Nevertheless, whilst findings should be interpreted in the context of the game, the context squares played an important role in keeping the game situated within the challenge of bTB. Moreover, participants commented that they found the process enjoyable and a helpful way of talking about cattle purchasing, and it was notable that the game play prompted conversations about why a decision had been taken between participants. Farmers were encouraged to talk through their purchasing decisions as they made their choices and explain their reasons after each purchasing event. Farmers were asked about each of the behavioural interventions during and at the end of the game. These discussions were recorded within Zoom, transcribed and cross-checked with notes taken during the game. Analysis of in-game cattle purchases identified and recorded each factor mentioned by farmers in their explanation of their purchase choice. Similar factors were grouped together and organised into five main categories. Transcripts were analysed thematically within Nvivo to elicit the key similarities between participants in relation to their views of the information provided and the rationales for their purchasing. Overall, the most frequently mentioned factors were the vaccination status of the animal and its status in relation to production diseases other than bTB. When purchase factors are aggregated into categories, the most important factors were related to aspects of the animal on sale and production diseases, followed equally by bTB and management factors. Farmers were particularly heavily swayed by the Johne’s disease2 status of each purchase choice, acting as an anchor or reference point for all other adverts. Around half of all disease factors were specifically about the vaccination status. This suggested that purchasing decisions were not multi-factorial but could be based on one criterion. As Player 3 commented for all his purchases, “Vaccination for major diseases, that’s what I am really looking for”. Years free from bTB was the third most frequently mentioned factor. This is likely to reflect the fact that it featured in every sale advert and suggests that information on bTB at the point of sale may provide a limited cue to some purchasers. Similarly, strawberry gutter system bTB compensation was only ever discussed in relation to adverts where compensation was mentioned.
Whilst the frequency of these factors is likely to be influenced by the information displayed in the adverts, results reflect previous research that has sought to identify the most influential factors in cattle purchasing . Table 3 shows how these factors vary between different purchase scenarios. For replacement dairy cows, production diseases were the most significant factor, followed by animal factors and then bTB. For purchases of calves, bTB was the least important factor, whilst management factors were the most important. For purchases of in-calf heifer calves the most popular factors were related to the animal, whilst bTB related factors were third. In contrast to the purchasing factors, adverts with high bTB ratings were chosen more frequently. In total, 39 in-game purchase choices were made which involved considering adverts with different bTB statuses. Over half of these in-game choices were of cattle with a high bTB rating . Fourteen in-game purchases were of cattle with the lowest bTB status . One further choice was of cattle whose status was on the midpoint and between the lowest and highest options. For all game players, ten consistently chose purchase options with the highest bTB rating, five the lowest, and three chose a range of options.Farmers suggested that the comparison needed more context to be valid: parishes could vary in size and by number of farms. A more reliable and standardised denominator may have more salience. However, discrepancies between parish and herd bTB ratings prompted some farmers to indicate that this was something that they would follow-up with the vendor to get an explanation. 20 of the 37 in-game cattle purchases involved cattle that would receive 100% of statutory compensation if the purchase was subject to a post movement test. Comparing choices made in each scenario reveals that most farmers did not have a preference for higher or lower compensation, five always chose options with higher compensation, and 3 chose options with lower compensation. Of the 18 in-game purchases, only four were of purchase options that had the highest rating or 95% satisfaction. The remainder were purchases of cattle with lower purchaser satisfaction. In scenario 4, the good farmer information featured on half of the purchase choices. Participants chose an advert featuring a good farmer logo in 14 out of 18 purchase choices. Choices were distributed equally between the highest and lowest good farmer ratings .
In reflecting on their purchasing choices and the information that was most salient to them, farmers articulated a purchasing strategy best described as ‘fitting the system’. This strategy aims to fit or match new cattle purchases to the farm system to ensure its continuity. When faced with a range of purchasing options, ‘fitting the system’ therefore acts as a kind of ‘radar’, honing on those factors that are most pertinent to the system. In-game purchases reflected the need to match systems in a number of ways. For dairy cows, players commented that cows that were cubicle trained were preferred. Information on what cows were being fed was not contained in any adverts, but players suggested that they would want to know that information to ensure a match to their own systems when possible. For calves, Player 16 chose advert 2, justifying the purchase because from the advert, it appeared that the ‘set up was very similar to what we’ve got in terms of the conditions, the vaccinations and the colostrum management’. The importance of a similar setup was to minimise the stress placed upon animals when they are moved and for them to have similar levels of immunity, so that they are not susceptible to illness. Whilst fitting the system provided an overall framework for cattle purchasing, dimensions of good farming were important in shaping how decisions were made. The challenges of fitting the system meant that trust and reliability in the seller became key factors in deciding what to buy. This was evident when farmers were asked to choose between an agent supplying cattle or buying from their neighbour. In this scenario, farmers highlighted the importance of local knowledge. For example, Player 3 commented that, “if it’s the same cow then you go for the neighbour, you know more stuff from driving past”. Similarly, Player 12 suggested that they “would walk away [from the dealer] and look at the neighbours’ [cows] because we know their farming system and they are in tune with what we are doing”. Other dimensions of local knowledge included the ability to draw on vets’ knowledge and their connections with other vets. Player 9, for example, suggested that their vet could speak to the vendor’s vet to “get into the nitty gritty and find out why the animals are on sale”. The effect of providing information on the good farming status of the vendor had a mixed effect. Firstly, purchase choices with high good farmer scores were not widely chosen, indicating that other systemic factors took priority. Nevertheless, farmers reacted positively to this rating, comparing it to ‘Amazon-style’ ratings and demonstrating the face-validity of this good farming metric. However, whilst farmers thought the principle of articulating vendors’ qualities in this way was good, it prompted further questions about what precisely the rating would mean, who would organise it, and how reliable it could be. Satisfaction of previous sales was generally seen as appropriate, but there were concerns about how easily this could be manipulated by ‘fake’ or misleading reviews arising from a genuine mistake by the vendor or purchaser.
Similarly, farmers were concerned about the ability to compare between vendors if one had fewer sales than the other. However, it was not always easy to elicit from the pictures the quality of the animal, farmer or farm, hydroponic fodder system prompting players to comment that they would prefer to be able to visit the farm. This offered farmers to gauge the trustworthiness and reputation of the vendor by being able to ask additional questions and determine from their answers whether they were ‘good farmers’ or not. This could include, for example, vendors’ knowledge of the animal’s history, and the records they keep. In this sense, purchases would partly be based on the farmer and the farm. Farmers commented that they would like to see that the farm was clean and tidy, the housing was of good quality and that the vendor had the ‘right’ attitude. Secondly, the challenges of ‘fitting the system’ also impacted upon the relevance of bTB information and its ability to reflect good farming. Whilst farmers generally preferred high status bTB cattle, their choices reflected their attempts to match cattle to their own circumstances based from other information available. In general, farmers valued purchases with a higher number of years bTB free. However, they also viewed the bTB test as an indication that an animal was ‘saleable’ and there was no real consensus on the threshold of what constituted a ‘safe’ herd. Five or more years was generally seen as good, although some farmers suggested lower. In each case, however, the scarcity of available cattle with high bTB status meant that a better guide was to buy no lower than their current status. The significance of bTB varied between purchase types and each players’ experience of bTB. Where farmers had experienced many outbreaks and farmed in expectation of an outbreak, information on bTB was less important. This reflects fatalistic attitudes towards bTB described in Enticott . However, where players had experienced a recent bTB outbreak, which had caused significant farm management problems, information about bTB was more important. Information on bTB was more likely to be salient when it was timely: farmers who were restocking following a bTB incident particularly valued this information. However, it was not the only factor: Player 9, for example, suggested bTB accounted for 50% of the purchase decision, and other factors could over-ride its significance. In this sense, fitting the system could reflect the wider epidemiological picture surrounding the farm. For example, Player 9 commented that “the closer geographically you are then closer to the same TB situation, [its best to] stick with the problems you know”.
However, for some animals, such as calves, some farmers suggested these dimensions of local fit were not important. Player 2, for example, suggested that “young calves spend so little time in the environment to pick up the disease”. In general, information on bTB appeared to play an ‘arbitrating role’ helping to differentiate between two equally ‘good’ animals for sale. This seemed to be most relevant for compensation incentives. Where adverts appeared to be of similar quality, the potential for additional compensation could sway the decision, all other things being equal . As full compensation was linked to the completion of post-movement testing, the attractiveness of this incentive also depended on the relative ease of completing this test. Where farmers were already frequently testing, the requirement to post-movement test was not considered onerous, meaning animals with full compensation were more attractive. Equally, the extent to which information could arbitrate between two adverts depended on the value of compensation itself. This paper has investigated the salience of different behavioural interventions to influence farmers’ cattle purchasing decisions. In this section, we consider the wider implications of our research. Firstly, the development and use of a scenario-based game has much to offer studies of bio-security and other land-use policy issues. Participants enjoyed playing the game and reported that it helped them to think and talk about their cattle purchasing decisions. Following Quine et al. , our purchasing scenarios were realistic, prompting some participants to reflect on times when the scenarios had played out in real life. Importantly, the use of the game also highlights the need for methodological triangulation when considering the impact of behavioural interventions within farming. Results from the game varied according to methodological and analytical techniques. Based on the analysis of purchasing rationales, results suggested that purchasing was primarily related to production factors. Analysis of the in-game purchases suggested that farmers preferred cattle from farms at a low-risk from bTB.