R-Stats resources for archaeologists – complete beginners!

The revaluation of Western European glass compositional data and the observation of patterns between different variables (chemical compositions, typology, chronology, etc…) is a key point of my MSCA project. Someone might even suggest that I had this idea so that I could FINALLY justify learning R and applying statistics to glass analysis… 😛

Months ago I started to put together a series of resources that could help learning both the basics of statistics and R-coding. So far I can do the really basics things such as plotting and some really crude PCA and hierarchal clustering analysis, but this clearly is not enough.

My approach was too naïve and disorganised and I clearly needed some help of some expert users to start taking seriously statistics and R-coding.

This is why I have asked for help to the R community through the #rstats hashtag on my twitter account:

There are some really good suggestions in the thread that I didn’t put here, so I highly recommend to check it out.

Moreover, there is nothing worse than keeping good resources to yourself. So, I will start to compile a list of R books, blogs, social media account that will be useful to beginner-users like me. This list does not want to be in any way comprehensive: just a small nudge in the right direction and a way for me to keep track of the most useful tools and books (or even the things I find interesting!). These resources are, of course, targeted specifically to other archaeometrists, but I really hope they will be useful to other researchers.

You are most welcome 🙂

Camilla

Books/articles:

Baxter, M.J., 2015. Notes on quantitative archaeology and R. Nottingham. https://www.academia.edu/12545743/Notes on Quantitative Archaeology and R

Baxter, M.J, Cool, H., 2016. Basic Statistical Graphics for Archaeology with R: Life Beyond Excel. Nottingham: Barbican Research Associates .

Baxter, M.J., Cool, H., Jackson, C.M., 2006. Comparing Glass Compositional Analyses. Archaeometry ,48, 399–414. https://doi.org/10.1111/j.1475-4754.2006.00263.x

Carlson, D.L., 2017. Quantitative methods in archaeology using R. New York, NY: Cambridge University Press. https://doi.org/10.1017/9781139628730

Drennan, R.D., 2009. Statistics for archaeologists: A Common Sense Approach. Boston, MA: Springer. http://doi.org/10.1007/978-1-4419-0413-3

Shennan, S., 1997. Quantifying archaeology. Edinburgh : Edinburgh University Press. https://doi.org/10.1016/C2009-0-21716-8 

Wickham, H., Grolemund, G., 2016. R for data science: import, tidy, transform, visualize, and model data. O’Reilly Media, Inc. Start here, this book is available online for free: https://r4ds.had.co.nz/

Wickham, Hadley, Navarro, Danielle, Lin Pedersen, Thomas, 2015. ggplot2: Elegant graphics for data analysis. Springer. Start here: https://ggplot2-book.org/

Websites:

https://www.r-bloggers.com/ – Incredible resource for R news and tutorials

RStudio education – RStudio (ultra-famous free integrated development environment for R) offers a series of resources based on your starting level (beginners, intermediates, and experts), Very very useful! Check also: cheatsheets and tutorials.

https://rladies.org/ – World-wide organisation to promote gender diversity in the R community. They are also have a blog: https://blog.rladies.org/post/

https://stackoverflow.com/ – question and answer site for professional and enthusiast programmers.

https://r-charts.com/ – Code examples of R graphics, ggplot2, and other packages.

https://rmarkdown.rstudio.com/ – Turn your analyses into high quality documents, reports, presentations and dashboards.

https://stat545.com/index.html – Data wrangling, exploration, and analysis with R by Jenny Bryan (excellent online course where you will learn to explore, visualize, and analyze data using R).

R-Packages:

https://www.tidyverse.org/ – The tidyverse is an opinionated collection of R packages designed for data science

Useful tags:

#rstats #DataScience #dataviz #Rchaeology

Twitter accounts:

@RLadiesGlobal

@Rchaeology1

Would you like to know where funnel beakers were made? Me too!

I don’t know if I have already mentioned it, but I have a couple of talks coming up in the next two weeks (one for Newcastle and another one for Harvard). Given that I have just started my research, my aim for both talks is to explain my research and the process that lead me to build my MSCA Global project.

When you focused on chemical analysis it becomes easy to get carried away by your data, but in my opinion archaeometrists (myself included) do not spend enough time reflecting on what the data can or cannot tell you.

I strongly believe that scientific data certainly it is a good start to describe glass production, but on its own they cannot explain its complexity. We can answer questions about what raw materials have been used or where they have been sources, if the fragment was recycled or not, but what about the other technological, economic, and social aspects involving glass production?  

And in my opinion the only way to do this to allow multidisciplinarity to enter the picture.

In my talk I want to make a practical example to make the audience better understand the big picture. The palm cup/funnel beaker serie is definitely one of the most characteristic type for the 7th-9th century given its broad diffusion in Western Europe.

Funnel beaker from Sweden (http://mis.historiska.se/mis/sok/fid.asp?fid=106824)
Photo by Gabriel Hildebrand SHM

One of the big question that remains unanswered is “where these vessels have been produced?”

A very appealing idea is that many glass-vessel types found in Scandinavia have a west European-Anglo-Saxon distribution (Feveile 2006, 252), different types of pam cup/funnel beaker included (Näsman 1990; Feveile 2006, fig.34, 236). The palm cup/funnel beakers distribution could therefore suggest a common origin of such vessel, maybe from the Frankish/Rhine area via Dorestad and Frisian Coast together with other materials such as Rhineland basalt and pottery (Feveile 2006, 252).
This hypothesis is very tantalizing, but it is based on common typological similarities and materials distribution, but does not include scientific analysis of glass composition and therefore in my opinion cannot describe realistic patterns of production.

Other works focused on the chemical composition seem to break this narrative:  differences in their major and minor oxides (Sanderson et al. 1984; Hunter and Heyworth 1998) seem to suggest that even if the palm cup/funnel beakers look similar, they were made with different raw glass. Could it be that the use of different raw glass implies the involvement of different secondary glass-making centres?

This certain a possibility, but the same scientific studies concerning 7th – 10th century vessel glass (Sanderson et al. 1984; Henderson and Holland 1992; Hunter and Heyworth 1998; Gaut 2011) need to be revisited in light of the past 30-40 years scientific literature to be included in the broader Early Medieval glass production narrative.   

Alone these disciplines only partially describe the complexity of glass production, but if we put them together, we might start to see patterns and trends that were hidden before!   

So far, WINDOWGLASSMED is the first project that will try to pair typology and chemical compositions data in conjunction with archaeological evidence to spot clear patterns that could mirror potential glass production areas and trends in Western Europe.

I hope that you start to see the potential of combining archaeological and scientific evidence together. This weekend I will dedicate my time reading a new publication by Broadley (2019) about Anglo-Saxon glass vessels and I am sure I will gather new data and exciting ideas for the talk and my research.

In the meanwhile, I hope I gave you a glimpse of the complexity of glass production in EM and why do we need more interdisciplinary projects in glass studies.

Stay safe out there 😊

Camilla

Bibliography:

Broadley, R., 2019. The Glass Vessels of Anglo-Saxon England: C. AD 650-1100. Oxbow Books.

Feveile, L., 2006. Hulglasskår fra markedspladsen i Ribe, ASR 9 Posthuset, in: Feveile, C. (Ed.), DET ÆLDSTE RIBE Udgravninger På Nordsiden Af Ribe Å 1984-2000. Jysk Arkæologisk Selskab, 195–277.

Gaut, B., 2011. Vessel glass and evidence of glassworking, in: Skre, D. (Ed.), Things from the Town: Artefacts and Inhabitants in Viking-Age Kaupang. Aarhus University Press, Aarhus.

Henderson, J., Holand, I., 1992. The Glass from Borg, an Early Medieval Chieftain’s Farm in Northern Norway. Mediev. Archaeol. 36, 29–58.

Hunter, J., Heyworth, M., 1998. The Hamwic glass. Council for British Archaeology.

Näsman, U. 1990: Om fjärrhandel i Sydskandinaviens yngre järn- ålder. Handel med glas under germansk järnålder och vikinge- tid. Hikuin 16. S. 89- 119.

Sanderson, D.C.W., Hunter, J.R., Warren, S.E., 1984. Energy dispersive X-ray fluorescence analysis of 1st millennium AD glass from Britain. J. Archaeol. Sci. 11, 53–69.

I have a plan.

Hi everyone!

I hope you all had a great week!

For me, this week has been mostly focused on planning and slowly return to a more stable working routine. The past four months have been an absolute personal nightmare between VISA bureaucracy (almost written bureauCRAZY there, how appropriate), moving house four times, and finding a stable place here in Boston. Moreover, today I finally got my Harvard credentials, which means I can finally access the libraries here!

Now that things are starting to look much better and promising, I had the time to sit down and reevaluate my big plan. I have realised that, even if my timeline has been affected, the main goals for the first months of work remain fundamentally the same:

  • Building a reference database for chemical data.
  • Making a personal literature review for Early Medieval typology focused in particular on the samples I already have secured.
  • Focus on adding new samples on my database.
  • Organising LA-ICP-MS analysis in collaboration with Dr Laure Dussubieux (Fields Museum – Chicago)
  • Learning how to use R and in particular to work on basic bi-plots, PCA, and cluster analysis.

The main priority right now will be acquiring new samples. I have some exciting news from Spain and Denmark, but I don’t want to share it just yet 😉

I have already my personal reference data excel spreadsheet; I just need to polish it, add more recent data, and readapt it to my current needs. Learning how to use R correctly will take some time for sure (looking at Drennan 2009 and Carlson 2017 at the moment), but I know that in the end it will be worth it! I have dabbled already with some statistics for my Comacchio (Italy) materials for my PhD, but in the end when I published the material (Bertini et al. 2020) I have decided to take a more straightforward approach (I simply was not ready to use statistics at the time).

Other than blindly going into the rabbit hole that typology is (at least for me), I have decided to take a more practical approach and look up the types of the samples I’ve already have. Next week I think I will introduce why I am also looking at glass types in conjunction with chemical data. It might sound like a quite normal thing to do, but in glass studies the dichotomy between typology and chemistry is still real, and I think there is still so much potential ready to be explored.

And this is it! Quite a short blog post, but I am so happy to finally be able to get things going into the right direction. I wish I had some pictures to show here, but for now this is the best I can do locked up in my Cambridge flat. I cannot wait to finally show some very exciting glass pictures to you all 🙂

Take care and stay safe out there!

Camilla

References:

Introduction to this blog

The aim of this blog is certainly to monitor the progress of my Marie Curie Global project focused on the study of Early Medieval glass, but I’d lie if I’d say that was the only reason; my other goal is to keep myself accountable and force myself to write on it regularly, while documenting my time here in the States and my return to the UK (and hopefully beyond!).

Since every form of in person networking has been taken away for the moment, presenting my research to a wide audience and reaching out to other scholars through the internet has become even more important. Moreover, I don’t want this blog to remain active only during the WINDOWGLASSMED project, but I want to document my trial and error process through my academic journey: who knows, maybe it be useful to other scholars that, like me, are always looking for new inspiration to fuel their creativity. And certainly I will learn some new tricks along the way!

I have decided that a decent goal would be to post at least once every two weeks to not only document every milestone (or lack of thereof!), but also to keep some sort of diary of my research. All posts will be tagged so the reader will be able to choose the type of content that wants to access at that time.

I also have an Instagram account (@theintrovertedemic) where I will post at least once a week about what I am doing and I will share the topic of each before posting here on the blog. You can also follow me on Twitter (@cami82) to get a daily feedback on my research.

Looking forward to share this journey with me!

Camilla