RMDA Workbook: Welcome to my mind
Welcome to my Msc Endangered Species Recovery and Conservation workbook! This is for our ‘Research Methods and Data Analysis’ module. I’m here because I want to work in conservation despite having chronic illnesses and disability; there’s a stigma that the industry isn’t made for people like me, and I’m wanting to prove this wrong.
Please note: I am suspected of having dyscalculia (math dyslexia) so please do not rely on me to have the math right. I probably do not have the math right.
Above is a series of animals I really like, just so you get to know me a little more. That was made using images and frames sourced from the design app, Canva.
I’ll be updating this and publishing new work at the top; so if you’d like to see anything from the creation of this (23rd September 2024), you’ll need to scroll to the end; or use the contents box.
______________________________________________________________________________________________________
Life Updates: New Year, Same Me
It’s been a while since I updated the workbook, and I believe that I’ve come a little way since starting. It’s now January 2025 and I’ve really struggled on this module.
We’ve got our summative due, so I’ll be uploading that and hopefully it’ll be okay. I’ll be doing the summative in a new work book, but will probably either give a final update here - or drop the link.
I’ve (again) made the image on Canva, and used a Helen Keller quote on optimism to show I am trying to be hopeful. I’m at least grateful the summative is on sharks as it means I have an area I’m confident within.
______________________________________________________________________________________________________
Week Seven b: Formative
Due Date: 11th November 2024
Still feeling really unwell, but the show must go on (as they say). We were presented with an article that had the front page removed. Using only the base text, our assignment was to create a title, abstract, and some key words for indexing.
My attempt is as follows. Not seen here is the formatting from the submitted PDF document. Size 10 font in Times New Roman. Double line spaced. This is due to the standards required for scientific journals.
“Effects of Farmland Agriculture on the Giant Galápagos Tortoise Populations of Santa Cruz Island
Author/s Name. Affiliated with James Cook University. Email address of author.
Abstract
Behavioural analysis of how tortoises in the Galápagos are utilising the land space available to them on patches of farmland, abandoned farmland, and tourist areas.
Analysis was conducted using focal observation by sampling 1m patches of ground vegetation every 5 minutes during a 30 minute observation period; the team took note of the height, density, and percentage of live vegetation. Tortoises were observed in two seasons (wet and dry) with the use of a FLIR camera for identification and temperature statistics; this was to monitor the activities seen from walking, eating, and resting. Behaviour was modelled using the Dirichlet regression method due to being the best fit for multiple proportions.
Our results found that the tortoises were diurnal and preferred vegetation that was low-density for eating and walking, in contrast they preferred high density abandoned agricultural areas for resting.
It has been found that the farmland agriculture is playing a large part within the conservation efforts of Galápagos tortoises; and the properties of vegetation determines the behavioural activities taking place within each agricultural zone of Santa Cruz Island.
Keywords: Conservation | Agriculture | Galapagos | Herpetology | Chelonians | Behaviour”
So, I learned that the Dirichlet regression technique is a thing as I’d never heard of this previously. I also learned about key word indexing as it’s how people may find your article.
The usual feelings collage is below; but I’m happy with the task as it didn’t involve a lot of numerical data, there was some, but it wasn’t too heavy on the matter. I also really like tortoises. Got invested even more so during an episode of Extinct or Alive with Forrest Galante where they found a Fernandina Island Tortoise.
______________________________________________________________________________________________________
Week Seven a: Mean Tests
Date: 5th November 2024
Absent due to being unwell, the cold mentioned a few weeks ago is still present and I’m still really struggling with it. I’ve tried my best to understand what’s happened and to my understanding, it’s all about the kinds of tests you can do. The penguins made a return!
I learned about some of the statistical tests we can do for various types of data.
Will take another look at this when I’m feeling better, but I will be doing the formative task set. Image collage just to make myself feel a little better.
______________________________________________________________________________________________________
Week Six: Frequency Data & Group Project
Date: 29th October 2024
Group Project: I worked with my good friends on this, which helped with my imposter syndrome. We looked at the Lynx data because all of us like big cats. My role was to formulate questions and do the background research. Here’s the work book in which we submitted the assignment: Link here!
Screenshot of group project goes here!
Lecture: Initally not confident, but then realised a frequency test was something I’d sort of done with week 5. So not overly bad. Still not learned how to code, but keep practicing.
______________________________________________________________________________________________________
Week Five: Choosing the Right Analysis
Date: 22nd October 2024
Week 5 and my positivity was short lived. Still not feeling great, still very tired. Now thinking back to my BSc, I was not good at choosing the right analysis, I was not good with any sort of statistical analysis. So week 5 is not a fun week.
I prefer categorical data, I’m a shark attacks girl, we’ve learned this. I like my locations, my provoked vs unprovoked, my species involved, all of that. Numbers, I’m not good with. Numerical data is not my forte.
The tasks are to formulate a question, make a graph, then assign it a statistical test. So fingers crossed, I’m also going out of my comfort zone and will be looking at ferrets. I collected this data myself while at a fundraiser for a ferret rescue, I adopted my own ferrets from a rescue after this event. My graph is below.
Question: Is there a correlation between sex and fur type?
Test: Chi Square due to the categorical nature and dependency testing.
Hypothesis:
Null Hypothesis (H₀): Fur type distribution is independent of sex.
Alternative Hypothesis (H₁): Fur type distribution is dependent on sex.
Feelings: I enjoyed using the data I collected at the charity event, and talking about ferrets. I did struggle with assigning a test and formulating a null and alternative hypothesis. Math is still confusing.
______________________________________________________________________________________________________
Week Four: Data Exploration & Scientific Hypotheses
Date: 15th October 2024
We are now into week 4, and I’m not feeling great in general. Trying to make some positive lifestyle changes, get myself into the “I can do this” mindset. Between my cold I’ve picked up from family, and adjusting to these lifestyle changes, it’s tough. Why is this relevant? Well, I’ve noticed that in the lectures I’ve clicked with, I feel good. I’m positive, I’m looking forward to them and getting stuck into the assignments. Here… it’s the opposite. I don’t understand it still so I’m struggling with it and then the assignments worry me, so I don’t have the same positive attitude - that then impacts my whole day and general sense of self for that day. Hence all these positive changes I’m trying to make.
Now, that little context session is over, lets get on with the write up.
Our task this week is to look at the following charts/graphs and describe which tests can be associated to them.
^^ Graph One: Box Plot
Box plots are usually used within t-tests, Mann-Whitney, anova, and Kruskal-Wallis. The benefit of using this type of graph is that it shows the outliers, and highlights the median area of the data.
^^ Graph Two: Density Plot
Density plots are usually used for normality testing such as t-tests and anova. They are popular due to their straight forward look; along side this their ability to show continuous data while also giving a comparative analysis makes them a great choice.
^^ Graph Three: Scatter Plot
Scatter plots are used for a range of tests; this particular plot also has a regression line. The tests they’re used for mostly consists of Pearson and Spearman’s Rank; they’ve also been used in anova, Kruskal-Wallis, and Mann-Whitney. They’re a good way of displaying data when you’re wanting to show relationships by identifying any correlation (positive, negative, or none) and if you’re wanting to make predictions.
^^ Graph Four: Bar Chart
Bar Charts are a familiar graph for many, these are (in my opinion) a timeless classic. Used for Mann-Whitney, Chai-Square, t-tests, and anova tests; they are a popular due to their simplicity and ability to show frequency and for making comparisons.
My feelings this week are as follows: I’ve been okay with identifying the graphs/charts, and then finding out what they’re used for. I’m still struggling with the coding element of the module, and I don’t see this changing. Overall, a more positive week and I do feel a little better about my work in this formative. I then found out we’d possibly be studying sharks as the summative topic and this, this is good news. I’m good with sharks. They’re my thing. Do we remember week one when I literally CHOSE to study their data? Maybe I’m not as much of a lost cause as I thought, I just need to get this coding malarkey down. As usual, the canva collage is below to show how I feel.
______________________________________________________________________________________________________
Week Three: Data Wrangling and New Data Set
Date: 8th October 2024
Week three and I still haven’t gotten to grips with this yet. After asking a friend to see what they’ve done so I know if I’m on the right lines - I’ve established I will probably (confidently) fail this module. There’s been tears, tantrums, and now the acceptance I’m just not good at this. The iconic Bakeoff scene with James Acaster springs to mind. The inner perfectionist in me hates this lack of confidence; and the GCSE student who never studied because “I don’t need to” feels very humbled. In positive news: The penguin data set has now been located!
We’ve been asked to come up with two questions for the diamond data set, one good and one bad. Here they are:
Good: Is the carat size the only factor when deciding the cost of the diamond?
Bad: What jewellery item are the diamonds being placed in?
As someone who is engaged, diamonds and their properties became a subject I looked into. Using this data to answer my good question. The pink arrows are for row 6, a 2.29 karat diamond; the green represent row 3, a 1.51 karat, in jewellery terms, the green is considerably smaller. Despite being a larger diamond in row 6; it is less value than row 3. The colour and clarity is more important here than the karat alone. The data set is from ‘R for Graduate Students’ written by Y. Wendy Huynh.
We can’t answer the bad question, as the data isn’t available for this. Someone wanting to study the uses of diamonds wouldn’t be able to use this data set for that particular question.
My feelings this week are represented by another canva collage, I’ve tried to show how I’ve felt this week. It’s all very confusing (bottom right); it feels like I’m doing the math and equations seen in the top left. We see I did manage to eventually load the diamonds data set. I do very much feel like each step forward is teetering on the edge (top left) and diamonds to represent the data set (bottom left). Lastly, the quote is the iconic one by James Acaster during a hard task on Bake Off as I mentioned it above.
________________________________________________________________________________________________
Week Two: Ethics
Date: 1st October 2024
We didn’t have a data set today, we looked at ethics and things which impact this, such as culture. I had typed out a huge paragraph on this, but I deleted it.
Why did I delete it? Because I understand that we all have different views on things, you can have siblings growing up in the same household which have different codes of morality. It’s such a deeply personal thing, and we will never share the exact same views and morals with people. Our life experiences, socio-economic status, family environments, culture… all of that shapes who we are. So this leads me to ask myself: why should I publish my sense of morals and ethics like it’s important when we all have those differences? The conclusion I came to was not to publish it.
I did learn a new term: Helicopter Science.
Image below is a small collage made on the design app, Canva. I asked for images about ethics. This just looks and feels nicer when I add something that isn’t plain text.
______________________________________________________________________________________________________
Week One: Introduction to ‘RStudio’
Date: 23rd September 2024
During todays introduction on the 23rd September 2024, I’ve learned that we’ll be using this software called ‘RStudio’ and honestly, I find this deeply confusing but I will try my best. I’m human, I have flaws and faults. Here seems a place to admit them. I’m doing this as part of my MSc in Endangered Species Recovery and Conservation. I’ve never done any coding before, I’ve never used R, this is a brand new experience. Please feel free to follow along as I attempt to learn how to do things.
I couldn’t find the penguin data, so I’ve used some other data from the Global Shark Attack File. Sharks are my comfort zone, and since this is week one, I’m sticking to a topic I feel ‘safe’ working within.
I will be looking at data from 1995, 2005, and 2015. Before looking at this data, my prediction will be that in each decade the number of figures will increase; and more shark species will be listed.
Figure 1 shows a bar chart detailing the number of shark attacks, each year has been split into provoked and unprovoked. My initial prediction was incorrect as 2005 had more human shark conflict than 2015.
As we see here; the small graphs [L-R] show that my suspected pattern of species increasing with decade did occur. Below shows each graph in a close up.
Figure 2. The shark data from 1995. The species here are mostly well identified; only two of the recorded species were unspecified, but recorded to be within the sand shark family.
Figure 3. The shark data from 2005. When looking at the data, it is surprising to see there were 44 species unidentified. With the geographic locations of the incident, bite wounds, and potential for investigation which would lead to bite radius, tooth indentation, and finding tooth fragments, it is unexpected to see this result being so high.
Figure 4. The shark data from 2015. Although nine unidentified species is still higher than expected, it is significantly lower than 2005. As expected, the range of shark species being recorded is much higher. It could be theorised that shark science is becoming more popular, and the knowledge of species has improved.
The learning: My charts and graphs have been exported in here as images, so it’s nice to say I feel comfortable with uploading images. I haven’t figured out how to change font colour, or make anything super fancy as I know my peers have. This is what I’m here to learn, I hope to understand what’s needed within R - for now, I’m just pleased I’ve been able to publish a file and make this somewhat nice to look at. Feeling very clueless and have decided to start putting images at the end to either show what I’ve learned, or how I’m feeling.
______________________________________________________________________________________________________