A prospective cohort study of game-based learning by digital simulation of a pig farm to train agriculture students to reduce piglet mortality

With the use of highly prolific sows in swine production, high piglet mortality before weaning remains an unsolved problem. The annual report from selected EU countries shows that preweaning mortality rate increased from 12.9% in 2014 to 13.3% in 2015 [1]. Hypothermia, starvation, and crushing are the primary causes of mortality in piglets during the first 48 h after birth [2, 3]. Therefore, improved management skills are crucial to reduce the mortality rate [4]. Despite general agreement that these skills are vital, there is little opportunity for students to practice them before being employed on a pig farm. At agricultural colleges novice students are presented to skills they must master when working on pig farms, but training is when they are employed on farms. The training of international labour is primarily the responsibility of the individual farmer. International labour has often no experience from pig production and start from zero when they are employed in Denmark. However, it is mandatory that all employees who are going to inject livestock animals must complete an 8-h medication course. To achieve better training of new international employees and to offer Danish agricultural schools new learning tools, we have developed “Game of Piglets.” The game is a virtual pig farm that has as its objective to allow students to practice procedures involved in external biosecurity and farrowing management. “Game of Piglets” belongs to the genre of simulation/ adventure games. Core competencies highlighted in the game are farrowing aid, recognition of sick sows after farrowing, aseptic conditions in surgical interventions, and optimal piglet environment. These competencies are developed by a series of tasks that must be solved in a particular order, just as in a real pig herd. The game’s scoring system is a piglet survival barometer. To see how the task of farrowing assistance are created please follow the link: https://www.youtube.com/watch?v=ukmxfhVT-zQ

If a task is performed incorrectly or omitted, it results in decreased piglet survival. At any time in the game, students have the opportunity to ask the farm manager about how to solve the tasks. When all tasks are completed and the student leaves the farm, an evaluation of tasks resolved correctly and/or a description of the mistakes that were made will be given. Screenshots from the game is displayed in Fig. 1.

Learning games are widely used in other disciplines such as health, business, military, and mathematics [5]. The most popular learning game genre by far is simulation, but other genres as action, adventure, strategy, and puzzle are also used [5]. Simulation training in health profession educations is associated with positive effects on knowledge acquisition, skill, and behaviour [6]. Virtual patients are extensively used for skills training in undergraduate nursing educations [7]. Virtual patients allow nurses to learn and practice in a safe, virtual environment where time can be paused. The pause allows time for self-analysis [8]. Thus far learning games for veterinary use is increasing, but only a few learning games have been developed specifically for agriculture education. In this paper we will introduce a learning game for agriculture students designed to help them improve their theoretical and practical skills in the field. The objective of the study is to measure knowledge acquisition in farrowing management among students only exposed to classroom instruction, students whose education was supplemented by “Game of Piglets,” and students who also had hands-on instruction. Furthermore, students’ perceived self-efficacy surrounding the practical skills of farrowing management was also measured.

The summarised mean score from the multiple-choice test is displayed in Table 2 and the box plot in Fig. 2. The simulation group had the lowest SD at 6.76. A significant difference between the simulation group and the hands-on group was found at 28% (P = 0.01). Between the simulation group and the classroom group we found a difference of 19% (P = 0.1), but not significant.

On average, the simulation group had slightly higher scores than the negative control group (difference = 8.4%) and the hands-on group (5.6%). In addition, there was a significantly smaller standard deviation among students in the simulation group compared to the hands-on group, which to some extent was caused by a lower number of students with wrong answers. However, the differences in SD were not significant between the simulation and classroom groups.

For the self-efficacy questionnaire, the distribution of answers is displayed in the diverged stacked bar charts in Fig. 3.

In response to the statement “I can choose the correct antibiotic for injections against MMA following instructions from the veterinarian,” the simulation group scored significantly higher (P-value 0.001) than both the classroom and hands-on groups. For the statement “I can perform the correct procedures for entry to a pig farm”, the hands-on group scored significantly lower than both the classroom and simulation (P-value 0.04) groups. In responding to the statements dealing with tail docking and castration, the hands-on group scored significantly lower than both the classroom and simulation groups (P-value 0.03 and P < 0.001, respectively). Answers to the satisfaction survey are displayed in Fig. 4.

After playing the simulation game, 84% of the students agreed that they had learned how to increase the survival rate of piglets in the farrowing unit. 84% also agreed that they enjoyed playing and wanted to play other learning games similar to “Game of Piglets.” However, 38% answered that they did not want to play this game in their spare time, and 40% found it difficult to play “Game of Piglets.” The difficulties were due to navigation and the lack of immediate feedback in the game. The lack of feedback was a deliberate design choice to encourage cognitive thinking rather than trial and error.

The present results indicate that students in Danish agriculture schools enjoy using digital learning games. The results of the multiple choice test were not significant, but indicated that using a simulation game in the curriculum had a positive effect on learning outcome and improved results in particular in the lower-performing half of students. It was anticipated that both the simulation and hands-on groups would perform significantly better than the classroom group in the multiple choice test. The multiple choice test was chosen to evaluate many students with low amount of resources. The difficulty of the test was carefully designed to assess skills and knowledge about farrowing management. A pre-test before instruction could have verified if students were able to answer correct even without instruction. The exposure time for both the simulation and hands-on groups was set to two hours. Two hours were the time that was planned in the schedule for a farm visit in the course on all schools. Therefore the time for gaming was also set for two hours to unify the groups as much as possible. This is likely too short an exposure time to have a significant effect on memorizing and retaining knowledge of farrowing management. The real farm is not an optimal learning environment if time is sparse, as there are many distractions on a real life pig farm. Therefore, the intended learning objectives may often become secondary on a first-time pig farm visit. The pig farm is more optimized for learning when students are familiar with the smell, sound and the condition in a pig farm. The simulator was easy to install, but it took some time and effort to become familiar with. Most of the students became competent at navigating and using the functions in the simulator during the exposure time. It is established that repetition and experience have a positive effect on memory and learning [13]. From this perspective, more exposure time to either simulation or hands-on experience could have changed the outcome of the study. Additionally, agriculture students claim that they learn much more from practical training than from classroom instruction. This opinion was found through a qualitative study conducted on Danish agriculture colleges in 2016 in which students were asked about the relationship between school and practical training [14]. However, the results of the multiple-choice test did not support that finding. The perceived self-efficacy questionnaire shows that the hands-on group scored significantly lower when it comes to procedures for entering a pig farm as well as for tail docking and castration. It was expected that the hands-on group would score highest since they had experienced reality and thereby have the best prerequisite for doing these procedures. Nevertheless, the low score could be attributed to the fact that these students experienced the complexity of reality. This may have increased their anxiety and reduced perceived self-efficacy because of exposure to reality without proper and thorough preparation. In surgical courses at veterinary schools, students have a high level of anxiety prior to surgical procedures on live animals. But with simulated skills lab training on models beforehand, anxiety levels before performing live surgery decreased significantly [15].

Assessment of the efficacy of educational interventions often includes Kirkpatrick’s Four-Level Model [16]. Level 1 is evaluation-reactions, Level 2 is evaluation-learning, Level 3 is evaluation-transfer, and Level Four is evaluation-results. The students’ positive response (Level 1) to the satisfaction questionnaire indicates that they had a positive experience with “Game of Piglets.” Although these positive reactions are subjective and are not the sole determinant of learning, they play a key role in assessing student engagement that in turn enhances active learning and understanding. Still, such a survey data collection tool has the inherent limitations of subjectivity, that is, self-report bias. Evaluating progress in learning (Level 2) moves beyond learner satisfaction and attempts to assess whether students have advanced in skills and/or knowledge as a direct result of the educational intervention. To this end we developed the multiple-choice test as a tool for measuring knowledge acquisition, and the perceived self-efficacy questionnaire for assessing skills. Despite this, it is still difficult to attribute the improvements in knowledge and skills acquisition solely to the use of tailored educational games. Measuring Levels 3 and 4 is costly and time consuming, as the students’ acquired professional knowledge and skills must be evaluated through demonstration and clinical examination. These are the recognized limitations of this study.

The results of this study show that custom designed digital educational game simulations of pig herds can be used as an additional training resource for undergraduate agricultural students. Though not significant, the results indicate that the simulation game had a positive effect on learning outcome and tended to improve results, especially among lower-performing students. The students’ commitment to game-based learning was positive and they requested more game modules with which to practice their skills. Digital games offer the new generation of teachers an opportunity to mix traditional approaches with simulation training. More studies are needed to explore the pros and cons of game-based learning in agricultural education.