Mapping the cultural authority of science across Europe and India

Lead Research Organisation: London School of Economics & Pol Sci
Department Name: Social Psychology

Abstract

Modern societies performs their understanding of issues in conferences, opinion polling, scoping exercises, everyday conversations and mass media coverage where the sciences are granted variable cultural authority. This Indo-European network will construct a system of science culture indicators based on news analysis and on attitude measures. Trends in science news (1990-2010; intensity, positioning, issues) are linked to trends in public attitudes to science (interest, attitude, knowledge, engagement: Eurobarometer & NCAER surveys, 1989-2010). The network mobilises and develops joint expertise in discourse analysis, computer-assisted text analysis, large scale survey research, and sophisticated statistical techniques to track the cultural authority of science.

Planned Impact

Not Applicable

Publications

10 25 50
 
Description 1. OECD Blue SKY Thinking Papers on indicators of 'science culture'
Martin Bauer had been invited by the OECD working group ST&I indicators, twice before, once in Paris (March 2014) and another time in Tokyo (Oct 2015) to present the MACAS ideas on science culture indicators. In September 2016, the OECD organised the Blue Sky thinking meeting on ST&I indicators in Ghent, and they asked me to organise a symposium on indicators of science culture. The MACAS team presented two papers contributing to the discussions: first, a paper on mass media mapping to create cultural indicators from news reference to ST&I (Suerdem & Bauer, 2016); second, a paper on mapping the cultures of science across Europe (Bauer & Suerdem, 2016). In the latter we developed a new analytical procedure to deal with repeated cross-sectional data on public perceptions of science as this is available in Eurobarometer (six waves. 1989-2013). We examined the data with 'data mining' techniques, combining missing data analysis, imputations, IRT, EFA and CFA to test a 3D model of science culture comprising the dimensions, knowledge, evaluation and engagement. Consistent with previous observations the evaluation of science splits into two sub-dimensions, utility oriented Progress, and value-based Reserves. On the basis of this stabilised system of indicators, we are able to cluster European countries into Four Culture of Science. Currently we are extending this analysis to regional level analysis and hope to be able to classify European regions along a similar number of clusters.
This analysis of existing data, having curated the EB data base for our purposes, is variously significant. Theoretically, we are revisiting and revitalising older model of Triarchical attitude structures. For a characterisation of science culture, this basic MACAS 3D+ model is very useful and can be put to test on a global scale. Empirically, we are moving away from performance comparisons to structural comparison of 'science culture'. The comparison is no longer at the level any indicator, but on the correlations between the indicators for each culture unit. This constitutes a new way of looking at things. It demonstrates, that the public understanding of science needs to be looked at in the light of a set of model, one shoe does not fit all. The question will be: how many models do we need on a global scale, and what are the boundary conditions for each of them? Thirdly, for the OCED, this idea of culture indicators is new territory because it suggest to use 'subjective data' for ST&I indicators. Hitherto, these OECD indicators are entirely based on performance data (x doing y), while our suggestions are putting mood and atmosphere indicators into the centre of an innovation culture. Clearly, this created interesting discussions with economists. This work will continue in all three directions. Project MACAS has secured a book contract with Routledge Series, Science, Technology and Society, to be published in 2018.
• Suerdem A and MW Bauer (2017, in print) Construction indicators of science culture through mass media mapping [OECD seminar papers, Gent Sept 2016]
• Bauer MW and A Suerdem (2017, in print) Relation science culture and innovation [OCED seminar papers, Gent Sept 2016]
• Bauer MW, Pansegrau P, and R Shukla (2018) The Cultural Authority of Science - Comparing across Europe, India, China, Americas and Africa, London, Routledge

2 Examining the culture of science simultaneously across EU, India and China
Project MACAS has curated three cross-sectional and in some cases repeated cross-sectional data on the public understanding of science across Europe, India and China. While India was a funded partner, the Chinese colleagues became affiliated partners. Successful training sessions and exchanges of young research between the teams took place over the past four years.
The purpose of this all was to co-ordinate the analysis of all data into a similar logic and conceptualisation, as we have been developing for the OCED discussions. Data mining was applied on India and China data, including missing data analysis, imputations, IRT, EFA and CFA to stabilise the MACAS 3D+ model comprising indicators of knowledge, evaluation and engagement. It was possible to stabilise the fit this 3D+ model on all the data, which seems to suggest that indeed the triarchical attitude structure is universally portable and adequate for the data. Where the cultural element is clearly manifest is in the correlations between these three dimensions, and the clusters that emerge across the provinces. For this analysis, we treat Indian and Chinese provinces in the same way as we treat European countries.
Our approach is from the start culturally sensitive. Rather a strict model of measurement equivalence, we are relaxing this condition to configural and weak equivalence only, so that we can examine the correlations among the dimensions as our key cultural indicators. The fitting process separately in the three cultural contexts, and we are comparing the resulting correlation structures and cluster the provinces on these. The analysis reveals a set of 'science cultures', four in Europe, two in India, and Five in China (preliminary), which can be characterised by the particular correlations patterns. In line with our working principle, we are not comparing cultures on performance, but on patters of variables.
This idea and procedure of handling data has been taken up by a new research unit in China, NAIS (National Academy of Innovation Strategy). Project MACAS will continue in China creating an All China index of science culture. These efforts demonstrate that the public understanding of science needs a several models; one shoe does not fit all. The question will be: how many models are needed on a global scale, and what are the boundary conditions for each of them?
• Falade BA & MW Bauer (2017) 'I have faith in science and in God': Common sense, Cognitive polyphasia and public attitudes to science in Nigeria, Public Understanding of Science, DOI: 10.1177/0963662517690293
• Bauer MW, YY Li, BA Falade & AG Jonsdottir (2016, submitted) Fundamental attitudes to science and technology - modus DAO or modus MAO, Public Understanding of Science.
• Bauer MW (2016) Written Submission to House of Commons Science and Technology Committee (COST) enquiry on Science Communication, 29 April 2016 [3000 words, 11p]

3. Construction of media corpus of science news and an integrated database on perceptions
We started work from previous research, the British Science News Corpus, 1946-1995 (Bauer et al, 1995). The idea of project MACAS, module 2 media mapping, was to update this database to the present (up to 2015). The initial Science News corpus was manually collected and hand coded following classical content analysis. The challenge of project MACAS was to establish continuity in the database, but with entirely different updated technical means of corpus constructions: on-line text scraping, keyword selection, and dictionary based text classification as in computer-assisted content analysis. The technological progress in this area is enormous, in particular the bottle neck on data is removed, however to create equivalences between manual and machines selection, and between human and machine coding is a complicated affair and very time consuming. Project MACAS gave us a very clear idea of what is involved in doing such text-based research.
Project MACAS developed a comparable strategy for news selection applicable in Indian, German and British newspapers. This took into account the available new on-line sources and constructed a stratified sampling rationale based on a systematic sampling of a two random week (15 random days per year) over the period for a considered set of newspapers in each context. We down loaded the news texts, and adapted it to the format for software QDA Miner. We constructed dictionaries with comparable features in all three contexts and conducted classification of news into themes, actors, scientific disciplines (OECD categories) and geographical locations of news events. On the basis of this we can characterise the science news flow over time and in comparison. KW searches also allows us to construct smaller sub-corpora on specific topics for more fine-grained analysis of themes. All corpora a fully operative in QDA Miner and ready for further research.
• Bauer MW and BA Falade (2017) Science News - Steps towards computerised frame analysis, technical report, Project MACAS, module 2: mass media reporting, LSE, Department of Psychological and Behavioural Science, March 2017.
• Bauer MW, Falade BA & A Suerdem (2017) Science in the British Media, 1990-2014; Project MACAS Report Module 2: mass media mapping; LSE, Department of Psychological and Behavioural Science, March 2017

4 Examine new techniques of 'computerised text analysis, topic modelling; examining the limits of these techniques with regard to HOTIs
One of the main purposes of this project MACAS was knowledge transfer and training of colleagues and young researchers. Project MACAS has done very successful work in that direction. A multi-national team of researchers is now fully trained on computerised text analysis, including corpus construction from on-line sources, dictionary extraction, topic modelling and difficulties of analysing HOTIs [higher order text intuitions such as narrative, rhetorical features, discourse, frames]. We have examined several software packages for this purpose including ALCESTE and QDA Miner, and also R.
Project MACAS also organised the technology transfer of a prototype science news monitoring system from Brazil to Turkey, UK and Italy. Project SAPO, developed at the University of Campinas by the Brazilian partners was introduced to project members. The idea was then adapted and implemented in parallel projects in Turkey (Bilgi University, Ahmet Suredem) and at Universith of Padua (Neresini et al). These systems are up and running, and have potential for commercial exploitation.
Project MACAS has organised a series of training session on computerised corpus construction and text analysis in London, Delhi, Stellenbosch, Istanbul and Salvador in Brazil. The international team and affiliates met three times for workshops in Istanbul (Bilgi University) with invited experts on text analysis for training purposes and exchange of ideas and experiences. Ideas developed for project MACAS also went into a book publication on foundational texts of text analysis in the social sciences.
• Bauer, MW, A Suerdem, A Bicquelet (2014) Textual Analysis - Sage Benchmarks in Social Research Methods, London, SAGE (4 volumes) [ISBN 978-1-4462-4689-4]
Exploitation Route A book contract has been secured with Routledge for a Scholarly book publication on the prestigious series Studies in Science, Technology and Society. The working title of the book is 'the cultural authorithy of science'
Sectors Creative Economy,Culture, Heritage, Museums and Collections,Other

URL http://www.oecd.org/sti/097%20-%20OECD%20Paper%20attitudes%20and%20innovation_v4.0_MB.pdf
 
Description there is currently no impact on the non-academic spheres of life of which we are aware of, obviously this does not mean there is none, only we have not detected any; but this will be updated should things develop and catch our attention.
 
Title EB Biotechnology 1991-2010 
Description project MACAS curated a European database, by integrating several waves of Eurobarometer data pertaining to biotechnology from 1991 to 2010. The database contains N=138972 observation across 32 European countries on some 30 variables. 
Type Of Material Database/Collection of data 
Provided To Others? No  
Impact scholarly publications are in the making 
 
Title EB science and society 
Description project MACAS curated a European database pertaining to public understanding of science, integrating seven waves of Eurobarometer data pertaining to perception of general science from 1989 to 2013. The data base contains n=143270 observations on close to 50 variables in 32 European countries. 
Type Of Material Database/Collection of data 
Provided To Others? No  
Impact A special country report for Portugal was produced by project MACAS Bauer MW and S Howard (2014) Modern Portugal and its Science Culture - Regional and Generational Comparisons, Report prepared for Ciencia Viva, Lisbon; London, LSE Enterprise information from this database also was the basis of advice given to BIS when designing the British Attitudes to Science in 2013. 
 
Title Science news corpus UK 1990-2013 
Description A systematic sample of science news, selected on a two-week artificial week, annually 1990 to 2013. The news items are selected via NEXIS database on the keyword basis. The corpus has various versions and ranges from 4000 items to 16000 items, with various degrees of 'precision' and relevance. 
Type Of Material Database/Collection of data 
Provided To Others? No  
Impact scholarly publication are in preparation 
 
Description Culture of science in Africa 
Organisation University of Stellenbosch
Department Centre for Research on Evaluation, Science and Technology
Country South Africa 
Sector Academic/University 
PI Contribution We brought the entire MACAS team to South Africa for a 4 day workshop to CREST.
Collaborator Contribution CRIST sponsored the entire meeting 16-18 Sept 2015 bringing the MACAS network to Stellenbosch with colleagues from Europe, US, Brazil, Argentina, China, India and Taiwan. The meeting was attended by high level officials from the Ministry of Science and Research, and also one day by the SA Minister of Science.
Impact a contribution to the joint scholarly publication is in preparation
Start Year 2015
 
Description Culture of science in China 
Organisation China Research Institute For Science Popularization
Country China 
Sector Charity/Non Profit 
PI Contribution training of post-doctoral staff at CRISP and in London
Collaborator Contribution access to Chinese data
Impact will contribute to scholarly book publication on science culture in China
Start Year 2012
 
Description MACAS Germany 
Organisation Bielefeld University
Country Germany 
Sector Academic/University 
PI Contribution Internationally collaborative resarch; project co-ordination, German data collection on science news
Collaborator Contribution brought in linguistic expertise and competence in discourse analysis of science news
Impact joint book project with Routledge is in production Bauer MW, Pansegrau P, and R Shukla (2018) The Cultural Authority of Science - Comparing across Europe, India, China, Americas and Africa, London, Routledge
Start Year 2013
 
Description MACAS Italy 
Organisation University of Padova
Country Italy 
Sector Academic/University 
PI Contribution Provided access to UK data for comparative purposes; initiated the discussion and organised the meetings in London and elsewhere
Collaborator Contribution developing a prototype science in the media mapping machine for italy and other countries; contributed to conceptual and empirical discussions all along the project; participation in the meetings and training of participants.
Impact A joint scholarly book publication is in preparation
Start Year 2012
 
Description SAPO project Campinas 
Organisation State University of Campinas
Department Institute of Chemistry
Country Brazil 
Sector Academic/University 
PI Contribution gave Brazilian collagues a plartform to present their prototype of a media monitoring machine called SAPO, which crawls, extracts and counts basis science news on a continuous basis
Collaborator Contribution Stimulated our understanding of the issues of computer-assisted mass media mapping
Impact making a contribution to the scholarly publication coming out of project MACAS
Start Year 2012
 
Description Science literacy Taiwan 
Organisation National Sun Yat-sen University
Country Taiwan, Province of China 
Sector Academic/University 
PI Contribution training of staff on Taiwan and London; advising in the developement of the local database on public understanding of science
Collaborator Contribution sharing of data on public perception of science
Impact making contribution to scholarly book publication joint paper publication in the making • Bauer MW, YY Li, BA Falade & AG Jonsdottir (in preparation) Fundamental attitudes to science and technology - modus DAO or modus MAO, Public Understanding of Science.
Start Year 2012
 
Description culture of science Brazil 
Organisation Federal University of Minas Gerais
Department Department of Sociology
Country Brazil 
Sector Academic/University 
PI Contribution bringing Brazilian colleagues into the network of MACAS researchers; adivsing Brazilian colleagues on the development of similar data in Brazil and guiding the analysis of existing data
Collaborator Contribution fascinating case studies of work in Brazil, new idea for measurement of culture of science in complex circumstances
Impact making scholarly contributions to planned book publication
Start Year 2012
 
Description science and society Switzerland 
Organisation University of Lausanne
Department Faculty of Social and Political Sciences
Country Switzerland 
Sector Academic/University 
PI Contribution bringing Swiss colleagues into an international network of comparative research on science culture; bring able to advise Swiss collagues on developing similar data streams
Collaborator Contribution contributing statistical expertise into the project and mobilising the WWS with relevant data for project MACAS
Impact making a contribution to a scholarly publication
Start Year 2012
 
Description Stellenbosch University workshop 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Third sector organisations
Results and Impact This was an additional day of public engagement. Participants at the 3-day technical workshop, presented results and observations on their research in front of a larger audience for general discussion and information. This created a lively discussion among participants during the entire day.
Year(s) Of Engagement Activity 2015