Quantcast
[ 3 / biz / cgl / ck / diy / fa / g / ic / jp / lit / sci / tg / vr / vt ] [ index / top / reports / report a bug ] [ 4plebs / archived.moe / rbt ]

/vt/ is now archived.Become a Patron!

/sci/ - Science & Math


View post   

[ Toggle deleted replies ]
File: 12 KB, 300x100, 34.png [View same] [iqdb] [saucenao] [google] [report]
11104896 No.11104896 [Reply] [Original] [archived.moe]

"1. Do data visualization and show some descriptive statistics for subsamples of the data set. Then generate your hypothesis (e.g. gender wage gap).


2. What is the difference in (predicted) wages between men and women with the same characteristics? How do you pin down your regression specification.
Talk about the identification, estimation and inference along the way of your analysis."

How the hell do I do this on Rstudio? How do I go on about it?

>> No.11104902 [DELETED] 

Maybe if you bothered reading some actual gender theory you'd know when feminists talk about the wage gap, they're not saying women are less than men with the same profession and experience (although that too exists, it's less than 5%).
The entire point is questioning is why is there an experience gap, why do professions that are traditionally female-dominated get paid less, and why are they female-dominated in the first place.

>> No.11104903

Maybe if you bothered reading some actual gender theory you'd know when feminists talk about the wage gap, they're not saying women are paid less than men with the same profession and experience (although that too exists, it's less than 5%).
The entire point is questioning is why is there an experience gap, why do professions that are traditionally female-dominated get paid less, and why are they female-dominated in the first place.

>> No.11104905

Go back through the tutorials you skipped.

>> No.11104915

>>11104905
mate I wish there were tutorials, can you direct me to one that works

>> No.11104990

>>11104915
This one https://stats.idre.ucla.edu/r/seminars/intro/ smashes through everything you need to know at a fairly brisk pace. This book https://r4ds.had.co.nz/ covers similar stuff but more in depth. Hang on to that stats.idre.ucla.edu url it's a good one.

R studio has four big panes - the document you're working on, the stuff you have in memory, the console which is basically a REPL that shares memory with your document and the documentation window. The documentation/figures window is a main pane for a reason. Type ?<command> in the console to bring up the docs in the docs window. Do this a lot.

>> No.11105084
File: 5 KB, 600x335, Screenshot_2.png [View same] [iqdb] [saucenao] [google] [report]
11105084

>>11104990
highly appreciated man, I managed to reach this stage with it

although the graph is handicapped and says nothing about the relation between gender and wage, anyone got ideas?

>> No.11105097

>>11104903
>they're not saying women are paid less than men with the same profession and experience
that's a blatant lie, they do stake most of their claims on the presumed fact that women are paid less compared to men for the same labor which is not supported by economic data on wages on first world world nations
>why do female-dominated professions get paid less
they're less valuable and the talent they recruit went through less schooling to earn their credentials. Nursing is less of an investment than medical school, financial management and quant fin are much more competitive than social work and hospital administration they also produce a larger amount of wealth for the people who invest in those industries.
>why are they female-dominated
genetically determined personality traits that predispose women towards more socially oriented, lower competition, and less intellectually demanding tasks like teaching, clerical work, healthcare.
>>11105084
>anyone got ideas
you're an idiot

>> No.11105103

>>11105097
I know man, it's just the course did not mention much, as the professor was on vacation and did not write the description. Therefore, I did not know what this class was really about. Now I am just stuck here not knowing what to do

>> No.11105108

>>11105084
Convert 'female' into a factor and draw boxplots instead. Hint: use lapply() to apply as.factor() to the values in the column.

>> No.11105190
File: 5 KB, 132x317, Screenshot_3.png [View same] [iqdb] [saucenao] [google] [report]
11105190

>>11105108
hey thanks for the idea, I followed it but how can I split a column into 2?

my head is going to bits, I have a female column where 1 is female and 0 is male, how can I split this column?

>> No.11105220

>>11105108
thank man i finally split it, ill make a boxplot now

>> No.11105243

>>11105190
You can split it into two tables like [math]\texttt{df.men<-df[df$female==0]}[/math] or something. if you're trying to split the column into one for female and one for male in the main table don't because you'll fuck up model fitting and you can't use that approach for most problems anyway so don't learn bad habits.

P.S. I'm going to bed so now you have to learn the most important skill in all of programming: finding questions on stackoverflow that apply to your problem

>> No.11105284

>>11105243
I'll go to bed soon, but I'll work a bit before it, but thanks for all the help and have a good night

>> No.11106874

>>11105084
Violin and boxplots (as >>11105108 said) can help with this.

>> No.11107082

>>11104896
Go to youtube.

Put into the search bar 'how does R work'

Watch one of the 1 hour vids that explain how R works. Obviously the exercises wants you to (1) learn how to make visualizations of descriptive statistics (2) test your ability to generate a null hypothesis with alternative hypothesis (3) do a statistical test (prob know the diff between non parametric and parametric) between variables, make sure they correlate and perform a regression containing features you chose.

The identification estimation and inference is all dependent on the variables you choose to do your regression on. But I think the assignment just wants you to separate male and female data first. And then draw a prediction line based on some variables (the same variables for both **) and see the differences.

Fucking undergrads having their first stat course around this time of the year....

>> No.11107094

>>11104896
There's a library called Swirl which will help you learn R within Rstudio which might be helpful.

>> No.11107094,3 [INTERNAL] 

Hi, thank you very much for the interesting post. If you are interested in such games or are interested in finding something new for yourself, you can visit http://onlinerealmoneyslots.com. Where you will find out at the end of this year about the most popular games.

>> No.11107094,4 [INTERNAL] 

I absolutely agree with you, the casino is a very entertaining thing. Especially slots, I like microgaming, for me it's the brightest and most themed slots https://spin4play.com/en/ besides, this provider has a demo version for all games

>>
Name (leave empty)
Comment (leave empty)
Name
E-mail
Subject
Comment
Password [?]Password used for file deletion.
Captcha
Action