Daily: Stream of Consciousness

[ltsippel]

10/25/23

Made this. Started to add links to diop Wiki.

[abigailmondin]

10/25/23

I updated the Wiki with dates, descriptions, and links of some of the things we've done so far.

[sophiehaber]

10/31/2023

I updated the data acquisition item & will bring it in for review during office hours to allow time for changes to be made

[abigailmondin]

10/31/23

I uploaded the do files for the informant and respondent data from the HCAP 2016 to the main page of our group repo. I also added the files to the wiki under the page for our dataset.

[abigailmondin]

11/1/23

I updated the wiki to better organize our master documentation (if anyone doesn't like the new setup we can always change it).

[ltsippel]

11/1/23

I uploaded a link to hypothesis formation and causal diagrams to our Issue.

[xorabear]

11/1/23

i did the literature review links and updated the master wiki on the wiki page

[kbuzard]

@ecn310/diop Here is my feedback on your documentation:

• I love what you've done with the master documentation on the wiki!
  • You might move the "Stream of Consciousness" line to the previous section
• In general, this is really well organized and is everything I was hoping for!

[ltsippel]

11/8/23

Finalized my Descriptive Statistics Exercise and we as a group are going to figure out our best next steps today!

[abigailmondin]

11/8/23

I unzipped the files for the 2016 RAND HRS dataset, which seems to have valuable data on education as well as dementia and cognition. The do file I created for the dataset is on the main page of our group repo and is linked in the wiki. The dataset (.dta file) is in the OneDrive.

[abigailmondin]

11/8/23

Just so everyone knows I tried to add the new dataset (2016 RAND HRS) onto Github and the file size seems to be too big to do that. We can always double check with Professor Buzard or Dylan next class, but for now the dataset is in the OneDrive.

[abigailmondin]

11/10/23

I added the codebook for the 2016 RAND HRS dataset to wiki and as a group we spent a majority of class combing through the data for the variables we will actually need. Sophie created a discussion for the variables we think we will need/use, and I've linked this discussion on the wiki.

[sophiehaber]

11/14-15

I finished reviewing the entire codebook for our project and began going into our data on STATA to utilize the keep function to keep the variables we deem important out of 7,904

[ltsippel]

11/15/23

• Specifically looked at PD101, PD102, PD501, PD502 all rate memory (self described memory) before and after the cognitive tests.
• *Note: only a partial interview for PD501 and PD502, but will try and figure it out accordingly.

[ltsippel]

11/15/23

-Plan for me is to make up graphs for Rate memory data and share with group.

[xorabear]

11/8/23

• I looked for specific variables that relate to our hypothesis and I had added 2 new links under data adjacent not all of them

[abigailmondin]

11/16/23

I added links to each of the descriptive statistics we created to the wiki master documentation. I added a link for Luke's, Sophie's, and my own, but I could not find a post made by Ixora.

[abigailmondin]

11/17/23

I created an issue for us to upload any and all charts or graphs that we create, I also linked this issue to the wiki master documentation. In class today, Professor Buzard went over LaTeX and the formatting of the final pdf that we're working towards. Luke made an Overleaf account, and I believe he will be the one taking the lead on that as we move forward.

[ltsippel]

11/17/23

I made an overleaf account and started linking up sections of our report to the introduction. I also made a pie graph for the variable pd101 with the intention of comparing it to pd501. Since each data has different amount of people that's why I chose a pie graph to represent the data via percentages.

[xorabear]

11/29/23

I uploaded my descriptive statistics work in the exercise reposition and inserted the link along with the description to the master wiki.

[ltsippel]

11/29/23

Entire Group figures out variables

• Rate memory pd101 `pd501'
• The Chance that they'll develop dementia pv264 V.S. Years of Education pz216
• Highest level of education `pb014'
• Ever had dementia pc273
• Ever had Alzheimer's pc272
• If they've earned high school diploma `pb016' 'pb015'
• Forgetful during daily activities `pv009'
• Get lost in familiar places pd554

[abigailmondin]

11/30/23

I created a bar graph using Stata for the variables pz262 (pw dementia) and pz216 (r years of education). This, along with a description and code used to make the bar graph, are located under the Charts and Graphs issue.

Update: I also created a frequency histogram that shows pz216 (r years of education). This is also located in the Charts and Graphs issue and includes a description and code.

[abigailmondin]

11/30/23

I edited the do file to reflect some of the data exploration I've been doing. Right now the updated file is in the OneDrive because when I go into GitHub desktop to commit the changes to our repo the do file isn't showing up in the changes area.

@kbuzard @eldreddyl Could one of you help me with this?

[kbuzard]

@abigailmondin The first thing I'd check: where is the version of the do-file that you've edited? As in, what folder is it in?

Then, in Github Desktop, making sure you're in your project repo, chose "Show in Explorer" from the Repository menu. Is the folder it brings up the same one that your edited do-file lives in?

[ltsippel]

12/1/23

Sophie: Histogram between whether or not dementia and college degree and high school/GED Abigail: Lost in familiar places and Forgetful during daily activities/Dementia and over education Luke: Tables, r^2 between education dementia and Alzheimer's Ixora: Years of education and highest education with Alzheimer's

[kbuzard]

@ecn310/diop There's no mention here of making your plan on Wednesday, or where to you're going to be updating it. I don't see it on your wiki either.

• There are a lot of outstanding tasks on that plan; they should be converted into task issues in this repo (maybe not a separate issue for each one; there may be ways to group them together that make sense and prevent an overwhelming number of open issues).

It would be better if some of your commit messages were more informative (this is going forward; it's a pain to edit commit messages from the past). For example, you have three commits all called "Update Sophie 2016_RAND_HRS do file.do" ; there's no information about what was updated each time.

The commit message when Abigail added the dataset is much clearer. However, I don't see where the do file is that created that file. You should have task documentation around creating this new file, where it's being stored, why you made this change.

[abigailmondin]

However, I don't see where the do file is that created that file.

12/2/23

I updated our running do file to include the code that created that file, as well as code to opening the new dataset with the reduced number of variables.

[sophiehaber]

12/5/23

I created the two tables rather than the graphs, as Prof. Buzard advised and used cross tabulation to ensure all values are present. I left the code for the graph in the do-file for now just in case it can be amended & used to help someone else before we have to finish everything for tomorrow. I compared the values of wether or not someone has dementia with if they have a high school diploma/GED or a college degree

[abigailmondin]

12/5/23

I created four bar graphs each posted under the Charts and Graphs issue:

1. pd554 over pz216 - get lost in familiar places over r years of education
2. pd554 over pc273 - get lost in familiar places over ever had dementia
3. pv009 over pz216 - forgetful during daily activities over r years of education
4. pv009 over pc273 - forgetful during daily activities over ever had dementia

I've updated the running do file to contain the code used to create each of the bar graphs and each post under the issue has the code that correlates with the bar graph it created.

[ltsippel]

12/6/23

Sophie and I figured out how to screenshot a table on Stata which enabled us to upload it on overleaf.

[abigailmondin]

12/6/23

I created an issue for the documentation plan, this is not 100% finished so feel free to make adjustments.

[xorabear]

12/6/23

I drafted the final literature review and emailed it to one of my peers for feedback and I'm going to incorporate the feedback later today

[kbuzard]

Sophie and I figured out how to screenshot a table on Stata which enabled us to upload it on overleaf.

You should either make the table natively in Overleaf by manually create the table, or use something like outreg to create it. Using a screenshot from Stata is just NOT done.

[abigailmondin]

12/7/23 - 12/8/23

@kbuzard @xorabear @ltsippel @sophiehaber I've been working on the hypothesis/theory section of our final report, and to be perfectly honest I'm not totally sure how this should be written or what should be included. As I've been writing, I've noticed some overlap between this section and the literature review section which I know Ixora worked on. What I've written so far is a jumping off point. Feedback and/or recommendations would be greatly appreciated. Let me know if the document doesn't open for anyone. ECN 310 - hypothesis & theory finalization.docx

[abigailmondin]

12/8/23

These are notes for myself to remember:

• For the frequency histogram, add the numbers for all the values across the bottom
• For the education graphs, make the variables proportional in graph bar code

[abigailmondin]

12/9/23

@kbuzard @xorabear @ltsippel @sophiehaber I posted this on here yesterday in an above post, but I just wanted to put this in here again in case anyone missed it. I would really appreciate feedback on what I've written for the hypothesis/theory section for the final report. Like I mentioned in the previous post about this, I've noticed some overlap between this section and the literature review that Ixora wrote. ECN 310 - hypothesis & theory finalization.docx

[kbuzard]

@ecn310/diop Here are the things I didn't say during your presentation since we were running short on time:

1. The labeling/spacing on the horizontal axis is weird on the first graph--makes it look like some years have no observations.
2. Don't include variable names in the titles of your plots; those variable names are not helpful at all for a reader. Your documentation should clearly identify the variable names, but unless you rename them so they have some meaning, I would leave them out of your report completely and instead use some kind of descriptive label.
3. Make the labeling of the horizontal axis more intuitive for the graphs that right now just have codes for different responses.

[kbuzard]

I posted this on here yesterday in an above https://github.com/ecn310/course-project-diop/issues/5#issuecomment-1846697050, but I just wanted to put this in here again in case anyone missed it. I would really appreciate feedback on what I've written for the hypothesis/theory section for the final report. Like I mentioned in the previous post about this, I've noticed some overlap between this section and the literature review that Ixora wrote.

@abigailmondin There's no problem with having some overlap between the lit review and the theory section, as long as you cite everything and everything fits together. You may be able to streamline a little, in that you can refer back from the theory section to what was shared in the literature review. You can say things like "We follow XXX in focusing on ...."

It would be better to paraphrase the direct quotations you have; you can turn it around to say something like

As Banks and Mazzonna (2012) point out, it is difficult to clearly identify the specific mechanism...

Also, your last plus sign should be a negative sign, shouldn't it?

You don't have to include the causal diagram, but if you do, make sure to label it as a figure and discuss it in the text. Right now, that's the main thing that's missing: a verbal description of the theory and a summary of your hypothesis (if you have a verbal description, then you don't necessarily need to have the diagram--that's just a matter of what you prefer). You can move the specifics about the variables you'll use in the data section, and then connect the variables to the hypothesis in the analysis section.

[kbuzard]

@ecn310/diop I was playing around with your data a little bit, and I think a reason that you're not seeing some of the strong results you'd hoped for is that for some of the variables you're looking at, you just don't have data for that many people (the "gets lost" variable versus the dementia diagnosis variable, for instance).

• This points to something I don't think I emphasized in class: you may need to include some of the summary stats on individual variables in your report; this usually goes in the data section. Readers will want to know how many observations you have for each variable, and what percentage of people answered "yes" for each question, what the age distribution of respondents is, what the average level of education is, and maybe more). This usually all goes in one table together; you might need to recode some of your variables (the mean of all those 1,5,8,9's doesn't actually have meaning; what is important is the number of people who actually answered, and the proportion that answered yes).

[abigailmondin]

12/10/23

I wrote a new documentation plan based on the feedback Professor Buzard provided, this is located under the issues tab of the repo. Additionally, I created an issue for the hypothesis/theory section of the final report, which I assigned to myself. Under this issue, I revised the hypothesis/theory section based on Professor Buzards feedback and tagged her for additional feedback on the revisions.

[kbuzard]

@ecn310/diop Your updated documentation plan looks good. @abigailmondin I'll get back to you on the revision to the theory section as soon as I can (tonight if possible, but I have to go to a meeting soon).

I'm going to give you a "2" for the documentation check, but I'm not convinced that everything is getting recorded here (for example, someone clearly made the presentation, but I don't see anything about it on this issue except the feedback I provided). With everything that needs to get pulled together in the next 9 days, you're going to need to make sure to communicate clearly and keep updating your documentation and finals submission plans to make sure everything stays on track.

I'm also a little worried that some of the upcoming tasks don't show up in your list of issues.

• For instance, the draft of the analysis section is due in less than 48 hours, but I don't see an issue for it. See guidelines here.

[abigailmondin]

12/10/23

I created a new issue for the drafting and finalization of the analysis section of the final report.

• According to our overall plan for the final report, @sophiehaber will be writing this section
• @ltsippel has the Overleaf account, so I assume they will collaborate to get the file compiled in Overleaf for the assignment submission
• Important: This is due on Tuesday, Dec. 12 at 8am

Additionally, I updated the plan for final submission to include more detailed notes about certain tasks, and updated some target dates.

Update from Dec. 7-8: Our presentation on Dec. 8 was completed before class on the 8th. All of us contributed to the presentation in terms of creation and actually presenting in class.

[sophiehaber]

12/10

• I created the data replication README and @abigailmondin will input the unzipping information since she handled that for our group
• I worked on drafting the analysis section but am still waiting on @xorabear to upload/send the graphs so I can include it in the analysis
• The whole group can look over the analysis section prior to submission if preferred

[abigailmondin]

12/11/23

I updated the frequency histogram for "r years of education" to includes values for all the bars along the x-axis. I also added the steps for unzipping the dataset to the data acquisition issue that @sophiehaber created.

[sophiehaber]

12/11

I completed the analysis and data sections of our report and finished uploading all graphs and tables, as well as labeling them all. I will also continue to work on the README document and remove the unnecessary information that I had previously added.

[abigailmondin]

12/11/23 - 12/12/23

I tried to make edits to one of the graphs (using the graph editor) based on the feedback given to us by Professor Buzard after our presentation in class. I made the changes to the ever had dementia/gets lost in familiar places graph. I'm not 100% sure these are the changes she's was imagining, so I tagged her in the post and will hopefully receive feedback soon.

[kbuzard]

I didn't get any notification. If I missed something, I'm sorry. Can you point me to where you want me to look?

If you edited a previous post instead of leaving a new comment, adding a tag doesn't trigger a notification.

[abigailmondin]

@kbuzard I did add the tag in an edited post, so I guess that would be why you didn't receive any notification. I was looking for feedback on this post under the Charts and Graphs issue. This post now has the original graph I created, that was included in our presentation, and the updated graph that I made last night. Before I go ahead and make similar changes to the other graphs, I just want to make sure this is what you were looking for.

[abigailmondin]

12/12/23

I unzipped the folder that Sophie had uploaded for our analysis section of the final report. The unzipped folder is called "DIOP.Analysis.Draft" and located on the main page of repo.

[kbuzard]

@kbuzard I did add the tag in an edited post, so I guess that would be why you didn't receive any notification. I was looking for feedback on https://github.com/ecn310/course-project-diop/issues/14#issuecomment-1842030937 under the Charts and Graphs issue. This post now has the original graph I created, that was included in our presentation, and the updated graph that I made last night. Before I go ahead and make similar changes to the other graphs, I just want to make sure this is what you were looking for.

@abigailmondin Yes! The updated chart looks great! I think you should be able to get the chart editor to give you the underlying code so you can see how to apply it to other charts. You need to have the code in your do-file anyway to make the chart reproducible.

[abigailmondin]

12/12/23

I updated all four of the graphs that I created based on Professor Buzards feedback given to us after our presentation. I posted the updates on this post in the Charts and Graphs issue of our repo. These graphs are still not 100% as I can't figure out how to code certain aspects, so I've tagged Professor Buzard and Dylan in the post that I made with details about what I need help with to hopefully get this figured out. Additionally, I updated the do file with the updated code to produce these graphs.