Mapping Transnational Motherhood: The Case of Romanian “Children Left Behind”

Team Members

Donya Alinejad, Claudia Minchilli

Introduction

The issue we want to map revolves around an online campaign in Italy: “Te iubeste mama! / Mamma ti vuole bene!” (trans. Mum Loves you!)

This is the most important campaign among Italian NGOs addressing the phenomenon of the so-called “children left behind” and is promoted by the Italian-Romanian women’s organization A.D.R.I. (Associazione Donne Romene in Italia). What’s the wider context?

• The campaign was a response to the process of “care drain” from Romania to Italy. “Care drain” describes the process by which women from the Global South leave behind their own children to move to First World countries in order to gain a higher income and fulfil the demand for paid work as the carer of First World children.
• Italian and other NGOs signalled that depression among the migrant caregiving women, and suicide rates and their children “left behind” in Romania, were disproportionately high.
• At the same time, these working women are stigmatized for being bad mothers, and the children were seen as victims.

We want to investigate this issue online in order to understand more about how its has spread via different media forms, across different national mediascapes.

Initial Data Sets

• URL list from Google search query: “Te iubiste mama!” from Romanian and Italian domains
• Issue Crawler output based on input of top 100 sites from Google scrape - also for both national domains
• Facebook page like network data for the campaign website
• Youtube video network data for the main campaign youtube video

Research Questions

• To what extent does the issue mobilize different parts of the Italian-based Romanian diaspora on different platforms?
• We expected that the issue would spread differently via specific platforms. And we thought we might find differences between how the issue circulates in Italian and the Romanian online spaces.

Methodology

Day 1

Transnational Motherhood: Romania and Italy

Steps in Netvizz: We did a search in Netvizz Search. The query was the name of the campaign in Romanian: Te iubeste mama!

Three page IDs: 1. 857566424339033 - Mama te iubeste – Personal blog 2. 124511094318810 - Proiectul "Te iubeste mama!" - websites 3. 1443491459309349 - Mama Mea Puterea Meah Te Iubește Fata Tah – Local Business

1. Page like network for this page did not have any connections or any other nodes
2. For this page we were able to make a network visualization
3. This page also didn’t have any connections

So we will focus the analysis only on page number 2, the page of the campaign that is made by the Italian organization ADRI, but that has its description/content in Romanian: Proiectul "Te iubeste mama!"

Steps in Gephi:

• In Gephi we imported the zip file
• Force Atlas 2
• Lin Log mode on (to find nodes)
• Prevent overlap
• Increased Scaling to 5.0
• Scaled the node labels to Degree

What we can see:

• The network overall is closed. In the sense that even the peripheral nodes are linked to other peripheral nodes
• 3 major nodes in the network: The central facebook page that was queried, the Italian organization responsible for the campaign (ADRI – based in Milan), the person’s page who founded the organization.

We didn’t see any see any visible clusters appearing in the above. So starting from scratch again, we added Modularity Class algorithm to come up with “communities”.

Making a list for Issue Crawler:

• Downloaded a new version of Firefox.
• (Went to preferences. Under Search, turned off suggestions. Under Privacy, kept on Never Remember History).
• Then went to tools.digitalmethods.net and chose Google Scraper tool.
• Then installed DMI toolbar.
• Query is entered under keyword (leave URL field empty): “Te iubeste mama!”
• Clicked Advanced Options
• Changed RO under Google Local Versions for Romanian domain. Left all the other fields unchanged. (Claudia chose IT for Italy’s domain).

Donya and Claudia did two different things from here:

Donya: Under Output at the top of the page, chose txt format. Copy pasted the text from the page into Text Wrangler. Saved As tsv file. Opened in Excel. Deleted all the columns except for the list of URLs.

Claudia: Under Output at the top of page, chose to open as CSV file. Then opened the file in Google Sheets in Google Drive In order to clean the data, copied the list of URLs only. Pasted into a new Google Sheet file.

In Issue Crawler: Then we did the same thing as each other again:

Then we made an Issue Crawler account each, and copy pasted our different URL lists into Issue Crawler With 1 iteration of method, 3 deep, co-link, privilege starting points. One list for Italy, one list for Romania.

Day 2

Claudia: tried to see if the Youtube Channel of the Association ADRI can show and confirm connections with the kind of organisations that the study on Facebook shows.

1. Hence, I went on ADRI Channel: https://www.youtube.com/channel/UCRz4CrxJAsdCU_KFTT7O3aw
2. I copied the ID of the Channel : UCRz4CrxJAsdCU _KFTT7O3aw and put it as seed on Youtube Data Tools “Channel Network”. That led to no results.
3. Hence, I decided to make a new research to see which are the videos associated to the main video of the campaign “Te iubeste mama! Mamma ti vuole bene” published by ADRI Channel on Youtube.
4. This time, I put the ID of the video https://www.youtube.com/watch?v=rPX-np7sjmY (ID= rPX-np7sjmY) on Youtube Data Tool “Video Network”.
5. I then opened the .gdf document on Gephi and visualised the network using Modularity to see the different clusters of videos associated to the seed one:

Though, the different labels say nothing about the character of the channels associated to the published videos, since they only report the titles. In order to understand the different character of the channels associated to the nodes/videos, I exported the .gdf file of video networks into a .csv, then opening it on Google sheet to clean it up, adding a further column which describes the nationality of each account associated to the videos. In the meanwhile, we returned on Gephi Facebook visualisation. We decided to delete the nodes “Silvia Dumitrache” and “A.D.R.I - Associazione Donne Romene in Italia”. The reason was that the presence of those two nodes, so strongly interrelated with the project’s one (the first is the page of the main funder of the project and the second is the one of the association that promoted the project), was causing the inclusion of nodes - and so clusters - not minimally related to last. Finally, we decided to delete also the main node “Te iubeste mama!”, in order to understand the degree of importance of the pages linked to it and re-formulate the clusters.

Manual coding countrys’ associations and colour in the graph.

Findings

Facebook Page Like Network for campaign page:

• 3 main nodes: campaign page (seed), organization initiating the campaign, page of individual who founded the campaign. (No visible clusters)
• Local Italian cluster: Italian (women’s) associations cluster, Democratic Party.
• Romanian clusters: diaspora/international Romanian TV channels; libraries in Romania (collaborating with a campaign project), transnational motherhood

The most striking overall finding was that there was no Italian national cluster in the visualization. Rather, the local dimension of the issue network in Italy was clearly dominant. The local Italian cluster included a variety of different organizations based in specific cities, in particular Milan. These included local sections of the Italian Democratic Party, women’s organizations, and other non-governmental associations. Aside from the locally-oriented Italian clusters, there was a broad section of the network that related to Romanian diaspora issues more broadly. These included pages about diaspora/international Romanian TV channels, on transnational Romanian motherhood, and also a sub-cluster of pages of libraries based in Romania (who worked together with the project on events that placed mothers and children in contact with each other via Skype).

In addition, we found a marginal but recurrent cross-diaspora connection to Moldova (when also taking into consideration the URLs from Google Scrape) in the form of a joint Moldovan-Romanian diaspora organization based in Italy appearing in the network, and hyperlink (in the URL list - next section) to an Italian language article mentioning a case of a Moldovan woman migrating to Italy, leaving behind her child.