Corrections to final project for use as sample

rpseely commented 4 months ago

Tracking Corrections to Final Project

As instructed by Professor Buzard, I will be making some edits to the final report write-up and the do-file for potential use as a sample project in future sections of ECN 310. I will be using the critique from the grading rubric used by Prof. Buzard to guide my editing of the project.

kbuzard commented 1 month ago

@rpseely On the quote about Hellerstein and Morrill (2011), I don't see the connection between their hypothesis and the evidence you refer to. Your evidence can't say anything about the mechanisms, can it? I think it's fine to say that you can't address this question.

Relatedly, I don't remember if you have more female or male nepobabies. If there were a lot more women than men nepobabies (relative to the total number in the sample), you could at least nod to the risk aversion hypothesis.

There is a portion of writing in the data section that addresses how we determined to subset the data in terms of age, specifically it includes notes on a logistic regression and chi-square analysis. My thought in putting that in the data section was to explain why I chose to only focus on the data from observations in which the respondent was younger than 30. Here is my idea: copy and paste the section you highlighted somewhere into the beginning of the results section. Then, add a sentence in the data section where that text was and say something like "we choose to focus on adults younger than 30 and you can read all about why in \ref{sec:result}" so that one can jump to the explanation of why we made that cutoff. Would that be more appropriate?

I think it's okay to either leave it where it is or move it to results. If you leave it where it is, just strengthen the language around it. It just read like a bunch of results to me--I didn't take away from that the idea that you were including these details in order to justify excluding the people who were older in the results section.

rpseely commented 4 weeks ago

Thursday, June 6, 2024

Lit Review Edits

On the quote about Hellerstein and Morrill (2011), I don't see the connection between their hypothesis and the evidence you refer to. Your evidence can't say anything about the mechanisms, can it? I think it's fine to say that you can't address this question.

Relatedly, I don't remember if you have more female or male nepobabies. If there were a lot more women than men nepobabies (relative to the total number in the sample), you could at least nod to the risk aversion hypothesis.

We have more male nepobabies, but I think it is still fair to assume risk aversion is a factor in becoming a nepobaby. I changed the language to reflect that the idea that we take from Hellerstein and Morrill (2011) is the risk-aversion hypothesis. Then that explains part of the reasoning of a) why nepobabies are common and b) why we thought they might be more prevalent during higher unemployment times. I changed it to this...

Hellerstein and Morrill (2011) describes women as particularly risk-adverse to explain why they may follow into their parents' professions, specifically their father’s occupation to prevent the risk of being jobless after their graduation. This finding, of young adults with a risk-averse quality following into a parent's profession, provides evidence for why nepotistic parent-child relations are formed and why they are common, and why we hypothesized that nepobabies would be more prevalent during higher unemployment times. Our finding, that approximately ten percent of all young adult workers surveyed are defined as nepobabies, demonstrates that a large portion of the workforce utilizes their familial relationships to establish careers and does provide evidence of this potentially risk-averse behavior.

Data Section Edits

For the age cutoff explanation, I decided to keep it in the data section and I added the bolded parts for clarification:

To determine a cutoff age for what we define as a "young adult," we used a logistic regression model, where hiring age is a predictor of nepobaby status. There is a 3.3 percent decrease in the likelihood of being a nepobaby associated with being hired each year older. Likewise, a chi-square analysis, testing the association between nepobaby status and age of hiring as 29 and younger or 30 and older shows a significant association at a 99.9 significance level. This age cutoff keeps the assumption of entering the job market and looking to parents for support as reasonable while simultaneously keeping the number of observations as high as possible.

Reproducibility Package

@kbuzard Wow! It is super cool that just adding ".md" to the end of that file made it into (what I assume is) GitHub-flavored markdown. Anyway, I added a good deal of information about the documentation for the sensitivity analysis and reformatted some of the reproducibility package. Tomorrow I will add a data appendix to the .tex file, and then as a last step I am going to clean up the "code" page of the nepobabies repo with some folders.

rpseely commented 4 weeks ago

I have done some work create a data appendix, which took me 30 mins. I will come back to this later and fix the issues in compiling and organization. So far I have only put a few explainers and some hyperlinks. I also got the hyperlinks to stand out using some cool LaTeX code with help from ChatGPT. Previously, it was not apparent in any way that they were hyperlinks which made it pointless...how would someone reading it know to click on them?!

kbuzard commented 4 weeks ago

@rpseely I think both paragraphs are great! And I'll look forward to your data appendix!

rpseely commented 3 weeks ago

Friday, June 6, 2024

Data Appendix

The data appendix is created! And probably more importantly, the files, github repository, and reproducibility package are all much more organized than they were before. Anyway, the data appendix is two short paragraphs describing where to find all of our code, data, and documentation with hyperlinks to the GitHub repository and OneDrive. @kbuzard I imagine that you would like me to do a little shaping up on the data appendix. If you can point me in the general direction or to a resource that offers a good example, I will absolutely do that! It should be fairly quick now that everything is organized in the right place. I uploaded a new draft if you would like to take a look!

Organization

I spent a good chunk of time reorganizing the GitHub repository. Everything is now in a folder and not just spread out all over the place. Check the code page of this repo to bask in its well-filed glory!

nepobabies.do file

I added in some comments to the code where it was lacking explanation. I also deleted lines of code that was old and not being used in the analysis, mostly old two-sample t-tests that were improper, i.e. they were trying to measure what a chi-square test is intended to. Additionally, I updated the nepobabies.log file which is now simply the code and results of running the nepobabies.do file

Small change to results

While reviewing the code and the log file, I noticed that the result for the two-sample t-test for average income between nepobabies and non-nepobabies was worth noting in the report (p = 0.0866). Similar to the t-test for years of education, I noted that the p-value offered some weak evidence and that it was close to conventional levels, and that future study with a larger n could examine this more.

OneDrive

I asked Rachel, who owns the OneDrive folder, if she could give view access to anyone with the link to the OneDrive folder (if someone wants to access it from the data appendix). She is not able to grant access to the folder to anyone with the link unless she gives them edit access. I cannot imagine that anyone would want to mess with the folder containing intermediate-step data for the economics research project of two undergraduate students, however, we would prefer not to grant edit access to anyone with the link, so we are seeing what we can do there.

kbuzard commented 3 weeks ago

@rpseely I just read over the report, and it looks great! Please go ahead and finalize it.

I'm going through the reproducibility package now. I'll keep digging, but I wanted to let you know that the link in the readme in the Final Report folder has a broken link: the one to the reproducibility package doesn't work. That doesn't affect any of the links in the Data Appendix, so you can finalize that and then fix this link.

kbuzard commented 3 weeks ago

@rpseely NEVERMIND. The link to the .do file in the Data Appendix gets a "file not found" message, as does the link to the .log file.

It would also be good to make the data appendix and actual appendix section so it shows up in the table of contents. It's super simple. See this explainer.

Then, please drop a reference to the data appendix into the data section somewhere (it could be in a footnote if that makes the most sense). You can hyperlink to the section by doing something like the following with the data appendix setup:

\appendix
\section{How I became inspired}
\label{sec:appendix}

And then use the reference in a sentence, like "The \ref{sec:appendix} provides details of the data acquisition, manipulation and analysis." (or it might have to be some variation on this depending on how things are set up--the \ref statement may print out the section number instead of the name "Data Appendix." Bottom line: the \ref command will insert a hyperlink, and you can manipulate what it looks like, for instance, with the nameref package (see first answer here).

rpseely commented 3 weeks ago

[x] Fix data appendix links
[x] Make data appendix an actual section
[x] Make references an actual section
[x] put hyper link and short sentence in data section referencing the appendix
[x] Check all links in README, reproducibility package, and appendix

rpseely commented 3 weeks ago

Tuesday, June 11, 2024

Fixes

I fixed all the links in the README (minus the final draft link), reproducibility package, and appendix.

Appendix and Sections

I added the code from the website you suggested above and I set up a hyperlink in the data section so that it can be accessed from there.
At first I tried playing around with a different option that also included the references section as one of the numbered sections, but it kept producing a numbered section header and then another un-numbered section section header right below it and I could not resolve that issue (I presume it has something to do with reading in the .bib file with the \bibliographystyle{apalike} command).

@kbuzard I attached a file below so you can see what it looks like. The only thing I am not 100% about is how the references section is different from the appendix and the rest of the sections. NepoBabies.6.12.24.pdf

kbuzard commented 3 weeks ago

@rpseely I think the section stuff looks fine for both the references and the appendix! Don't worry about pushing it any further. There's probably a way to make it work the way you were trying (which I agree would be great), but I see this all the time in papers and people are used to looking for the references in the right place so not a big deal.

All the links in the data appendix work! I don't think it needs any further revisions.

For the OneDrive issue, the most elegant way around it would be to provide information about where the data is downloaded from on the GSS website. Then someone could download that big file themselves.

Then, have GSSclean_noRDs.dta in your Data folder on OneDrive, and everything can flow from there.

For the purposes of putting this on my website, I think it's good to go as soon as you upload the .tex and .pdf and make sure that the right version is linked on the main readme.

Two more things I suggest:

Clean up the branches on your repo. When I click on the final report on the main readme, I think it's taking me to a different branch b/c the files aren't organized.
Add the location of the report at the bottom of the reproducibility readme (with a note that it was written on Overleaf but that the final .tex file is available on your repo as well).

rpseely commented 3 weeks ago

@kbuzard I added a comment in the appendix on accessing the original GSS dataset on their website and added a hyperlink. I uploaded the .pdf and .tex for the final report and they are in the final report folder on this repo.

rpseely commented 2 weeks ago

Wednesday, June 19, 2024

I have been looking for the gss7222_r1 release with little success. On my first keyword search on google, there was only website match that came up, which was the link to the Nepobabies project on Prof. Buzard's website, so that was funny. I submitted a contact form on the GSS website asking for a link to the original data release that we used. I tried uploading the file directly to GitHub but as we discussed it was too big.

@kbuzard I went on your website and noticed that the hyperlink to the accidents project leads to the PDF of the nepobabies project.

kbuzard commented 2 weeks ago

@rpseely Another option: you could try downloading the version that is currently on their website and see if you get the same results. You could even start off by using something like this to compare the two datasets. If your results are the same, then you can just link to the current version of the dataset.

rpseely commented 2 weeks ago

@rpseely Another option: you could try downloading the version that is currently on their website and see if you get the same results. You could even start off by using something like this to compare the two datasets. If your results are the same, then you can just link to the current version of the dataset.

@kbuzard Unfortunately I am not able to open the r1 or r3 versions of the GSS data in Stata on the RDS because the files are too big. I tried opening them in R but it stopped responding (this happened just now and when I tried it last week).

kbuzard commented 2 weeks ago

Unfortunately I am not able to open the r1 or r3 versions of the GSS data in Stata on the RDS because the files are too big. I tried opening them in R but it stopped responding (this happened just now and when I tried it last week).

I just emailed the research computing people in Maxwell about this. I'm sure they'll have a solution. I'm going to be in the car for the rest of the afternoon so won't be able to respond to them. So, if they ask a question that you can answer, please go ahead and reply.

rpseely commented 1 week ago

@kbuzard I have phenomenal Nepobabies news! The results from the gss7222_r3 are the exact same as those from the gss7222_r1! Once I learned about unzipping folders/files, I was able to get the r3 results. I then made these edits on overleaf and I have attached a PDF here. If you like the changes in the appendix, I will go ahead and update the final report links on the GitHub repo. NepoBabies.6.25.24.pdf

kbuzard commented 1 week ago

@kbuzard I have phenomenal Nepobabies news! The results from the gss7222_r3 are the exact same as those from the gss7222_r1! Once I learned about unzipping folders/files, I was able to get the r3 results. I then made these edits on overleaf and I have attached a PDF here. If you like the changes in the appendix, I will go ahead and update the final report links on the GitHub repo. NepoBabies.6.25.24.pdf

@rpseely The updated PDF looks good! Will you also adjust the first item in Final Report/ReproducibilityPackage.md to reflect the data coming from the GSS website?

rpseely commented 1 week ago

@rpseely The updated PDF looks good! Will you also adjust the first item in Final Report/ReproducibilityPackage.md to reflect the data coming from the GSS website?

@kbuzard I have updated all the reproducibility package, the file organization with the updated PDF and TEX files, and all of the links.

kbuzard commented 18 minutes ago

Great job! I'm sending it to Professor Zhu now!

ecn310 / course-project-nepobabies