News

Here you can find the latest updates on the Collaborative Research Centre "Global Dynamics of Social Policy": summaries of current research results, references to our latest publications, outcomes of events and more news from the projects and their staff members.

A group of researchers from the CRC 1342 and BIGSSS attended a two-day workshop given by Dr Saskia Schottelius on “The Art of Self-Presenting for Female Scientists”. The workshop covered several aspects related to self-awareness, self-confidence, and self

A group of researchers from the CRC 1342 and BIGSSS attended a two-day workshop given by Dr Saskia Schottelius on “The Art of Self-Presenting for Female Scientists”. The workshop covered several aspects related to self-awareness, self-confidence, and self-presentation; focussing particularly on the additional challenges that women encounter in academic institutions that have historically been dominated by men. Through theoretical inputs, practical exercises, and peer-to-peer coaching, the participants were able to reflect on their own strengths and goals, while benefiting from the mutual encouragement provided by the group. The many themes covered will require continuous practice, and the participating researchers agreed that the workshop had benefited them for their academic paths going forward.


Contact:
Dr. Anna Wolkenhauer
CRC 1342: Global Dynamics of Social Policy, Institute for Intercultural and International Studies
Mary-Somerville-Straße 7
28359 Bremen
Phone: +49 421 218-57099
E-Mail: anna.wolkenhauer@uni-bremen.de

Members of the project’s expert network GIST
Members of the project’s expert network GIST
On the 5th and 6th of May, the B01 project held a workshop at the University of Bremen. They were joined by members of the project’s expert network GIST (Group Inclusion and Social Policies over Time), an international group of social policy scholars from

A central aim of the project is to compare the temporal sequence of inclusion into social security schemes in 20 countries around the world. The project aims to explore national social security legislation to find out what groups were covered by social security in what temporal sequence. The purpose of the workshop was to draw conclusions from the comparative assessment of group constructions (in laws on old-age security, survivors' pensions and unemployment insurance) for exploring the legitimation patterns for the inclusion and exclusion of groups. Moreover, the network planned its publication strategy and laid down next steps for future projects and cooperation.


Contact:
Dr. Johanna Kuhlmann
CRC 1342: Global Dynamics of Social Policy
Mary-Somerville-Straße 7
28359 Bremen
Phone: +49 421 218-58574
E-Mail: johanna.kuhlmann@uni-bremen.de

@ Pixabay (RosZie)
@ Pixabay (RosZie)
Within the frame of the A07 project “Global Dynamics of Long-term Care Policies”, the team supervised by Heinz Rothgang and Simone Leiber is looking to implement innovative research on long- term care systems in international comparison.

A particular aspect of the A07 project is the application of case vignettes to investigate the inclusiveness of different benefits schemes in a selected sample of countries. The goal is to identify and assess the legislative strength and relevance of various factors influencing eligibility, with particular attention to care needs, cognitive limitations, and socio-economic conditions. To properly discuss the implementation and select the variables and eligibility conditionalities to include in the vignettes, the A07 team is organising two workshops with a group of renowned international experts with extensive knowledge on the topic of long-term care, who agreed to support and consult the team throughout the process.

Using vignette cases allows for combining the details and complexity of real-life scenarios with experimental designs’ variable modifications, which in turn provides the chance to assess the relevance of a single variable. While this methodology is mainly applied to interpret individual judgement, beliefs, and intended conduct, team A07’s approach to the methodology is different as it profiles the method to be used in the field of social policy, where vignette studies are still underused. In this application, the vignettes present different carefully constructed profiles of people needing long-term care through constructed descriptions, including various aspects of needs for care, economic situation, family constitution, and living conditions. The vignettes, obtained by the symmetrical and orthogonal combination of the selected variables from these fields, are then going to be presented through a questionnaire to a group of respondents with in-depth knowledge of the single long-term care systems eligibility assessment methodologies to evaluate which benefits could be granted to the different profiles in each country. Against the background of the A07 project’s aim, which pays particular attention to the risk of dementia in long-term care, the role of the syndrome in granting potential access to benefit schemes will be under scrutiny.

The results will then be analysed quantitatively to identify correlations between specific conditionalities and the inclusion in the target population for particular types of benefits.

The first and more informational meeting took place on Thursday, the 23rd of March, 2023. It produced a fascinating discussion on the methodology and the aspects and factors that need to be considered to make the application of the method feasible while balancing the intrinsic trade-off between depth and case complexity that using vignettes requires. Building on the feedback obtained in the first meeting, the A07 team plans to organise a draft formulation of the different vignettes and the questionnaire for data collection. The second meeting with the consulting experts will take place on the 24th of May, 2023, and will focus on reviewing and refining the drafted vignettes before finalising the questionnaire.


Contact:
Prof. Dr. Simone Leiber
CRC 1342: Global Dynamics of Social Policy, Faculty of Education
Universitätsstraße 2
45141 Essen
Phone: +49 201 183-2319
E-Mail: simone.leiber@uni-due.de

Prof. Dr. Heinz Rothgang
CRC 1342: Global Dynamics of Social Policy
Mary-Somerville-Straße 3
28359 Bremen
Phone: +49 421 218-58557
E-Mail: rothgang@uni-bremen.de

One of Italy‘s long-term care experts, Dr. Giovanni Lamura, who leads the Centre for Socio-Economic Research on Ageing (https://www.inrca.it/zz?uMH9) within the National Institute of Health and Science on Ageing (INRCA IRCCS), had exciting news to deliver

The main factors that made this much needed reform possible included the COVID-19 pandemic, post-pandemic resilience plans funded by the EU, the leadership of the previous government under Draghi, as well as the joint lobbying efforts of a large civil society coalition together with a group of long-term care experts. The coalition comprises almost 60 organisations including care providers, trade unions, employer associations, religious organisations, and the most expert academics in the field, including Lamura himself. Strongly supported by the bottom-up process initiated by this coalition, the framework law will allow for the creation of legislative decrees which in turn can implement its core elements via more detailed acts.

The elements of the framework law cover several aspects of long-term care for older people. The proposal made by the coalition to establish long-term care as a distinct sector of the welfare state was not included in the framework law, but one of the main elements of Italian long-term care will be significantly changed: The care allowance (indennità di accompagnamento) will see its first major overhaul since its creation in the 1980s. It will be replaced by a Universal Benefit which for the first time will foresee different levels of care needs. This will also require a standardisation of the assessment system, which so far had been characterised by wide regional differences in terms of practical implementation. The Universal Benefit will be granted either in the form of monetary transfers, or in-kind services. The latter could also include privately hired home-based care workers – half of which are currently employed in undeclared form – and thus facilitate regular contractual conditions. Other plans to integrate these home-based care workers (mostly female migrant workers) into the long-term care system foresee harmonised training paths and standards, as well as the reorganisation of existing contributions and tax reliefs for families who employ these care workers.

Other elements of the reform address the requalification of residential care, the horizontal fragmentation of the current system via an inter-ministerial committee, the vertical coordination of different governance levels, as well as improved support structures for family carers. Finally, the framework law also promotes active ageing and thereby follows a holistic and preventative approach to long-term care.

Whilst Dr. Lamura pointed to the difficulties on funding for the envisaged changes in long-term care – a policy field that is often subjugated to healthcare – he did underline the enormous step that was taken with this framework law. He is hopeful that the effective lobbying efforts of the civil society coalition that formed in this process will continue to positively impact the reform of long-term care for older people in the upcoming years.

To learn more about the pact for a new welfare in long-term care for older people, visit (only in Italian): https://www.pattononautosufficienza.it/.

Text: Marlene Seiffarth

Marlene who was a member of the CRC 1342 in the first funding period has analysed which and how actors have shaped the institutionalisation process of the migrant-in-the-family care model in Italy.

On Tuesday, 21 March 2023, Marlene Seiffarth successfully defended her PhD dissertation based on her research as part of the project “Transnational Service Provision in Long-term Care between Western and Eastern Europe” (https://www.socialpolicydynamics.de/projects/completed-projects/project-b07-2018-21-/en/) of the first funding period. Her cumulative dissertation entitled “The Perpetuation of the Migrant-in-the-Family Care Model in Italy” is comprised of three first-authored papers, all published in peer-reviewed journals (see links below). In a colourful and thought-provoking presentation, she presented the contributions of her papers to both the academic and policy world. All papers attest to the perpetuation of a care model based upon and reproducing global inequalities. These inequalities manifest in inadequate working conditions for the almost one million (mostly female) migrant care workers in the country, who mainly come from Eastern European countries (63%) and countries of the Global South (37%).  Although informal employment (payment off the books and without social security registration) is wide-spread in the migrant care work sector, the rate of informal employment decreased from 90% in 1995 to 52% in 2021. An important factor in this formalisation process in Italy are the efforts of the collective bargaining social partners – trade unions and employer associations – who have created irrevocable standards in the absence of national policies and outdated legislation, as well as a conducive environment for formalisation via their administrative and legal support services they offer to their members.

 

For more results of Marlene’s research, access her publications:

Seiffarth, M. (2023) Collective bargaining in domestic work and its contribution to regulation and formalization in Italy, International Labour Review, Accepted Author Manuscript, https://doi.org/10.1111/ilr.12382

Seiffarth, M. (2022) Potenziale für „gute Arbeit“ im Privathaushalt? Regulierung und Interessenvertretung migrantischer Pflegekräfte in Italien, WSI Mitteilungen 75(5), 386-393, https://www.wsi.de/de/wsi-mitteilungen-regulierung-interessenvertretung-migrantischer-pflegekraefte-in-italien-43650.htm.

Seiffarth, M & Aureli, G. (2022) Social Innovation in Home-Based Eldercare: Strengths and Shortcomings of Integrating Migrant Care Workers into Long-Term Care in Tuscany,
Int. J. Environ. Res. Public Health, Vol. 19 (17), 10602; https://doi.org/10.3390/ijerph191710602

Seiffarth, M. (2021) Crisis as Catalyst? Romanian Migrant Care Workers in Italian Home-Based Care Arrangements, Sociológia - Slovak Sociological Review, Vol. 53 (5), 502-520; https://doi.org/10.31577/sociologia.2021.53.5.19

Dr. Lorraine Frisina-Doetter (SOCIUM and CRC 1342 member) served as rapporteur for the WHO/Europe at their first-ever “Health in the Well-Being Economy” Regional Forum on 1-2 March 2023 at UN City, Copenhagen, DK.

Building on the growing awareness of the importance of health to well-being economies, the Forum demonstrated how countries are already shifting investment, spending and resources.
The Forum also focused on actions needed to be taken going forward, as the European Region faces the interlinked challenges of climate change, the war in Ukraine, the COVID-19 pandemic and the cost-of-living crisis. These challenges are placing health, social care and welfare systems under strain, and widening health inequities across the Region.

The event gathered high-level representatives from ministries of health, finance and economy, together with government advisors working on recovery, resilience and sustainable development policies, public health policy-makers, and representatives of national and international banks, nongovernmental organizations, the United Nations and European Union agencies.

Key speakers included: Dr Hans Henri P. Kluge, WHO Regional Director for Europe; Katrin Jakobsdóttir, Prime Minister of Iceland; Professor Mario Monti, Former Prime Minister of Italy and European Commissioner; Professor Sir Michael Marmot, Director, Institute of Health Equity, University College London.


Contact:
Dr. Lorraine Frisina Doetter
CRC 1342: Global Dynamics of Social Policy
Mary-Somerville-Straße 3
28359 Bremen
Phone: +49 421 218-58561
E-Mail: frisina@uni-bremen.de

Irene Dingeldey and Ulrich Mückenberger edited the latest issue of the International Labour Review, which focuses on the concept of legal segmentation that was developed in the CRC 1342.

The International Labour Organisation (ILO) has published the Special Issue Overcoming legal segmentation: Extending legal rules to all workers? in its journal International Labour Review. The volume was edited by Irene Dingeldey and Ulrich Mückenberger ( project A03) and features results from the first funding period of the CRC 1342 on the concept of legal segmentation (i.e. labour market segmentations that stem from state law) and related quantitative results. The Special Issue is supplemented by regional studies by recognised experts on South Asia, Latin America and Southern Africa, which essentially confirm the argument that law plays a segmenting role, which has only been moderated in recent decades by equalising normative elements. In this volume, the ILO also includes the critical view that its own norm-setting practice is characterised by legal segmentation and has only recently given more space to universalising regulation.

The ILO's International Labour Review is published in English, Spanish and French, which means that the results of project A03 enter into a global discourse among academics, practitioners and political decision-makers involved in social policy.

---

Ulrich Mückenberger & Irene Dingeldey (2022): Overcoming legal segmentation: Extending legal rules to all workers?, International Labour Review (Special Issue), Volume 161 (4). https://onlinelibrary.wiley.com/toc/1564913x/2022/161/4


Contact:
Prof. Dr. Irene Dingeldey
CRC 1342: Global Dynamics of Social Policy, Institute Labour and Economy
Wiener Straße 9 / Ecke Celsiusstraße
28359 Bremen
Phone: +49 421 218-61710
E-Mail: dingeldey@uni-bremen.de

Prof. Dr. Ulrich Mückenberger
CRC 1342: Global Dynamics of Social Policy, Faculty of Law
Universitätsallee, GW1
28359 Bremen
Phone: +49 421 218-66218
E-Mail: mueckenb@uni-bremen.de

Fritz Kusch (Photo: BIGSSS)
Fritz Kusch (Photo: BIGSSS)
Fritz Kusch is searching U.S. archives for evidence that protectionism and social policy are intertwined. In our interview, he takes stock of his initial findings and recalls scribbled notes by William II that he tried to decipher for his master's thesis.

Dear Fritz, you've been at the CRC for about half a year now - what did you do before that?

I studied history as a major and political science as a minor in Freiburg and graduated with a Bachelor's degree. During this time, I was also at a university in Istanbul, where I got to know and appreciate the Turkish language. I also started to study Turkish and Ottoman history. This then led me to start a second Bachelor's programme after graduating from Freiburg: Turkish Studies at the Free University of Berlin.

Parallel to your Turkology studies, you also completed a Master's degree ...

I did a Master's in History with a focus on North America. US history has been a passion of mine for a long time. There are very few points of contact between North American and Turkish-Ottoman history, but I actually consider that more of an advantage: it is definitely rewarding to look at completely different topics and contexts. This often gives you a new, fresh perspective on the individual topics.

What was your Master's thesis about?

About two statues of Friedrich Wilhelm von Steuben, erected in Washington in 1910 and in Potsdam in 1911. The subject combines the history of remembrance, migration and diplomacy. Steuben was a Prussian general who fought in the American War of Independence and became an ethnic hero for the ethnic German population in America. A kind of symbolic representative for the status of German Americans in the American immigrant society. Around the turn of the century, several ethnic groups in the USA campaigned for statues of their respective Ethnic Heroes to be erected in public spaces. In Lafayette Square behind the White House in Washington, for example, two statues of French generals from the War of Independence were erected. Other groups, such as the German-Americans, noticed this and demanded a statue of Steuben, which was also passed by Congress in 1902. Foreign powers, such as the German Empire, also noticed this and entered into this somewhat convoluted process of statue erections and gifts, whereby the prestige rivalry of ethnic groups also became a diplomatic prestige rivalry. The German Empire gave the USA a statue of Frederick the Great: a poorly thought-out gift that also went down rather badly in the USA - after all, Frederick was an absolute monarch! The American government did not erect the statue in the city, but very discreetly in the War Academy. In addition, the erection was delayed for more than two years; in other words, the statue was hidden away. It was not until 1910 that the German-Americans received their Steuben statue, which they and certainly the US government understood as a sign of their belonging to the American nation. The USA then in turn gave a copy of the statue to the German Reich as a diplomatic counter-gift to the Frederick statue. This copy was inaugurated in Potsdam in 1911. There you can see how German-American prestige aspirations intersected with German-American diplomacy. Moreover, the same Steuben statue was charged with very different meanings in Potsdam and Washington: For German-Americans, both statues honoured their Ethnic Hero and thus indirectly actually the group of German-Americans as a whole, but for the German Reich government, the Potsdam statue was a symbol of a supposed German-American friendship that had always existed.

How did you research the details of this story?

Unfortunately, I couldn't travel to Washington, so I researched everything in Berlin archives. My sources were mainly dispatches between the embassy in Washington and the government in Berlin, which I was able to see in the political archives of the German Ministry of Foreign Affairs. I also researched in the Secret State Archives. Among other things, I found out that the statues were clearly more important to the Kaiser and his staff than one would assume. Wilhelm the Second dealt with them in quite some detail: The idea for the Frederick statue goes directly back to him, he personally chose the site for the Steuben statue in Potsdam, he provided his court architect to redesign the square, and he wrote whimsical comments (which I was able to decipher to some extent, sometimes more, sometimes less) in the margins of the draft dispatches.

It must be quite fascinating to hold documents in your hand on which Wilhelm the Second scribbled with his fountain pen ...

Absolutely, although Wilhelm the Second did not use ink, but a pencil. Officials and diplomats at the time had a rather bureaucratised style of writing, which of course didn't apply to Wilhelm. This made it very easy to see what the Emperor himself had written: he simply wrote off the cuff.

A few weeks ago you were in the USA to do research for your work at the CRC. Where were you exactly what were you looking for?

I was in the US for two and a half months, mainly in Washington, but then also in Wilmington, Delaware, and in Hyde Park, New York. Our project investigates the connection between social policy and protectionism in the USA and Argentina, and I am responsible for the period 1890 to 1970 in the USA. The question was: Can I also demonstrate this connection, which is established in the social science discussion, by means of historians? I would need to find corresponding documents that show that the political actors actually discussed this connection, e.g. raising tariffs in order to fund social welfare programmes. I looked for it - but I didn't find anything on this, or only in rudimentary terms.

Where did you look for evidence?

I went to a whole series of archives. The Library of Congress, the Hagley Archives in Wilmington, the Franklin Roosevelt Presidential Library in Hyde Park, the New York Public Library and the archives of the umbrella union AFL-CIO, housed at the University of Maryland. I spent most of my time at the National Archives in Washington, or College Park, looking through the files of the Departments of Commerce, State and Labour. It's easy to follow customs policy because the responsibilities are clear. This is not the case with social policy, which makes research immensely difficult. At the federal level, there was virtually no universal social policy until the New Deal. There was veterans' care, which is roughly related to tariffs through its funding. Since the 1960s, the US has had Trade Adjustment Assistance (TAA), a social programme that provides assistance to the workforces of companies facing hardship from international competition: Benefits included retraining, relocation assistance and the like. That's interesting and hardly explored by historians so far, but I wasn't sure I could build a dissertation on it.

When did you realise that you weren't really finding what you were looking for?

Even before I had travelled to the USA. I had looked at the archive catalogues beforehand.

To what extent are the files there digitally searchable?

In the USA, this works very well. The catalogues are almost always online. In the National Archives, for very central files, the entire documents were scanned and are available online. So I can read them from Bremen. But that only applies to extremely important documents. For the main part of the collections, at least the Finding Aids are available digitally. Most of the time, you can see what's in an archive box down to the level of the individual folders. It says, for example, that the folders contain the minutes of Committee XY in a certain period, and if you're lucky, there's even a paragraph on what was discussed.

So I already had my doubts in Bremen that I would find enough material. At the same time, it was clear that I would never be able to look at all the material there - hundreds of archive boxes with hundreds of pages each. It would be unrealistic to look at all of it. Therefore, before my trip, I had thought about looking for something else. That's how it turned out: now I'm studying protectionist pressure groups, i.e. organisations that campaign for high tariffs. In the historical literature, these organisations do appear, but they are usually on the margins. The literature on tariff policy is often written by political scientists and economists. Neither have paid much attention to pressure groups. My initial research, however, showed that they were extremely engaged. Since the 1880s, there was a strong effort to bring protectionism into the public domain. This was also pursued with a lot of money, for example by the iron and steel industry.

Did these calls for more protectionism come only from the employers' associations or also from trade unions?

The associations are pretty clearly about the interests of capital. The largest federation I am researching is the American Protective Tariff League, founded in 1885. Together with a student assistant, I have already taken a closer look at who the actors were in it: Capitalists and Republican Party politicians. Tariff policy was partisan politics in those days: the Democrats were for low tariffs, the Republicans for high tariffs. When the American Protective Tariff League was founded, heavy labour conflicts were raging in the US, so industrialists were anything but worker-friendly. This is not to say that there were not protectionist sentiments in labour or that there were not attempts to win workers over to protectionist positions. Surprisingly, the interests of American workers appeared relatively frequently in the pressure groups' pamphlets and leaflets: The American standard of living and the relatively high wages compared to Europe depend, so the argument goes, on the protection provided by high tariffs. If this protection were to break down, the USA would be flooded with cheap European imports, as a result of which the American worker would no longer be able to enjoy this "great" prosperity. In the tariff issue, therefore, there is alignment of interests between corporations and workers, at least that is the reasoning. In addition, there is often the classic argument of infant industry: young industries must be developed under the protection of high tariffs; as soon as the industry is competitive, tariff protection is no longer needed. However, this argument became increasingly obsolete towards the end of the 19th century, as US industry was superior to European competition in many areas.

So from the US perspective, there were definitely good reasons to reduce protectionism.

Absolutely, especially of course in sectors that had strong export interests. Incidentally, this also led to tensions within the Republican camp: Some industrialists demanded lower tariffs so that the US would get better export opportunities in return. I follow the American Protective Tarrif League from the 1880s to around 1960, during which time its political environment changes radically: the USA begins to dismantle protectionism from 1934 onwards and quickly becomes the driving force behind international trade liberalisation after the Second World War. Even before that, there were first attempts at tariff reform. The US transformed itself from a high-tariff to a low-tariff country. I am curious to see how organisations like the American Protective Tarrif League have dealt with this change. These are, after all, organisations that have boasted of dominating the public debate and even putting presidents in office. They saw themselves at the levers of power. A few decades later, they are in total opposition. I want to look at whether and how the framing of the argument changed as a result. How much of the arguments still remain? How does the ideology change? What personnel continuities are there, which industries remain or become active? Do workers and trade unions join in at some point? I am also interested in the means of agitation. I would like to examine how attempts were made to popularise a rather technical and abstract topic like customs policy in such a way that it appeals to wider circles. This naturally leads to the question of why this worked better at some times and almost not at all at others.

Have you been able to collect enough material yet, or do you have to return to the archives again later?

Of course I would like to go on another archive tour. But I have already collected a lot of material - probably more than I can even read in detail. I'm currently working on my dissertation proposal: When I get to the detailed planning, it could be that I realise: There are still some gaps here or there that I need to fill in a second round of research. But before I set off on a second trip, it's the turn of the other project members.


Contact:
Fritz Kusch
CRC 1342: Global Dynamics of Social Policy, Faculty of Social Science
Universitäts-Boulevard, GW2
28359 Bremen
Phone: +49 421 218-58581
E-Mail: kusch@uni-bremen.de

Dr. Gregor Wiedemann
Dr. Gregor Wiedemann
Gregor Wiedemann recently gave a workshop at CRC 1342 on AI-based analysis of large text volumes. In this interview, the expert in computational social sciences explains the opportunities this approach creates for social sciences.

Dear Gregor Wiedemann, what applications of Natural Language Processing, i.e., automated text and language processing based on artificial intelligence, do you currently see in science and especially in the social sciences?

One focus of Natural Language Processing applications in the social sciences is the assessment of very large amounts of text for automated content analysis. This makes, for example, debates in social media or across many news media accessible for research that otherwise would be impossible to handle manually.

What are you using NLP for at the moment, personally?

I'm currently working on NLP methods for evaluating argument structures in Twitter debates. Among other things, we are observing how public discourse on the use of nuclear energy has changed in recent years. For example, environmental aspects play a role in pro-nuclear arguments much more often, which is a significant change from previous decades. In a second project, we extract protest events such as demonstrations, rallies or strikes from local media and prepare these data for protest event research.

Let's assume I have collected tens of thousands of text documents in my project that I want to analyze. How can Natural Language Processing help me with this?

Let's stay with the example of protest event research. Here, for example, a local newspaper reports on a Pegida rally in Dresden. To find this coverage automatically, we use a classification method that automatically detects whether articles containing the word "Pegida protests" are reporting on a protest event or whether the word is only used in a more general context. This allows us to reliably identify only those articles that actually report on protest events from the large set of all newspaper reports that contain these keywords. In a second step, we automatically extract information such as motto, number of participants and organizers from these articles. The end result is a structured dataset of all local protest events from four local newspapers over the last 20 years, which can then be analyzed by political scientists.

What requirements must my text data meet so that I can analyze it using Natural Language Processing?

Ideally, the training data for a model should come from the same population as the target data. The texts must also be available in digital form, of course, and they should not deviate too much from the standard language. That means, for example, transcripts of spoken language that depict strong slang, or historical documents that use a very old language, can sometimes cause problems. But there are solutions for this. In these cases, the language models used must be adapted for the target domain. By the way, it is now easily possible to use multi-lingual corpora (e.g. German and English) in the same analysis, or to make good predictions for German target data with English training data.

And what prerequisites do I as scientist have to bring along to work with NLP? How much programming knowledge do I need or other relevant prior knowledge?

For some years now, NLP methods are mainly used on the basis of multi-layer neural networks (also known as "deep learning"), because they perform significantly better than earlier approaches, which were based on word lists, for example. In order to use these neural networks for large amounts of text, it is necessary to deal with certain program libraries, which are usually written for the script language Python. Nowadays, there are also first so-called R-wrappers, which make the functionalities of the Python libraries available in the R programming language. However, there is currently no way around programming your own analysis scripts.

Before I can analyze my text data, I have to tell the algorithm what to look for and what to do with what it finds. How do you do that?

In order to teach a machine which texts to extract as relevant from a large set, which categories to sort a text into, or what information to extract from a document, I need to teach it what that information looks like using examples. To do this, a training data set is created, typically containing a few hundred or thousand positive and negative examples for the target category. With new technologies called Few Shot Learning, significantly fewer examples are already sufficient to train a pre-trained neural network on a target category.

What does my training data set have to look like so that the quality of the analysis is high and reliable afterwards?

The training data for training an NLP model should be as complete and as uniformly coded as possible. Complete means that there is a conscious category decision for all entities (words, sentences, or documents) presented to the model. Consistent (or reliable) means that if several coders create a data set, they must also reach the same judgment about category assignment in the same cases. If the machine is presented with inconsistent training data, it will not be able to learn a category adequately.

How can I tell that the NLP is working reliably?

To tell if an NLP model works reliably, it is tested with test data. Test data, just like training data, are hand-coded texts. The predictions of a trained model on the test data are then compared with the hand-coded categories. In this way, a statement about the quality of the automatic prediction is possible.

The technology in this area is rapidly progressing. What developments have advanced NLP in particular recently?

We are currently experiencing rapid progress in the field of artificial intelligence and machine text understanding with so-called large language models. The most recent example is ChatGPT from the company OpenAI. This language model is able to understand requests and responses from users and thus generate human-like dialogic communication.

What are the limitations of NLP from your point of view?

Although language models currently already perform impressively by presenting the knowledge stored in them to users in dialogic form, their internal structure lacks symbolic-logical knowledge representation that is transparently comprehensible from the outside. The goal of current research is to make this "black box" decodable within neural networks and thus also to make sure that more formally correct and better validated knowledge can be stored in these models.

---

About Dr. Gregor Wiedemann

Dr. Gregor Wiedemann is Senior Researcher Computational Social Science at the Leibniz Institute for Media Research │ Hans Bredow Institute (HBI), where he directs the Media Research Methods Lab (MRML) together with Sascha Hölig. His current work focuses on the development of methods and applications of Natural Language Processing and Text Mining for empirical social and media research.

Wiedemann studied political science and computer science in Leipzig and Miami. In 2016, he completed his doctorate in computer science at the University of Leipzig in the Automatic Language Processing department on the possibilities of automating discourse and content analysis using text mining and machine learning methods. He then worked as a postdoc in the Department of Language Technology at the University of Hamburg.

He was working on methods for unsupervised information extraction to support investigative research in unknown document repositories (see newsleak.io) and the detection of hate speech and counter-speech in social media.

Until taking over the leadership of the MRML, he was working in the DFG project "A framework for argument mining and evaluation (FAME)", which deals with the automatic detection and evaluation of recurrent argument structures in empirical texts.

(Source of short biography: Hans Bredow Institute)