Alycia Holwerda

            Since I majored in statistics, many of my courses directly correlated to research. In Statistics 450: Regression Analysis, Statistics 458: Categorical Data Analysis, and Statistics 492: Capstone; all of these courses followed the same format. My professors taught me various theories and methods to the related statistics course, and I would practice these methods through final research projects. The teacher would give the students a dataset or the students would find a dataset to analyze. Then, students would utilize various statistical methods to gain insights from the dataset. Finally, the students would write a report as well as present findings to the class. I have attached my final reports and presentations at the top of this webpage. In Statistics 450, I collaborated with a partner to research roller coaster data through regression. In Statistics 458, I worked in a group to analyze low birth weight data. Finally, in Statistics 492 Capstone, I utilized all statistical methods, processes, and theories to exemplify my statistic capabilities gained throughout my undergraduate career. For this class I chose Petfinder data to learn more about various characteristics as to which animals would be adopted from a shelter.

 

            When working with the dataset, I had to ethically use this information (Information Literacy: Level 2). In the statistics field this meant following the appropriate methods and assumptions. A huge part of working with datasets was having to clean the data. The data given was anywhere from 50 rows and 3 columns to millions of rows and hundreds of columns. The data was almost never completely ready to be used initially; data cleaning was always necessary. A statistician or data analysist needs to clean the data properly to be able to ethically use the information. For example, missing data was almost always present in datasets, and a statistician has to handle the missing data properly. They could not fill in the missing data with random numbers but had to either replace it with a proper value such as a 0, median, or mean, or not use that field at all.

​

            After cleaning the data, the next step was to apply statistical methods to gain insights whether it was creating a predictive model or learning characteristics about the data. When working with a huge dataset it was important to focus on the key variables. A statistician could spends years working with the same dataset. An important characteristic of a good data analyst was being able to find the key insights in a dataset to present to the data sponsor. Through my statistic projects, I found the key insights that were the most valuable to present that addressed the questions asked by data sponsors. For example, when using my Petfinder data I told which animals were more likely to be adopted. Humane societies could then use this information to focus their energy on helping the animals that did not follow these characteristics to help them be adopted. I had the ability to evaluate and incorporate selected information into knowledge base (Information Literacy: Level 3).

​

            Through my capstone project I was able to not only conduct original research but also to use information effectively and ethically to accomplish my research goal (Information Literacy: Level 4). My research goal was to see which types of animals were more likely to be adopted. I was able to discover key characteristics as to which animals were more likely to be adopted.

​

            My statistic capstone project was my individual original research project to gain knowledge. Through hackathons I was able to conduct original research, but this was in a group setting. When beginning this project, I had to find my own dataset that interested me. I looked at various data, analyzed which work workable, and then decided which would interest me most. I found a dataset about Petfinder.my. This data gave various information on dogs and that were or were not adopted in Malaysia. Since I love animals and believe in adopting, I chose this dataset. I began by proposing a research question that extended my knowledge and allowed me to practice my discipline (Original Research: Level 1). My research question was: Which type of animals are more likely to be adopted in the Malaysia area?

When working with this data I had to further identify what I meant by “type”. I decided to see what type of animal, color, health conditions, age, type of fur, and vaccinations. I was able to further develop my research question (Original Research: Level 2).

​

            After developing my research question, the next step was to analyze the data. I learned what each variable meant and then cleaned the data. I had to reorganize data labels, handle missing data points, and create binary variables. Next, I analyzed the distributions of each variable, odd ratios, and correlation. The most important matter was creating a predictive model through trying different forward, backward, and stepwise selection. I then identified the key characteristics of adoption and which characteristics had a higher adoption rate. Through this project I was able engage in creative practice that extends my statistics discipline (Original Research: Level 3). I was able to create new research on to what type of animals were more likely to be adopted in Malaysian. I exhibited my completed research to my class in front of my professor (Original Research: Level 4).

​

            After completing my statistic courses, I can now graduate with my Bachelor of Science degree in Statistics. Each day at my job at IBM I will apply the knowledge I learned at MNSU each day to gain more insights for IBM to further improve their company. Research will be a huge aspect of my career. I hope to be able to continue to research various data to help improve society and benefit the world. Various areas I hope to find solutions to include pet adoption, climate control, endangered species, world hunger, and cancer. Even though these are huge topics, I just hope to be able to use my skills to help benefit the world in some sort of way through research.