Attendance unrequired today as there is a flood warning. Class resumes on Wednesday. The paper is due by midnight and submission is through email.
Sent in the report for the week. Class on Friday presumably. Today was another “not really required” class.
The report is due on Wednesday. I have asked Dr.Davis when specifically on Wednesday. I did not ask about the submission format; whether it should be over email or if he will create a place to submit on Canvas. Nothing has been graded thus far so I am also wondering if there is going to be a “mid semester grade” posted soon. I hope to pass in corrections, if need be, for this paper after he grades and hopefully get a better idea of my standing in the course up until this point.
Formatting of the paper is APA, Arial, 11pt font, single spaced. According to one of the resources pages (found here), it only needs to be three pages (excluding the title page and the optional appendix). I am revising and trying to meet these goals while also recognizing the paper should include sufficient detail without being too lengthy.
10/20/25 marks the completion of the first 7 weeks of this course. For the remaining 7 weeks, we are to look at new data if we no longer wish to continue with the data from The Washington Post. From the “Dataset for Linear Regression” page (found here), “6. Fish Market Dataset”, “10. Energy Efficiency Dataset”, and “13. Student Performance Dataset” peaked interest. Alternatively, “3. Movie Revenue and Rating Prediction” from the “Beginner Regression Projects and Datasets” section on this page.
The next 7 weeks will focus on one of the above.
10/13/15 Indigenous People’s Day
No class.
A week exactly til the report is due. This is unsettling. Uncertain of my starting point. I’ve yapped a lot but I didn’t have a main pinpoint question to focus on. Dr.Davis suggested looking at the stuff that’s here so far and pick a focus.
Nicole mentioned working together for the paper so I have just spent some time reading through her posts. I appreciate all her graphs and general organization of data. It feels easy to follow and I can see a way to bring our information together in a sense.
I think we should focus on mental health and fleeing. Nicole has already made some graphs related to the two and I had started looking at the fleeing status myself previously. I think our question should be something of the two presumably but need to think about it further.
I would like to focus on correlation between the “threat type”, “flee status”, and “armed with.”
I think writing a step by step of the RStudio program to be able to go back to would be good. I wrote pieces before, unsure if it was comprehensive steps. I thought I saved a file for it but I couldn’t find it again.
Upon opening RStudio and already forgot how the tables were made.
Latest.
Replication within the program has not been the easiest. I think this is amusing because it should be fairly straightforward to do so. Just the first step of getting the data open again posing a challenge is bothersome.
Copy and paste terms:
-data<-read.csv (‘link’)
-threat_type
-flee_status
-armed_with
The mid semester report is due in exactly two weeks from today. I made a reminder for today to get a progress check. As the paper is due on Wednesday the 22nd, I thought it would be good to review the requirements for the report. It feels similar to a lab report in the structural set up and the parts that are required.
I have emailed about Dr.Davis reading the “weekend work” posts as I wonder how to proceed from where I am at. I believe there is a possible correlation between the event, but unsure how to investigate this deeper.
I have decided to try and spend more time over this coming weekend to do more understanding of last week’s findings.
Here’s the initial questions, followed by top answers from the tables in the last post:
-Given the threat types; what is the majority of the population qualified as?
‘Shoot’ threat type was the highest at 2924, followed by ‘threat’ at 2674, and ‘point’ at 1890.
-Given the flee status; what is the predominate response?
‘Not’ at 5575 was the highest response and ‘car’ with 1634 was second.
-Given the types of things people were armed with; what is the most common weapon, if any?
6040 cases were marked as armed with a gun and the second most weapon was a knife at 1776 cases.
There was no initial question in regards to race. But 4659 were white and 2486 were Black people. While the number is higher for white, the size of the population getting effected was not the same so there is actually more Black people getting killed.
The highest categories were ‘shoot’, ‘not’, and ‘gun.’ From this alone, it sounds like people were not running, but armed with a gun and threatening to or ready to shoot. I think there is a positive correlation between these pieces of information.
It’s a Saturday, it’s also my niece’s birthday. As I would rather not be overstimulated and surrounded by people, I am in her room doing homework.
In the last post, we got started on the use of R and RStudio. The areas of focus are threat type, flee status, and armed with. I did also mention race, but this will need to worked with not primarily independently but scaling is necessary so I will likely need to revisit “after the fact.”
The initial breakdown was of the threat types. “Str” is for structure, this showed the types and how many in totality. “Unique types” organized the following responses into categories and how many of each. After creating and sorting a table, the following was given.
Threat_Type Table:
| shoot | threat | point | attack | move | undetermined | flee | ” “ | accident |
| 2924 | 2674 | 1890 | 1490 | 599 | 538 | 192 | 68 | 55 |
I would like to repeat the following process for the other 3 categories I intend to look into. Having a table created for each gives me a better starting point / sense of understanding what is happening in each category.
Flee_Status Table:
| not | car | ” “ | foot | other |
| 5575 | 1634 | 1591 | 1345 | 385 |
Armed_With Table:
Race Table:
Happy October! Today is October 1st. Mid semester report is due in 3 weeks and the semester is progressing steadily. I feel as though I have not made thorough progress as I don’t have a background in programming. Today’s class was focused on everyone sharing their questions and progress in answering said questions. When Erika shared, she said that she had a rather specific question and continually broke it down into smaller and more manageable questions. I like this approach as it can get into the “nitty gritty” of things while not being the most “complex” approach. I would like to apply this thought of breaking it into manageable chunks to my own analysis.
The areas that I am most interested in searching are the threat_type, flee_status, armed_with, and race categories. I believe these four have possibility to overlap and connect in ways that can be searched or analyzed deeper in my search. I’d like to use R and R Studio to create the connections and see the possible areas of overlap.
I think starting with a breakdown of each category would be good with the following questions:
-Given the threat types; what is the majority of the population qualified as?
-Given the flee status; what is the predominate response?
-Given the types of things people were armed with; what is the most common weapon, if any?
I don’t have a specific question pertaining to race because as mentioned in class that it should be scaled since the population of different races vary this will be a later question rather than a starting point.
On Friday, I got to really start using R and RStudio. I had both downloaded, but had not opened RStudio so I was unable to get the data open initially. I really appreciate the assistance and the start to getting into breaking down the data.
Another Monday! Today in class we got to see some of what people have been working on and finding. Nicole presented about her findings.
Wednesday’s class was supposed to be focused on R and RStudio. Using the introduction to R page we were supposed to try playing around in the program.
Today starts with spurious correlations. We are looking at various graphs that have two variables that are seemingly entirely unrelated, but their data appears similarly within a single graph. Tyler Vigen spurious correlation page uses AI a lot to write papers based off the data it presents.
Starting off with a webpage update since not everyone had their latest post showing first. Then onto the “Software application notes” from page 18. It has an introduction for various programs that we can use to look at and analyze data in this course. “The easiest way to learn is to start by doing it.”
Wednesday’s class we were supposed to start using Mathematica, but a lot of the computers don’t have it working because the wifi or other issues. But I was able to download it to my laptop so I will be trying to get things started on my end. I have also downloaded R and R Studio to my laptop as well. I believe one program was more user friendly while the other had more capabilities. As I have never used either before, I am going in with no prior knowledge for either.
The plan for Friday’s class was to use Mathematica to go over Samuel’s question about body cams. As we currently don’t have access because the student subscription has expired, the plan has been adjusted. Looking at the data from the table and doing a T-test to compare the means of the data. I would like to start using R and R studio to try and see its capability in understanding the data better.