The development of a non-invasive tool to measure stress and pain in animals has significant applications in assuring the welfare standards of the New Zealand agricultural industry and boosting the value of its products in overseas markets. Here we are interested in stress involved in the disbudding (removal of horns) of cattle.
This video discusses how the emotional responses of cattle can be detected and quantified non-invasively through measurement of eye-temperature, using infrared thermography. Four treatments are compared in this designed study.
Other applications of this technology involve screening for swine flu at country borders, lie detection and detection of injuries in animals and consequential inflammation.
This video requires a web browser capable of HTML5 or Flash video.
If the video does not play, you can download it here (right-click and choose Save Link As), then open it with a video player such as VLC.
TasksThe data are presented in the Excel file Disbud stress.xls. These tasks make use of the Bootstrap Procedure, instructions for which can be downloaded here (PDF).
- Task 1: Discuss the importance of a randomised design. How many treatments are present? Why might one of the treatments have only 6 cattle? Is there a control treatment that other treatments are compared with?
- Task 2: For each treatment calculate the mean eye temperature and the standard error for each treatment mean (sem). Use this to construct a 95% confidence interval for each mean. (Use the large sample value of 1.96 or 2.00 in these calculations. In practice a t value should be used in each case. The t value will have 8-1=7 degrees of freedom for treatments 1, 2 and 4 but only 6-1=5 degrees of freedom for treatment 3).
- Task 3: For each treatment mean calculate the bootstrap confidence interval. Use the excel procedure bootstrapper.xls
- Task 4: Compare treatment means for the DB and LADB treatments. Calculate the difference between the means, the standard error of this difference and the 95% confidence interval for this difference (using the large sample approximation of 2 times the standard error of the difference between the means). What conclusion can be drawn from this about the effect of the local anaesthetic?
- Task 5: Calculate the bootstrap confidence interval for the difference between the two means in Task 4. Compare with the Task 4 confidence interval. Again use bootstrapper.xls
- Task 6: Set up the 95% confidence intervals for the other four comparisons of pairs of means. (Eventually at University you will see a better way of making all these comparisons using the analysis of variance procedure : ANOVA).