- Housekeeping
- Repetition Session 1
- Difference between females and males
I'm stuck at and the following code doesn't work for me:
Great video but the code might be outdated? Running the ggplot at , my Studio shows error "At least one layer must contain all variables used for facetting“”
> ggplot(data.combined[,], aes(x = Age, fill = Survived)) ++ facet_wrap(~Sex + Pclass) ++ geom_histogram(binwidth = 10) ++ ggtitle("Survival Rates Broken out by Sex, Pclass, and Age") ++ xlab("Age") ++ ylab("Total Count")Warning message:Removed 177 rows containing non-finite values (stat_bin).What code can be used to deal with the missing values in "Age"?
Dave, it somehow work. I was gettng the "aesthetics-must-either-be-length-one-or-the-same-length" while trying to plot ggplot(datacombine[,], aes(x = ticket.first.char, fill = Survived)) +geom_bar() +ggtitle("Survivability by ticket.first.char") +xlab("ticket.first.char") +ylab("Total Count") +ylim(0,350) +labs(fill = "Survived")
ggplot(data.combined[,], aes(x = Age, fill = Survived)) ++ facet_wrap(~Sex + Pclass) ++ geom_histogram(binwidth = 10) ++ xlab("Age") ++ ylab("Total Count")Warning message:Removed 177 rows containing non-finite values (stat_bin). why is this error occurring??
ggplot(data.combined[,], aes(x=Sex,fill=Survived))+facet_wrap("Pclass")geom_bar(stat="count")+xlab("Sex")+ylab("totalcount")+labs(fill="Survived")error: Cannot add ggproto objects together. Did you forget to add this object to a ggplot object?pls help
ggplot(data.combined[,], aes_(x = Sex, fill = survived)) +stat_count(width = 0,5) +facet_wrap(~Pclass) +ggtitle("Pclass") +xlab("Sex") +ylab("Total Count") +labs(fill = "survived")
you also faced same situation "summary(data.combined[,"Age"])" where n/a =177..
- Examining the age variable
Great channel and very helpful. I have a problem with a graph that you create at . R Studio keeps poping out a warning message:
at , you create the model of the age and sex distributions, but when I run the code, it gives me this warning message.
Pew
Hi David. Thank you for creating such high-quality content! One question... My histograms look very different from the ones you obtained (). I've even downloaded the data from your GitHub repo in order to obtain the same results, but I'm not able to obtain the same visualizations. Any idea why that is the case?
- Finding a proxy for age
Hi, hopefully someone can give some advice. At around , we're considering that young girls in 'Miss' travelling alone may have a different survival rate. When I try to pull out those variables using the code provided:
- Siblings, spouses, and parch variable
Thanks David for these videos and your explanation of approaching data analysis. You're awesome. One Q: at of this video, we're trying to plot a histogram of 'sibsp' after converting it into a factor. I just started using R Studio but in this version, a R Studio says "StatBin requires a continuous x variable: the x variable is discrete." In your video it seem to work but then it was like 4 years ago. What's your advise for this? btw, geom_bar didn't work either.
Miss is not a married woman, that means that SP in sibsp doesn't apply. Okay. But based on that, why can we logically infer that we are dealing with female children? What if I'm a miss, and has sibsp =1? This could indicate for example that I'm older sister traveling with younger brother/sister
- Feature engineering, family size
Hi David, thanks for the videos they are being very instructive. I would take issue with the inference that you are drawing from the data at "Large families are more rare than small families in the early 1900's" Large families were still quite popular in the 1900's I think the average family size in the 1900 was about 5 children, so I think that it may be more accurate to say that it was more unlikely for large families to travel, especially on a journey like this, than a small family. Large families tend to be more settled than a small family unless something happens to upset their situation e.g. redundancy.
Hi Dave! Thanks for adding me on Linkedin. I just wanted to briefly necro your video here and point out that at you're mentioning a decrease in survival area, but if you were to plot it as survival rate or % survival, the odds would look better. Sure, the # of observations go down and that may mess with your significance, but it is actually opposite of what you're saying.
