept of ‘saturation’, i.e. informational redundancy, in qualitative data as ‘data adequacy’ and explained it by ‘collecting data until no new information is obtained’. Consequently, the face-validation of the concourse by farmers may be seen as an acceptance of a ‘saturated concourse’. Q-methodology is about respondents ranking matters of opinion within a concourse to identify the existence of
families of perspectives. The result of a Q-factor analysis is useful to identify and describe a population of viewpoint and not, as in R, a population of people (Risdon et al., 2003). The difference between Q and R being that the issue of large numbers becomes rather unimportant in Q. Q-studies cannot obtain ‘true replication’ because: (1) an identical set of participants, contexts and experiences is impossible to find and; (2) the concourse, as it expresses itself in a Q-study, becomes context-bound to the particular participants, settings and contexts. The result of a Q-study is the distinct families of perspectives on a topic (as described by the concourse) that are operant, not the percentage of the sample (or the general population) that adheres to any of them. Consequently, the required number of respondents to establish the existence of a factor is substantially reduced for the purpose of comparing one factor with another when compared to traditional R statistics. 4.3. Future development The signaling effect that none of the farmers in this study were punished for defecting legislation on biosecurity may have increased to other farmers’ unwillingness to improve or invest in biosecurity measures. We believe that cooperation, compliance and biosecurity can be improved dramatically if it is possible to change the payoff-structure in a process considered by farmers to be both open and trustworthy, i.e. information about collective (in)efficiency plays an important role in the farmer’s decision to support a sanctioning system, as discussed by Mulder et al. (2005). The challenge is to present the decision of compliance or defection to the farmer as a decision situated within a business decision frame, where defectors and free-riders will be punished immediately and effectively. However, we would also like to suggest a system that rewards those farmers, who decide to improve biosecurity measures at farm-level. Consequently, we suggest the construction of an index system for biosecurity in dairy farms, i.e. a system that facilitates market-mediated biosecurity. Participating farmers should share both responsibility and costs associated with disease outbreak management. Participation should be voluntary and farmers deciding not to participate should be aware that they will get no sympathy or financial support in case of a disease outbreak at their farm. Probably, this would encourage such farmers’ financial lenders to engage in the discussion and subsequent decision. Obviously, financial institutions are aware that their risk will increase substantially if the farmer decides to be outside the collective system, especially since compensation from society will be reduced in the near future (EU, 2007). Within the collective system farmers should be allowed to decrease, sustain or improve biosecurity at farm-level to suit their production and ambitions; however, each index point should reflect an investment or effort to improve biosecurity and be directly linked to the individual farmers’ payment for being part of the suggested collective (insurance) system, i.e. some farmers must pay more than others to be a member. With this approach farmers can still be creative and innovative about their production facilities and earn the respect of other farmers (Burton et al., 2008).
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
The challenge to researchers will be to capture the fundamental relationships between farmers’ coping strategies and ambivalence towards biosecurity, incentive structures, risk (and risk perceptions) of animal diseases, the pricing of these risks and the sharing of the direct and consequential costs associated with a disease outbreak. 5. Conclusion In 2008 it was legislated that Danish farmers owning very large herds must develop a farm-specific biosecurity plan. A year later, none of the farmers participating in this study had complied. Participating farmers perceived the purchase of animals from established dealers as the highest risk for introducing disease to the herd. However, several of the participating farmers had recently purchased animals from such dealers. Several barriers to the improvement of biosecurity at farm-level are discussed, e.g. farmers’ view on biosecurity responsibility, cognitive dissonance and social dilemmas. Herd health advisors and policy-makers are advised to tailor communication to support farmers’ interest and compliance with biosecurity measures and legislation. We propose a framework to a biosecurity index system in a collective context. This system allows both creativity and innovation at farm-level and protects participating farmers from the exploitation of free-riders. Such an index system would contribute substantially to the development of sustainable cultural biosecurity changes at farm- and national level. References Brouwer, M., 1999. Q is accounting for tastes. J. Advert. Res. 39, 35–39. Brown, S.R., 1993. A primer on Q methodology. Operant Subjectivity 16, 91–138. Burton, R.J.F., 2004. Seeing through the ‘good farmer’s eyes: towards developing an understanding of the social symbolic value of ‘productivist behaviour’. Sociol. Ruralis 44, 195–215. Burton, R.J.F., Kuczera, C., Schwarz, G., 2008. Exploring farmers’ cultural resistance to voluntary agri-environmental schemes. Sociol. Ruralis 48, 16–37. Dawes, R.M., Messick, D.M., 2000. Social dilemmas. Int. J. Psychol. 35, 111–116. EU, 2007. From Farm to Fork [http://ec.europa.eu/food/animal/diseases/ strategy/index en.htm]. Festinger, L., 1957. A Theory of Cognitive Dissonance. Stanford University Press, California, pp. 291. Furnham, A., 1986. Response bias, social desirability and dissimulation. Person. Individ. Differ. 7, 385–400. Gächter, S., Fehr, E., 1999. Collective action as a social exchange. J. Econ. Behav. Organ. 39, 341–369.
129
Garforth, C.F., 2010. Motivating farmers: insights from social psychology. In: Proc. Annual Meeting, NMC , Albuquerque, US, pp. 60–67, 49. Gunn, G.J., Heffernan, C., Hall, M., McLeod, A., Hovi, M., 2008. Measuring and comparing constraints to improve biosecurity amongst GB farmers, veterinarians and the auxiliary industries. Prev. Vet. Med. 84, 310–323. Heffernan, C., Nielsen, L., Thomson, K., Gunn, G., 2008. An exploration of the drivers to bio-security collective action among a sample of UK cattle and sheep farmers. Prev. Vet. Med. 87, 358–372. Jansen, J., van den Borne, B.H.P., Renes, R.J., van Schaik, G., Lam, T.J.G.M., Leeuwis, C., 2009. Explaining mastitis incidence in Dutch dairy farming: the influence of farmers’ attitudes and behaviour. Prev. Vet. Med. 92, 210–223. Jansen, J., Steuten, C.D.M., Renes, R.J., Aarts, N., Lam, T.J.G.M., 2010. Debunking the myth of the hard-to-reach farmer: effective communication on udder health. J. Dairy Sci. 93, 1296–1306. Janz, N., Becker, M.H., 1984. The health belief model: a decade later. Health Educ. Q 11, 1–7. Kiernan, N.E., Heinrichs, A.J., 1994. Identification of farm manager types through cluster analysis of calf and heifer management practices. Prev. Vet. Med. 18, 225–236. Kristensen, E., Enevoldsen, C., 2008. A mixed methods inquiry: how dairy farmers perceive the value(s) of their involvement in an intensive dairy herd health management program. Acta Vet. Scand. 50, 50. Kristensen, E., Jakobsen, E.B., 2010. Evaluation of dairy herd health management. Updates on Ruminant Production and Medicine, World Buiatrics Congress XXVI, Santiago, Chile, pp. 53–64. Landry, M., Malouin, J.-M., Oral, M., 1983. Model validation in operations research. Eur. J. Oper. Res. 14, 207–220. LeBlanc, S.J., Lissemore, K.D., Kelton, D.F., Duffield, T.F., Leslie, K.E., 2006. Major advances in disease prevention in dairy cattle. J. Dairy Sci. 89, 1267–1279. Morse, J.M., 1995. The significance of saturation. Qual. Health Res. 5, 147–149. Mulder, L.B., van Dijk, E., Wilke, H.A.M., De Cremer, D., 2005. The effect of feedback on support for a sanctioning system in a social dilemma: the difference between installing and maintaining the sanction. J. Econ. Psychol. 26, 443–458. Onwuegbuzie, A.J., Leech, N.L., 2007. A call for qualitative power analyses. Qual. Quan. 41, 105–121. Patton, M.Q., 2002. Qualitative research and evaluation methods, 3rd ed. Sage Publications, 230. Rat-Aspert, O., Fourichon, C., 2010. Modelling collective effectiveness of voluntary vaccination with and without incentives. Prev. Vet. Med. 93, 265–275. Risdon, A., Eccleston, C., McCracken, G.C., 2003. How can we learn to live with pain? A Q-methodological analysis of the diverse understandings of acceptance of chronic pain. Soc. Sci. Med. 56, 375–386. Stephenson, W., 1935. Technique of factor analysis. Nature 136, 297. Stott, A.W., Gunn, G.J., 2008. Use of a benefit function to assess the relative investment potential of alternative farm animal disease prevention strategies. Prev. Vet. Med. 84, 179–193. Valeeva, N.I., Lam, T.J.G.M., Hogeveen, H., 2007. Motivation of dairy farmers to improve mastitis management. J. Dairy Sci. 90, 4466–4477. van Exel, N.J.A., de Graaf, G., 2005. Q methodology: a sneak preveiw. [http://www.qmethodology.net/index.php?page=1&year=2005]. Weber, J.M., Kopelman, S., Messick, D.M., 2004. A conceptual review of decision making in social dilemmas: applying a logic of appropriateness. Pers. Soc. Psychol. Rev. 8, 281–307.
Contents lists available at ScienceDirect
Preventive Veterinary Medicine journal homepage: www.elsevier.com/locate/prevetmed
Danish dairy farmers’ perception of biosecurity Erling Kristensen a,b,∗ , Esben B. Jakobsen b a b
Knowledge Centre for Agriculture, Cattle, Agro Food Park 15, DK-8200 Aarhus N, Denmark StrateKo Aps, Risagervej 5, DK-6690 Gording, Denmark
a r t i c l e
i n f o
Article history: Received 3 November 2009 Received in revised form 26 January 2011 Accepted 26 January 2011 Keywords: Biosecurity Social dilemma Q-methodology
a b s t r a c t To implement biosecurity measures at farm-level is a motivational challenge to dairy farmers as emerging diseases and their consequences largely are unpredictable. One of the reasons for this challenge is that outcomes are more likely to benefit society than the individual farmer. From the individual farmer’s point of view the impacts of zoonotic risk, international trade and welfare concerns appear less obvious than the direct costs at farm-level. Consequently, a social dilemma may arise where collective interests are at odds with private interests. To improve biosecurity at farm-level farmers must be motivated to change behavior in the ‘right’ direction which could provide selfish farmers with unintended possibilities to exploit the level of biosecurity provided by other dairy farmers’ collective actions. Farmers’ perception of risk of disease introduction into a dairy herd was explored by means of Q-methodology. Participating farmers owned very large dairy herds and were selected for this study because Danish legislation since 2008 has required that larger farms develop and implement a farm specific biosecurity plan. However, a year from introduction of this requirement, none of the participating farmers had developed a biosecurity plan. Farmers’ perception of biosecurity could meaningfully be described by four families of perspectives, labeled: cooperatives; confused; defectors, and introvert. Interestingly, all families of perspectives agreed that sourcing of animals from established dealers represented the highest risk to biosecurity at farm-level. Farmers and policy-makers are faced with important questions about biosecurity at farmlevel related to the sanctioning system within the contextual framework of social dilemmas. To solve these challenges we propose the development of a market-mediated system to (1) reduce the risk of free-riders, and (2) provide farmers with incentives to improve biosecurity at farm-level. © 2011 Elsevier B.V. All rights reserved.
1. Introduction In 2008 it became Danish law that biosecurity plans were mandatory in large dairy herds (>330 cows per year). Supposedly, the biosecurity plans should force farmers to
∗ Corresponding author at: Knowledge Centre for Agriculture, Cattle Health, Welfare and Food Safety, Agro Food Park 15, DK-8200 Aarhus N, Denmark. Tel.: +45 4160 4715; fax: +45 8740 5010. E-mail addresses: ekr@vfl.dk (E. Kristensen), [email protected] (E.B. Jakobsen). 0167-5877/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.prevetmed.2011.01.010
improve on precautionary measures, disease surveillance and controls in order to reduce the risk of introducing animal diseases into the dairy herd and minimize the impact of outbreaks, should they occur. This was in line with the intentions described in the EU initiative, Animal Health Strategy (EU, 2007). However, to motivate farmers into changing their daily routines is a well-known challenge (Burton et al., 2008; Garforth, 2010) and biosecurity measures seem particularly difficult to implement by law as outcomes are more likely to benefit society than the individual dairy farmer, i.e. a social dilemma. Social dilemmas are situations characterized by two factors: (1) at any given
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
decision point, individuals receive higher payoffs for making selfish choices than they do for making cooperative choices regardless of the choices made by those with whom they interact, and (2) everyone involved receives lower payoffs if everyone makes selfish choices than if everyone makes cooperative choices (Mulder et al., 2005). The situation in Denmark brought two social dilemmas into the open: (1) farmers were forced to prioritize between shortterm selfish interests or the long-term interests of dairy farming, i.e. collective interests are at odds with private interests, and (2) farmers were forced to make a decision to defect or accept legislation, i.e. a surveillance and sanctioning dilemma. Interesting to this context is the discussion by Gächter and Fehr (1999) on how subjects’ beliefs about other subjects’ contributions to a social dilemma situation will ultimately determine how much they contribute. Weber et al. (2004) related this dilemma to group mechanisms, involving emotions as fear or greed. Further, being part of a group may define and frame a social dilemma to the group as a situation in which selfish behavior is unlikely to be sanctioned or punished socially (Dawes and Messick, 2000). Farmers’ decision making is not confined to herd health (Stott and Gunn, 2008). Consequently, the underlying assumption behind rational choice models, i.e. the vigilant and calculating decision maker focused on maximizing his expected payoff is unlikely to give a valid picture of farmers’ attitudes and behavior (Valeeva et al., 2007). In fact, rational choice models often downplay social influence processes and overall utility which limit the explanatory power of such models when applied to most social dilemmas (Weber et al., 2004; Rat-Aspert and Fourichon, 2010). It follows, that rational choice models have little predictive effect on farmers’ attitudes and subsequent behavior at farm-level (Burton, 2004; Jansen et al., 2009). Rather, as discussed by Janz and Becker (1984), peoples (preventive) measures are determined by (1) the perceived threat (perceived vulnerability and perceived severity) and (2) the perceived effectiveness of proposed measures (perceived benefits and barriers). Identical findings have been reported from the field of animal health by Heffernan et al. (2008) where farmers’ mutual mistrust (and general mistrust in the authorities) prevented improvements of collective biosecurity measures. Lately, herd health management has been characterized by an integrated, holistic, proactive, data-based and economically framed approach to preventive medicine (LeBlanc et al., 2006) combining disciplines like sociology, psychology, behavior science and communication with veterinary epidemiology (Kristensen and Enevoldsen, 2008) into what we suggest should be labeled ‘social epidemiology’ (Kristensen and Jakobsen, 2010). This approach offers new methodological possibilities to animal and herd health researchers as the social sciences have a longer tradition to account for individual differences, which may be viewed as central to (1) understand the cognitive dissonance between farmers’ biosecurity attitudes and their behavior, and (2) to tailor communication to groups of farmers sharing opinions on biosecurity (Heffernan et al., 2008; Jansen et al., 2010).
123
The objectives of this study were: (1) to explore dairy farmers’ perception of risk of introducing disease into a dairy farm; (2) to contribute to the discussion about social dilemmas in a surveillance and sanctioning system that presently is being imposed on Danish dairy farmers, and (3) to suggest a solution to social dilemmas associated with biosecurity that improves biosecurity at farm- and national level. 2. Materials and methods In this study we explored dairy farmers’ perception of risk of introducing disease into dairy farm by asking the participating farmers to rank a number of statements on a layout guide according to their individual perception of the term ‘biosecurity’. The core research tool of this study was Q-methodology, which was first described by Stephenson (1935) and provides a foundation for the systematic study of subjectivity that is ‘a person’s viewpoints, opinion, beliefs, attitude, and the like’ (Brown, 1993). It follows that Q-methodology does not aim at estimating proportions of different views held by the ‘farmer population’ (Brouwer, 1999). Rather, Q identifies qualitative categories of thought shared by groups of respondents, i.e. dairy farmers. The applied research methodology followed the approach described in Kristensen and Enevoldsen (2008). 2.1. Study population In a stratified design (to make sure that all regions in Denmark was represented) including reflections on sample size, as discussed by Onwuegbuzie and Leech (2007), we selected 27 farmers from a list of 168 farmers with herds larger than 330 cows per year. Within each region, we selected the largest farms and a number of farms that largely reflected cow density in Denmark. To provide the study population with the largest possible comprehensiveness we tried to avoid including more than one farm associated with the same practicing veterinarian simply by selecting the next farmer on the list. Thus, the study population was a sample of dairy farmers, who were likely to have an interest in biosecurity and probably would have clear and interesting viewpoints on the subject, and, because of that quality, could define a factor. The selected farmers were invited to participate in the study by a covering letter and a subsequent phone call by the last author within the following week. Twenty five farmers accepted the invitation (two farmers declined; one farmer had just sold the farm and the other farmer had a large family event coming up at the time data was collected in this region). Farmers did not receive any compensation for their participation. 2.2. Data collection In Q-methodology a ‘concourse’ refers to ‘the flow of communicability surrounding any topic’ (Brown, 1993). A concourse is the technical definition for a contextual structure of all the possible statements that respondents might make about their personal views to answer a single research question. In this study, we constructed the
124
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
-4
-3
-2
-1
0
1
2
3
4 Agree
Disagree mostly
mostly
16
8
10
4
3
1
2
6
5
17
19
14
9
7
13
12
23
27
25
11
15
18
22
20
24
21
26 Fig. 1. The 25 participating farmers ranked the 27 statements (Table 1) from the concourse on a lay-out guide to address the question: from your point of view – which of these issues are most effective in preventing introduction of disease into a dairy herd? The ranking shown on this particular lay-out guide originates from farmer #16.
concourse based on our reflections on viewpoints in the literature, experience, input from the Danish National Board of Health and previous interviews and discussions with veterinarians, researchers, dairy farmers, financial lenders in the agro-business, the dairy industry, consumer and animal welfare organizations etc. Essentially, we included those on the farm; those surrounding the farm and those with a possible interest in dairy farming. This approach was time consuming, but necessary to provide the concourse with enough breath and comprehensiveness to cover the subject under study. Subsequently, we broke the concourse down into as many, yet distinctly different, statements that potentially could provide an answer the research question: from your point of view – which of these issues are most effective in preventing introduction of disease into a dairy herd? Farmers were asked to rank the statements with minimum interference from our side. The statements were sorted on a layout guide along a quasi-normal distribution (mean 0, SD 2.25) ranging from ‘agree mostly’ (+4) to ‘disagree mostly (−4). Each statement was typed on a separate card and marked with a random number for identification (Fig. 1). Following the sorting procedure farmers were asked to elaborate on the contextual structure and thematic saturation of the concourse and we asked all the participating farmers if they had: (1) developed the biosecurity plan, and (2) implemented a system to monitor and improve biosecurity at farm-level. The interviews took place at the individual farms and where administered by the last author, who provided the ‘conditions of instruction’ (Brown, 1993) before the sorting procedure. 2.3. Data analysis, quantitative Farmers’ rankings of statements were analysed by means of ‘PQMethod’ (http://www.lrz-muenchen.de/ ∼schmolck/qmethod/#Q-References) that is tailored to the requirements of Q-methodology. Specifically, the program allows easy entering of data the way it is obtained, i.e. as ‘piles’ of the random statement numbers. ‘PQMethod’ computes correlations among respondents (the variables or columns in the data matrix) that were characterized by the sorting procedure, i.e. each of the 27 statements was represented by one row in the matrix. This is equivalent to reversing the correlation matrix
used in traditional ‘R-factor analysis’, which is based on correlations between variables characterising respondents. Respondents, who were highly correlated with respect to their ranking of statements, are considered to have a ‘familiar’ resemblance, i.e. those statements belonging to one family being less correlated with statements defining other families. A principal component analysis was chosen in ‘PQMethod’ to estimate the total explained variance and the variance attributable to each identified factor (family of perspective). Factors with eigenvalues smaller than 1.00 were disregarded. A factor loading was determined for each respondent as an expression of which respondents were associated with each factor and to what degree. Loadings are correlation coefficients between respondents and factors. The remaining factors were subjected to a varimax (orthogonal) rotation to obtain the necessary rotated factor loadings. Scores and difference scores of each factor were estimated to present the normalized weighted average statement score of respondents defining that factor. The weight (w) was based on the respondent’s factor loading (f) and was calculated as: w = f/(1 − f2 ). The weighted average statement score was then normalized (with a mean of 0.00 and SD = 1.00) to remove the effect of differences in number of defining respondents per factor thereby making the statement’s factor scores comparable across factors. Thus, we took into account that some respondents were closer associated with the factor than others by constructing an idealized Q-sorting for each selected factor. The idealized Q-sorting of a factor may consequently be viewed as how a hypothetical respondent with a 100% loading on that factor would have ranked all the statements (Table 1). The difference score was the magnitude of difference between a statement’s score on any two factors that was required to be statistically significant. ‘PQMethod’ offers the possibility to identify the most distinguishing statements for each family of perspective and consensus statements between families of perspectives (not shown), i.e. those statements that do not distinguish between any pair of families. The limit for statistical significance of a factor loading was calculated as according to van Exel and de Graaf (2005): factor loading/(1 divided by the square root of the number of statements). If this ratio exceeded 1.96 we regarded the loading as statistically significant (P < 0.05).
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
125
Table 1 List of Q-statements and participating farmers’ ranking of biosecurity measures according to their perception of ‘effectiveness’ in preventing introduction of disease into a dairy herd. Statements derived from the concoursea 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 % Variance attributable to each family of farmers’ perspectives (unrotated factors (rotated factors))
Not possible to separate different age groups Staff not adequately educated Not possible to wash and disinfect hands Responsibility for product quality is unclear Reduce international sourcing of animals Lack of facilities to place animals for rendering Geography (e.g. wind direction) Lack of legislation on quarantine when buying animals Lack of an industry codex to regulate biosecurity Pests (e.g. rodents and birds) Visitors’ movements within the herd area Not self-perpetuating with heifers Unclear boundaries surrounding the herd area No use of claw bath Unclear distinction between clean and unclean roads to and from the herd area Sharing or renting machinery Legislation on biosecurity is too weak Grazing areas close to other herds Housing facilities (beddings, floor) Lack of spare clothes and boots for visitors Established dealers’ movement within the herd area Financial lenders are not willing to support investments in biosecurity No specific demands to the health status of purchased animals Lack of knowledge of biosecurity among farmers Unclear work procedures (e.g. manure handling) No clear logistics in feeding management No purchase of animals from established dealers 34/21
Family 1 cooperatives
Family 2 confused
Family 3 defectors
Family 4 selfish
−1 2 1 −1 4 3 −3 1
4 −1 3 2 0 1 −1 −1
0 −1 1 0 0 2 −2 −3
3 0 −1 0 0 1 −2 −4
−3
0
−3
−1
−2 0 2 0 −2 0
−4 −1 3 −2 4 −4
−1 0 3 1 4 −2
1 −3 1 0 3 1
−4 −2 0 −1 3 2
−2 2 −2 1 0 0
−1 −4 2 4 1 −2
−2 0 −4 −1 −3 2
−4
1
2
2
1
0
1
−2
1
−3
−1
−1
0
1
−4
2
−1 4
−3 2
0 3
4 4
12/8
8/11
7/12
a A concourse is a ‘view of the world’ constructed by the researcher from various sources of data. In Q-methodology the concourse is broken down by the researcher into a number of statements that respondents rank according to ‘my point of view’, i.e. how well the individual statement answers the research question: from your point of view – which of these issues are most effective in preventing introduction of disease into a dairy herd?
2.4. Data analysis, qualitative Following the quantitative analysis we did three different qualitative analyses to understand and describe the logic within and differences between the identified families of perspectives. First, we classified all statements into four thematic groups related to the origin of biological risk and the farmer’s possibilities to rapidly improve biosecurity as described by the statements (Table 2). This procedure did not by itself provide us with a clear picture of differences between farmers’ perspectives on biosecurity. Next, we sorted the statements according to their in-built level of abstraction. Statements were classified into three categories: (1) feasible and describing measures/activities that are easily connected to farm-level biosecurity; (2) either feasible or describing measures/activities that are easily connected to farm-level biosecurity, and (3) neither feasible nor describing measures/activities that are easily connected to farm-level biosecurity (Table 3). Last, we grouped the statements related to rules and legislation
(lack of legislation on quarantine when buying animals (8); lack of an industry codex to regulate biosecurity (9); legislation on biosecurity is too weak (17)) and introduced four ‘somebody else is responsible’ indicator statements (pests (10); legislation on biosecurity is too weak (17); established dealers’ movement within the herd area (21); financial lenders are not willing to support investments in biosecurity (22)) to study possible patterns of defection from legislation and expressions of introvert or selfish attitudes.
3. Results None of the participating farmers had developed the mandatory biosecurity plan or implemented any procedures comparable to a systematic biosecurity program. Further, none of the farmers had received any kind of punishment or had been contacted by the responsible veterinary authorities.
126
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
Table 2 Classification of statements (biosecurity measures) according to the origin of the biological risk and the farmer’s possibilities to rapidly improve biosecurity. Thematic group
Statements included in thematic group
External biosecurity
Not possible to wash and disinfect hands Reduce international sourcing of animals Lack of facilities to place animals for rendering Geography (e.g. wind direction)b Pests (e.g. rodents and birds) Visitors’ movements within the herd area Unclear boundaries surrounding the herd area Sharing or renting machinery Grazing areas close to other herds Lack of spare clothes and boots for visitors Established dealers’ movement within the herd area No specific demands to the health status of purchased animals No purchase of animals from established dealers Staff not adequately educated Responsibility for product quality is unclear Not self-perpetuating with heifersc No use of claw bath Unclear distinction between clean and unclean roads to and from the herd area Financial lenders are not willing to support investments in biosecurity Lack of knowledge of biosecurity among farmers Unclear work procedures (e.g. manure handling)d Lack of legislation on quarantine when buying animals Lack of an industry codex to regulate biosecurity Legislation on biosecurity is too weak Not possible to separate different age groups Housing facilities (e.g. beddings and floor) No clear logistics in feeding management
Internal herd managementa
Rules and legislation
The stablea
a In this study we defined ‘internal herd management’ and ‘the stable’ as different thematic groups. Reasons for this decision were time and costs. Likely, the farmer has few possibilities to rapidly change his daily working procedures to improve biosecurity measures within the existing stable. Other measures, e.g. educating staff or increasing focus on reproduction efficiency, may be implemented faster and within the existing buildings. b Geography (e.g. wind direction) was classified as external biosecurity because geography is beyond the farmer’s control. However, it has been argued that geography is in fact within the farmer’s control, as he is free to move his dairy enterprise to a safer location. c Poor reproduction results and subsequent shortage of heifers in the herd to maintain herd size was classified as an internal biosecurity breach because the solution to the problem, improving reproduction management, is a classical management decision within the farmer’s control. d Work procedures may follow a written protocol or a clear procedure communicated to all personnel working at the dairy farm. Equally, work procedures may be unclear, e.g. driving machinery with dirty tires across the feeding alley.
Table 3 Statements (biosecurity measures) classified according to the in-built level of abstraction within each statement. Level of abstraction in statement
Statements in each group of abstraction
Feasible and easy to understand
Not possible to separate different age groups Not possible to wash and disinfect hands Reduce international sourcing of animals Lack of facilities to place animals for rendering No use of claw bath Lack of spare clothes and boots for visitors Established dealers’ movement within the herd area Lack of knowledge of biosecurity among farmers No purchase of animals from established dealers Staff not adequately educated Geography (e.g. wind direction) Pests (e.g. rodents and birds) Not self-perpetuating with heifers Sharing or renting machinery Legislation on biosecurity is too weak Grazing areas close to other herds Housing facilities (e.g. beddings and floor) Financial lenders are not willing to support investments in biosecurity Unclear work procedures (e.g. manure handling) No clear logistics in feeding management Responsibility for product quality is unclear Lack of legislation on quarantine when buying animals Lack of an industry codex to regulate biosecurity Visitors’ movements within the herd area Unclear boundaries surrounding the herd area Unclear distinction between clean and unclean roads to and from the herd area No specific demands to the health status of purchased animals
Either feasible or easy to understand
Neither feasible nor easy to understand
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
The concourse consisted of a number of issues which were separated into 27 statements. A large proportion (61%) of the variation in statements could be explained by 4 factors (families of perspectives) with eigenvalues >1. Family 1 identified risk of introducing disease to the herd as a threat from the outside world (compare Tables 1 and 2). Family 2 and Family 3 were rather unsystematic in their perception of biological risk to the herd. Family 3 ranked statements related to rules and legislation (lack of legislation on quarantine when buying animals (8); lack of an industry codex to regulate biosecurity (9), and; legislation on biosecurity is too weak (17)) much lower (in total −10 point) than Family 2 (in total +1 point). Family 4 perceived internal herd management procedures as the best way to improve biosecurity. It was a general trend for all families of perspectives that statements that were easy to relate to farm-level biosecurity received a higher ranking; however, Family 1 seemed more willing to place a positive value on statements related to external biosecurity even if those statements were classified as ‘neither feasible nor describing measures/activities that are easily connected to farm-level biosecurity. In contrast, Families 2–4, especially Family 2, tended to place a negative value on statements classified to be on a higher level of abstraction. Family 4 ranking the four ‘somebody else is responsible’ indicator statements higher than Families 1–3 (in total −6, −1, −5 vs. +5 point, respectively). Interestingly, all four families of perspectives ranked the purchase of animals from established dealers’ (statement 27) among the statements associated with a high risk of disease introduction (Table 1). We tested this common perception identified among the participating farmers and from the Danish Cattle Database we realized that nine of the participating farmers had purchased animals from more than three established dealers in 2009. Consensus statements (non-significant at P > 0.05) were: lack of facilities to place animals for rendering (6); lack of an industry codex to regulate biosecurity (9); not self-perpetuating with heifers (12). These statements were considered revelatory by virtue of their salience as most farmers agreed upon the importance of statements 6 and 12 and the minor importance of statement 9. The idealized Q-sortings were assigned with informative names (labels) with input from the most distinguishing statements for family of perspective; the consensus statements and the qualitative analyses. The process of giving names to the idealized Q-sortings according to its characteristics serves to facilitate the discussion and communication of the findings (Kiernan and Heinrichs, 1994). To summarize: Family 1: Family 2: Family 3: Family 4:
Cooperatives; oriented towards external biosecurity Confused; unsystematic approach, trouble with more abstract statements Defectors; unsystematic approach, disregarded rules and legislation Introvert; focused on internal herd management, highest score on ‘somebody else is responsible’ indicator statements
127
4. Discussion 4.1. Participating farmers’ perception of biosecurity Farmers’ perception of biosecurity could meaningfully be classified into four families of perspectives. Family 1 (cooperatives) were knowledgeable about preventive measures and perceived the outside world to present the highest biological risk to the herd. None the less, a year after the implementation of biosecurity legislation none of the farmers participating in the study had developed the mandatory biosecurity plan. This leads to the notion that some of the farmers associated with Family 1 (to some extent) could represent a response bias, e.g. social desirability, which may shatter the picture of a cooperative approach. Social desirability represent respondents’ tendency to distort self-reports in a favorable direction, i.e. to deny socially undesirable traits and to claim socially desirable ones and the tendency to say things which place the respondent in a favorable light (Furnham, 1986). Social desirability could potentially explain why these farmers, despite the apparent ‘correct’ perception of biological risk to the herd, defect legislation even though they, by their ranking of statements, expressed a point of view that appeared to be cooperative with the intentions in the current biosecurity legislation. If this was the case, then some of farmers in Family 1 (cooperatives) may actually perceive biosecurity more like Family 3 (defectors; disregarding rules and legislation related to biosecurity). The theory of cognitive dissonance consists of the notion that humans try to establish internal harmony and consistency among opinions, attitudes, knowledge, and values (Festinger, 1957, p. 260). The substantial difference between the perspective on biosecurity expressed by Family 1 (cooperatives) and the fact that none of the farmers had developed a biosecurity plan point to the suggestion that farmers associated with Family 1 could be in a state of cognitive dissonance. To solve such dissonance and motivate the farmer to change his perceptions and attitudes in the ‘right’ direction is, however, no easy task (Heffernan et al., 2008; Jansen et al., 2010). Farmers associated with Family 2 (confused) expressed a point of view on biosecurity that appeared rather confused, according to the groupings used in this study. No obvious pattern was identified by their ranking of statements; however, we realized that biosecurity measures on a higher level of the abstraction-ladder tended to ranked low compared to the other families. Thus, we contend with the observation that Family 2 may have specific needs for tailored and ‘not too-complex’ communication. Family 3 (defectors), which may also include some of Family 1, disregarded statements related to rules and legislation. Improving biosecurity at farm-level may be constrained if such improvements are supported by official rules (Gunn et al., 2008). We interpreted Family 3 as a defector strategy prompted by the farmers’ perception of improvements of biosecurity measures as a business decision (see section on decision frames below) rather than a decision related to the common good. Farmers associated with Family 3 probably experienced little cognitive dissonance from not developing a biosecurity plan because
128
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
their ranking of statements appeared to be in harmony with their (defector) strategy. However, if these farmers allow their negative impression of rules and legislation to have influence on their intent to implement the mandatory biosecurity measures some of these farmers may experience some degree of cognitive dissonance simply because they know that defecting rules and legislation is likely to problematic to any citizen. Family 4 (introvert) was characterized by focusing on internal herd management procedures and the perception that ‘somebody else is responsible’ for the herds external biosecurity. None of the farmers in the study had developed the mandatory biosecurity plan. We speculate if this legislative non-compliance to could be explained by (1) the farmers’ lack trust in other farmers’ motivation or ability to maintain adequate biosecurity, as discussed by Gunn et al. (2008); (2) the farmers were uncertain if their colleagues would contribute to the common good at all (Gächter and Fehr, 1999; Rat-Aspert and Fourichon, 2010); (3) the farmers perceived the risk of disease introduction as too low to demand their attention, or (4) the farmers expected no social punishment from the social dilemma of (non-)compliance, as discussed by Dawes and Messick (2000). To explore the farmers’ reasoning to these decisions, however, is beyond the scope of this study. To summarize: ‘the assumption of complete connectedness between groups of farmers is inappropriate’ (Heffernan et al., 2008) indicating that each group of farmers (or perhaps every farmer) must be addressed individually, when it comes to biosecurity. Social desirability is a response bias that must be taken into account when evaluating biosecurity at farm-level. Participating farmers agreed that the highest risk of introducing disease into a dairy herd is the purchase of animals from established dealers. 4.2. Validity of results This study aims to obtain insight into a phenomenon as experienced by a range of individuals selected for this study because of their ‘information richness’ (Patton, 2002). Therefore, the results from this study are only directly applicable to the particular participants, settings and contexts, as discussed by Onwuegbuzie and Leech (2007). This study could not explain all the variation between statements which indicates that more may be learned about farmers perception of ‘effectiveness’ in preventing introduction of disease into a dairy herd. However, the active participation of end-users, i.e. farmers and other stakeholders, in the modeling–validating process (when constructing the concourse) is emphasized as an important part of the usefulness dimension of validity in operations research (Landry et al., 1983). Morse (1995) defined the concept of ‘saturation’, i.e. informational redundancy, in qualitative data as ‘data adequacy’ and explained it by ‘collecting data until no new information is obtained’. Consequently, the face-validation of the concourse by farmers may be seen as an acceptance of a ‘saturated concourse’. Q-methodology is about respondents ranking matters of opinion within a concourse to identify the existence of
families of perspectives. The result of a Q-factor analysis is useful to identify and describe a population of viewpoint and not, as in R, a population of people (Risdon et al., 2003). The difference between Q and R being that the issue of large numbers becomes rather unimportant in Q. Q-studies cannot obtain ‘true replication’ because: (1) an identical set of participants, contexts and experiences is impossible to find and; (2) the concourse, as it expresses itself in a Q-study, becomes context-bound to the particular participants, settings and contexts. The result of a Q-study is the distinct families of perspectives on a topic (as described by the concourse) that are operant, not the percentage of the sample (or the general population) that adheres to any of them. Consequently, the required number of respondents to establish the existence of a factor is substantially reduced for the purpose of comparing one factor with another when compared to traditional R statistics. 4.3. Future development The signaling effect that none of the farmers in this study were punished for defecting legislation on biosecurity may have increased to other farmers’ unwillingness to improve or invest in biosecurity measures. We believe that cooperation, compliance and biosecurity can be improved dramatically if it is possible to change the payoff-structure in a process considered by farmers to be both open and trustworthy, i.e. information about collective (in)efficiency plays an important role in the farmer’s decision to support a sanctioning system, as discussed by Mulder et al. (2005). The challenge is to present the decision of compliance or defection to the farmer as a decision situated within a business decision frame, where defectors and free-riders will be punished immediately and effectively. However, we would also like to suggest a system that rewards those farmers, who decide to improve biosecurity measures at farm-level. Consequently, we suggest the construction of an index system for biosecurity in dairy farms, i.e. a system that facilitates market-mediated biosecurity. Participating farmers should share both responsibility and costs associated with disease outbreak management. Participation should be voluntary and farmers deciding not to participate should be aware that they will get no sympathy or financial support in case of a disease outbreak at their farm. Probably, this would encourage such farmers’ financial lenders to engage in the discussion and subsequent decision. Obviously, financial institutions are aware that their risk will increase substantially if the farmer decides to be outside the collective system, especially since compensation from society will be reduced in the near future (EU, 2007). Within the collective system farmers should be allowed to decrease, sustain or improve biosecurity at farm-level to suit their production and ambitions; however, each index point should reflect an investment or effort to improve biosecurity and be directly linked to the individual farmers’ payment for being part of the suggested collective (insurance) system, i.e. some farmers must pay more than others to be a member. With this approach farmers can still be creative and innovative about their production facilities and earn the respect of other farmers (Burton et al., 2008).
E. Kristensen, E.B. Jakobsen / Preventive Veterinary Medicine 99 (2011) 122–129
The challenge to researchers will be to capture the fundamental relationships between farmers’ coping strategies and ambivalence towards biosecurity, incentive structures, risk (and risk perceptions) of animal diseases, the pricing of these risks and the sharing of the direct and consequential costs associated with a disease outbreak. 5. Conclusion In 2008 it was legislated that Danish farmers owning very large herds must develop a farm-specific biosecurity plan. A year later, none of the farmers participating in this study had complied. Participating farmers perceived the purchase of animals from established dealers as the highest risk for introducing disease to the herd. However, several of the participating farmers had recently purchased animals from such dealers. Several barriers to the improvement of biosecurity at farm-level are discussed, e.g. farmers’ view on biosecurity responsibility, cognitive dissonance and social dilemmas. Herd health advisors and policy-makers are advised to tailor communication to support farmers’ interest and compliance with biosecurity measures and legislation. We propose a framework to a biosecurity index system in a collective context. This system allows both creativity and innovation at farm-level and protects participating farmers from the exploitation of free-riders. Such an index system would contribute substantially to the development of sustainable cultural biosecurity changes at farm- and national level. References Brouwer, M., 1999. Q is accounting for tastes. J. Advert. Res. 39, 35–39. Brown, S.R., 1993. A primer on Q methodology. Operant Subjectivity 16, 91–138. Burton, R.J.F., 2004. Seeing through the ‘good farmer’s eyes: towards developing an understanding of the social symbolic value of ‘productivist behaviour’. Sociol. Ruralis 44, 195–215. Burton, R.J.F., Kuczera, C., Schwarz, G., 2008. Exploring farmers’ cultural resistance to voluntary agri-environmental schemes. Sociol. Ruralis 48, 16–37. Dawes, R.M., Messick, D.M., 2000. Social dilemmas. Int. J. Psychol. 35, 111–116. EU, 2007. From Farm to Fork [http://ec.europa.eu/food/animal/diseases/ strategy/index en.htm]. Festinger, L., 1957. A Theory of Cognitive Dissonance. Stanford University Press, California, pp. 291. Furnham, A., 1986. Response bias, social desirability and dissimulation. Person. Individ. Differ. 7, 385–400. Gächter, S., Fehr, E., 1999. Collective action as a social exchange. J. Econ. Behav. Organ. 39, 341–369.
129
Garforth, C.F., 2010. Motivating farmers: insights from social psychology. In: Proc. Annual Meeting, NMC , Albuquerque, US, pp. 60–67, 49. Gunn, G.J., Heffernan, C., Hall, M., McLeod, A., Hovi, M., 2008. Measuring and comparing constraints to improve biosecurity amongst GB farmers, veterinarians and the auxiliary industries. Prev. Vet. Med. 84, 310–323. Heffernan, C., Nielsen, L., Thomson, K., Gunn, G., 2008. An exploration of the drivers to bio-security collective action among a sample of UK cattle and sheep farmers. Prev. Vet. Med. 87, 358–372. Jansen, J., van den Borne, B.H.P., Renes, R.J., van Schaik, G., Lam, T.J.G.M., Leeuwis, C., 2009. Explaining mastitis incidence in Dutch dairy farming: the influence of farmers’ attitudes and behaviour. Prev. Vet. Med. 92, 210–223. Jansen, J., Steuten, C.D.M., Renes, R.J., Aarts, N., Lam, T.J.G.M., 2010. Debunking the myth of the hard-to-reach farmer: effective communication on udder health. J. Dairy Sci. 93, 1296–1306. Janz, N., Becker, M.H., 1984. The health belief model: a decade later. Health Educ. Q 11, 1–7. Kiernan, N.E., Heinrichs, A.J., 1994. Identification of farm manager types through cluster analysis of calf and heifer management practices. Prev. Vet. Med. 18, 225–236. Kristensen, E., Enevoldsen, C., 2008. A mixed methods inquiry: how dairy farmers perceive the value(s) of their involvement in an intensive dairy herd health management program. Acta Vet. Scand. 50, 50. Kristensen, E., Jakobsen, E.B., 2010. Evaluation of dairy herd health management. Updates on Ruminant Production and Medicine, World Buiatrics Congress XXVI, Santiago, Chile, pp. 53–64. Landry, M., Malouin, J.-M., Oral, M., 1983. Model validation in operations research. Eur. J. Oper. Res. 14, 207–220. LeBlanc, S.J., Lissemore, K.D., Kelton, D.F., Duffield, T.F., Leslie, K.E., 2006. Major advances in disease prevention in dairy cattle. J. Dairy Sci. 89, 1267–1279. Morse, J.M., 1995. The significance of saturation. Qual. Health Res. 5, 147–149. Mulder, L.B., van Dijk, E., Wilke, H.A.M., De Cremer, D., 2005. The effect of feedback on support for a sanctioning system in a social dilemma: the difference between installing and maintaining the sanction. J. Econ. Psychol. 26, 443–458. Onwuegbuzie, A.J., Leech, N.L., 2007. A call for qualitative power analyses. Qual. Quan. 41, 105–121. Patton, M.Q., 2002. Qualitative research and evaluation methods, 3rd ed. Sage Publications, 230. Rat-Aspert, O., Fourichon, C., 2010. Modelling collective effectiveness of voluntary vaccination with and without incentives. Prev. Vet. Med. 93, 265–275. Risdon, A., Eccleston, C., McCracken, G.C., 2003. How can we learn to live with pain? A Q-methodological analysis of the diverse understandings of acceptance of chronic pain. Soc. Sci. Med. 56, 375–386. Stephenson, W., 1935. Technique of factor analysis. Nature 136, 297. Stott, A.W., Gunn, G.J., 2008. Use of a benefit function to assess the relative investment potential of alternative farm animal disease prevention strategies. Prev. Vet. Med. 84, 179–193. Valeeva, N.I., Lam, T.J.G.M., Hogeveen, H., 2007. Motivation of dairy farmers to improve mastitis management. J. Dairy Sci. 90, 4466–4477. van Exel, N.J.A., de Graaf, G., 2005. Q methodology: a sneak preveiw. [http://www.qmethodology.net/index.php?page=1&year=2005]. Weber, J.M., Kopelman, S., Messick, D.M., 2004. A conceptual review of decision making in social dilemmas: applying a logic of appropriateness. Pers. Soc. Psychol. Rev. 8, 281–307.