BIOLOGY 1615 ARTICLE ASSIGNMENT #3 Article Summary Colin Wood
1 I. Introduction Guide dogs, which are used to help any number of people whom are disabled, have long been known to have an ideal weight and size. This ideal weight leads to a much safer and more effective guide dog. The goal of the experiment performed by Helmink, Shanks, and Leighton is to identify a mathematical model which accurately predicts whether or not a given dog will land within this target range. The experiment will analyze the growth curves well over one thousand different dogs of the two different breeds and also compares the difference in the growth patterns in males and females. Depending on the size of the handler, the optimum range for Labrador Retrievers and German Shepherds has been found to be 18-32kg.The experiment will assist in identifying another tool to accurately predict whether or not a dog will reach this range. By having a dog that has enough mass those who are blind are able to be directed more easily. If a dog is too large, blind handlers are prone to not being capable of handling the animals appropriately. II. Materials and Methods During the experiment 18,235 recording of 1558 German Shepherds and Labrador Retrievers weights were recorded. Because growth of such a complex organism can take so long the experiment required a large data set to gain an accurate average value. This resulted in an experiment that had to be conducted over a 5 year time period, 1992 1997. The dog s weights were recorded initially at very frequent intervals. This was done because animals tend to have an extremely fast growth rate in the earlier period of life. AS they get older eventually their growth will stop which means that there is no longer a need to analyze the data. The experiment limited the maximum age of dogs to be measured to 551 days.
2 The next step involved in the experiment was the method for analyzing the data. The Gompertz function was selected as the mathematical model that would best fit the needs required in this experiment. The reasons for using this model were stated as, adequate fit of the data, having a suitable point of inflection for modeling dog growth, and previous use of the Gompertz function in the literature. (Helmink, Shanks, & Leighton, 2000) This function seemed to be ideal as it models the growth rate of something that starts off slow, then speeds up in the middle, and lastly slowly comes to a stop near the end of the growth cycle. III. Results The vast majority of the dogs landed within the optimum range of 18-32 kg however, it was uncovered that many of the animals have a tendency to gain weight when with caretakers rather than at the facility. What this meant was that, having an animal measured while living at the facility was much more accurate than having it measured while living with its caretaker. When living with a caretaker far too many factors such as diet and activity level can force an animal to stray far from the average growth or median growth size. At the end of the experiment the team recognizes that the growth curves which they have uncovered should merely be used as an aid allowing them to predict whether or not a given animal might hit a target range. IV. Discussion The results identified an average growth path/model that breeders and trainers can use to assist in their selection of guide and Seeing Eye dogs that will fit within the ideal weight range of 18-32kg. The issue with some of the data was the fact that when an animal left the facility they could no longer control the factors that might contribute or take from a dog s
3 growth rate. It is important to note that there is no possible way for us to predict with certainty that a dog will attain a certain size. The best we are able to do given the current technology available is to collect a data set that is large enough to weed out anomalies and find some common characteristics within that data set to make and educated estimation.
4 Reflection This assignment has allowed me to become much more of an objective reader and has enabled me think much more critically. Seeing as how I hope to be a professional software engineer in not too long I believe being able to read through others work and identify how they came to their conclusion is going to be critical on a daily basis. I cannot tell you how often I find my self doing exactly what I did in this assignment when trying to learn to write a new algorithm. I have to go through someone else s work and identify exactly where they went right and where they went wrong. Being able to then take someone else s work and explaining it in my own words will be invaluable in the work place. This ability to communicate effectively is something that you must be able to do to survive. All to often I believe that being able to regurgitate is something we don t learn well when all we are hoping to do is fill in the blank and get the right answer in college. This assignment has forced me to read through an article, gather my thoughts, and in my own words, tell the story these folks told without changing any of the data. What an amazing exercise for the brain!
5 Works Cited Helmink, S., Shanks, R., & Leighton, E. (2000). Breed and Sex Differences in Growth Curves for Two Breeds of Dog Guides. Journal of Animal Sciences, 27-32.