Last Friday I had a great time presenting on work constructing synthetic cohorts and using linked administrative data using the UKHLS.
See below for a snapshot of those results or follow the link to my website to download the slides and find out more here lnkd.in/eKanfR2a
Thank you Mike!
Thanks Dave!
As of yesterday, I have officially passed my Viva! Very surreal that the PhD journey has come to an end.
For those interested in Handling Missing Data Methods, see via the link some SGSSS training materials that cover a wide range of methods and materials in both R and Stata format:
scott0atley.github.io/Scott0atley/...
#sociology #socsky #quant
Fantastic opportunity to network and present with some fantastic quant people!
Very happy to announce that I will be joining the ELSA team as a Research Associate over at the University of Manchester. Super lucky to get the chance to work with some great people!
#academicsky #sociology
I've finally been able to set up a Jupyter Notebook that can use Stata, R, Python, and Mplus all at the same time. The base Python kernel means that for Mplus to run, it has to run the Mplusautomation package through R which is then run through Python. The time delay is surprisingly fine so far!
As of a few days ago I have officially (officially, officially thanks admin) submitted my PhD.
This is my formal apology tweet to all the emails that have gone unanswered over the past few months. An apology email tour is planned.
Now on to submitting this growing pile of paper manuscripts!
It is also (more likely) their browser, switching to an alternative usually does the trick.
Just looking at the backend now. Appears to be a user spike issue. I’d suggest getting them to pair up for the time being. This has been reported before, we are waiting on some funding to increase capacity.
That’s amazing to hear! Do let me know if your class has any comments or run into any issues.
Introducing: socstrat
A new command available on ssc for stata users.
This command cleans and generates a full analytical 8-category NS-SEC based on either SOC 90, 00, 10, or 20 as well as RGSC on SOC 90, 00, or 10.
Hopefully this makes life a little easier for people.
#academicsky #stata
Bit of a Hail Mary ask but does anyone have a link to the SOC 2020 derivation of NS-SEC that includes the 1.1 & 1.2 breakdown of NS-SEC 1?
The official documentation appears to have abandoned this practice from SOC 90-2010. Or are you now forced to code into operational classes -> analytical now?
NS-SEC Stata program coming circa 2025 possibly...
... (though something similar could exist and I am unaware?). You are of course right, doing something like this is ultimately a trade-off between measurement error and interpretability.
It certainly seems better than studies that stick SA clusters into a MLM and calls it a day.
Promising in the sense that this new method appears to provide the ability to make such a tradeoff. The bootstrapping provides a relative threshold mechanic to assess the relative cluster stability. I think this is what has been missing from models that use SA clusters...
Just reading this article on the use of clusters from SA in modelling and finding it really interesting. I've always avoided (and been told by others) to avoid doing this because of the data reduction issues involved, this method seems... promising?
link.springer.com/article/10.1...
Update for the update:
1) Dynamic Button selection improves user quality
2) tied server load to button selection to speed up results
3) slimmed down latent trait code so crashes should now be a thing of the past
please do comment/message any more suggestions. Code available in about us page!
The Class Calculator has been updated!
Check out our new and improved teaching tool here:
eosscc.shinyapps.io/Edinburgh_Cl...
Comes with 7 social class schemas to interact with.
#academicsky #socsky
There are attendance monitoring systems in place. Though I have no idea why any member of academic staff would use them. I am not a immigration enforcement officer and realistically that is why attendance is recorded.
Thank you for the suggestion. Off the top of my head the differences in R^2 are very small, but I will re-run some analysis to test for sure.
My general conclusion also - though I do feel there is some interesting effect size differences going on when duplicating the 2007 paper's model on auth-lib scales?
If all are measuring the same thing, I suppose that opens the "Is it measuring status or social distance?" conversation once more.
Had some really great feedback on this talk yesterday. I'm still not entirely convinced myself here, but I think it is a useful exercise nevertheless.
Things to consider:
Gender sensitivity
How to account for women falling out of work?
Negative loading of social intercourse
...
Linear Correlation of Measures
High correlations between measures. The correlation between my measure and C-G is equivalent to the correlation of the C-G measure and proportion of A'levels in the 2004 paper.
I just gave a talk on social status to the @largesocialsurveys.bsky.social group. To check out the presentation head over the my website and download it:
scott0atley.github.io/Scott0atley/...
If you want to follow along with tomorrows training “Survey and Questionnaire Construction: the Shiny Package in R” that I’m hosting @edinburghup.bsky.social find the training materials on my website:
scott0atley.github.io/Scott0atley/...
#academicsky #socsky #Rtraining
Wee jaunt to University of Stirling for a super useful and informative weights workshop delivered by Pierre Walthery from the UK Data Service. Cheers @kateohara.bsky.social for organising through the Large Social Surveys group (time to make a blue sky account so I can tag you pls).
Yes, this works very well & easy to do - use ORCID ✌️
#socialpsyc
#psychSciSky
#AcademicSky
