emmaghl
commented
2 years ago Thank you for your feedback!! Will definitely incorporate the formatting you mentioned for links!
Open callumstewart742 opened 2 years ago
emmaghl
commented
2 years ago Thank you for your feedback!! Will definitely incorporate the formatting you mentioned for links!
Describe the goal of the project.
The goal is to examine how chances of a passenger surviving the Titanic disaster relate to factors such as sex, age and socio-economic status.
The data set used was accessed via Awesome Public Datasets on Github and they believe the data was collected by a British Government commission that interviewed witnesses following the disaster in 1912. The data frame has 891 observations and 12 variables, including a name and unique ID number for each passenger as well as data such as their age, sex and class.
They outline 3 hypotheses and their plans to investigate them.
1: They hypothesise that the survival rate was higher for men than women and plan to use a segmented bar plot by sex and survival status to test this. They also include some summary statistics which suggest their initial hypothesis was correct.
2: They hypothesise that the proportion of survivors was higher among the young than the old and plan to investigate this with a histogram of age, filled with survival status. They plan to use the mutate function to create 10-year age ranges in their data frame, as this will allow for easier analysis. Again, they use summary statistics which suggest their initial hypothesis was correct.
3: Lastly, they hypothesise that a higher class passenger would have better odds of survival. They suggest they could again use a bar plot to visualize this relationship, but also that they could use a linear model of fare price vs survival rate.
They ultimately plan to use their visualizations to recreate a historical data visualization of the Titanic disaster made at the time by G.Bron.
Their proposal is clear and easy to follow.
The visualizations they already have are clear and informative.
I would suggest that their data set is ideal for building a model, which could incorporate some or all of the predictor variables mentioned above and predict the chances of survival based on various factors. One of their open issues suggests this is planned, which I think would add a lot to their project.
As far as minor improvements: Their narrative would be more readable if broken up into smaller sections and not single paragraphs. The introduction has multiple links in it, these would look cleaner with
[link text](https://example.com)formatting.I'm most interested in seeing their finished recreation of G. Bron's historic visualization, which I think is a unique and very good idea.
No problems with code organization. Possibly remove the unneeded README.html in the data folder, which is created when you preview the README.md file.
Nothing else to add, good luck with your project!