Linked datasets provide researchers with the opportunity to assess the limitations of administrative data and develop quantitative methods of analysis that can overcome such limitations... more

 

Linked datasets provide researchers with the opportunity to assess the limitations of administrative data and develop quantitative methods of analysis that can overcome such limitations. The projects are primarily concerned with areas of research where administrative data are currently being used without full understanding of the potential biases being introduced by the weakness of administrative data in particular respects, namely missing and limited covariates.

PI: Anna Vignoles

ADMIN homepage

Linked datasets also provide researchers with the opportunity to assess the limitations of survey data, particularly panel survey data, and develop quantitative methods of analysis that can overcome problems commonly associated with survey data, such as attrition, non-response and measurement error... more

 

Linked datasets also provide researchers with the opportunity to assess the limitations of survey data, particularly panel survey data, and develop quantitative methods of analysis that can overcome problems commonly associated with survey data, such as attrition, non-response and measurement error. Our proposed research includes four projects that will focus on improving survey data by relying on linked datasets to address important topical policy questions.

PI: James Brown

ADMIN homepage

Non-response in surveys can arise for many different reasons, and biased inference can result if factors associated with non-response are also associated with the question under study... more

Non-response in surveys can arise for many different reasons, and biased inference can result if factors associated with non-response are also associated with the question under study. Researchers at BIAS II are developing different approaches to modelling non-response bias. In one approach, we use graphical models to develop a type of influence diagram representing both the process under study and the potential causes of non-response. The conditional independence assumptions encoded by such diagrams allow us to separate the process of non-response from the mechanism of inferential interest. We can then develop biased-corrected estimates (e.g. using post-stratification strategies) of the parameters of interest by conditioning on appropriate variables separating the parts of the graph. Our second approach borrows ideas from the propensity score literature to model non-response bias using latent variables. Instead of directly introducing variables responsible for non-response in a graphical model, we model the effect of the bias globally through a latent "bias parameter" which is used to adjust the parameter estimates of interest. As part of this work, we are investigating, within the Bayesian framework, how to incorporate external information, e.g. via a validation data set, in the specification of the prior distribution of this bias parameter.

Duration: July 2008 - March 2011

People working on the project:

Node staff: Dr Sara Geneletti, Prof Nicky Best, Prof Sylvia Richardson Collaborators: Dr Lawrence McCandless (Simon Fraser Uni), Prof Paul Gustafson (UBC)

BIAS homepage

Researchers at BIAS II are developing Bayesian space-time models to characterise stability and estimate change over time in small area indicators... more

Researchers at BIAS II are developing Bayesian space-time models to characterise stability and estimate change over time in small area indicators. Our methods are designed to distinguish random temporal fluctuations in stable areas from areas in which real change has occurred. Our approach is to include different types of space-time interaction terms in the basic small-area model with spatial and temporal main effects, with a particular emphasis on choosing an appropriate hierarchical structure for these terms that facilitates classification of areas/time periods as 'predictable' (i.e. smoothly changing) or 'abruptly changing'. Together with our collaborators at the Office for National Statistics (ONS), we are using these space-time models to identify time trends and areas with low levels of income and employment, poor health and housing conditions, using small area data from e.g. General Household Survey, the Integrated Household Survey and the Family Resources Survey. BIAS II researchers are also collaborating with Prof Bob Haining at Cambridge University to apply our methods to analyse the space-time pattern of criminal offences in Cambridgeshire. We are investigating how stable patterns are and whether there is evidence of repeat victimisation and "spatial" repeat victimisation (where an offence such as burglary does not occur in the same household but within some radius). In addition, we are trying to detect sudden increases in the crime activities by modelling space-time interactions.

Duration: July 2008 - June 2011

People working on the project

BIAS II staff: Prof Nicky Best, Prof Sylvia Richardson Collaborators: Mr Philip Clarke (ONS), Prof Bob Haining (Cambridge)

BIAS homepage

Researchers at BIAS II are collaborating to develop and apply a Bayesian hierarchical modelling approach for the synthesis (meta-analysis) of multiple cross-national longitudinal datasets... more

Researchers at BIAS II are collaborating with Prof Scott Hofer at Oregon State University USA, who directs a large international collaborative research network on longitudinal studies of ageing - IALSA (http://lifelab.tss.oregonstate.edu/) - to develop and apply a Bayesian hierarchical modelling approach for the synthesis (meta-analysis) of multiple cross-national longitudinal datasets. Our basic synthesis model assumes a common set of variables measured in all studies and has a hierarchical structure with random effects capturing differences between, e.g. countries, studies with different population bases etc. Our model can then be elaborated to include constructs and covariates that are measured differently across the studies, by relating these to underlying latent variables, and to include several outcomes simultaneously. Substantive questions that BIAS II researchers are addressing using these synthesis models include: (a) to document general patterns of population average (between person age differences) and individual variation (within person changes in age) in change in cognitive capabilities; (b) evaluate the effects of age, education, and sex on intra-individual trajectories of cognitive outcomes across studies; (c) identify/describe the level and rate of change in Mini Mental State Exam at the individual and aggregate population level prior to diagnosis of dementia.

Duration: July 2008 - June 2010

People working on the project:

BIAS II staff: Dr Jassy Molitor, Prof Nicky Best, Prof Sylvia Richardson Collaborators: Prof Scott Hofer (Oregon State University)

BIAS homepage

During BIAS I, we developed hierarchical related regression (HRR) methods for combining random samples of individual level data with aggregate data on the same variables in order to reduce ecological bias and increase power compared to analyses based on a single data source... more

During BIAS I, we developed hierarchical related regression (HRR) methods for combining random samples of individual level data with aggregate data on the same variables in order to reduce ecological bias and increase power compared to analyses based on a single data source. BIAS II researchers are collaborating with Dr Steve Fisher at Oxford University to apply these HRR methods to analyse electoral behaviour data. One question we are addressing concerns ethnicity and vote choice. The vote choice of ethnic minorities in Britain is hard to estimate with opinion polls, or even with the British Election Study (BES) surveys, because the sample sizes are too small to yield sufficient numbers of ethnic minorities. We are using HRR models to combine the individual level BES data with aggregate census data, which includes ethnicity and religion, and election results for parliamentary constituencies, to improve estimates of the strength of association between ethnicity and vote for the 2001 and 2005 general elections and thereby assess the extent to which there was a realignment of British ethnic minorities away from Labour.

Duration: July 2008 - June 2010

People working on the project:

BIAS II staff: Dr Jane Key, Prof Nicky Best, Prof Sylvia Richardson Collaborators: Dr Steve Fisher (Oxford University)

BIAS homepage

This strand of the Hub's research asks 'what happens when perspectives from different disciplines come together?'... more

 

This strand of the Hub's research asks 'what happens when perspectives from different disciplines come together?' It will consider interdisciplinarity as it applies to relations between the social science disciplines, such as economics, psychology and sociology, but may also range beyond this to consider interactions with natural sciences and the humanities. Of particular interest are the fit between methods and disciplines, and the issue of the achievement of genuine engagement between approaches which compete over the same territory.

Current outputs from this strand of Hub research include

• Five Strategies for Practising Interdisciplinarity. NCRM Working Paper.
• What Are Academic Disciplines? Some Observations on the Disciplinarity vs. Interdisciplinarity Debate. NCRM Working Paper.
• Annotated Bibliography: Interdisciplinarity. Project Report.

Hub homepage

This area of work seeks to explore how and why developments or 'innovations' in methods achieve breakthrough status in terms of uptake and use in the social science community while others do not... more

 

This area of work seeks to explore how and why developments or 'innovations' in methods achieve breakthrough status in terms of uptake and use in the social science community while others do not. Specific research questions are:

• What is methodological innovation?
• What is the history of specific innovations in terms of their uptake and use?
• How are they used and adapted in different disciplines?
• How are the methods publicised and promoted; what impact has this had on their uptake?

Current outputs from this strand of Hub research include:

• Annotated bibliography

Hub homepage

In this research strand we focus on the development and extension of latent class and Markov transition models for modelling patterning of offences over time... more

In this research strand we focus on the development and extension of latent class and Markov transition models for modelling patterning of offences over time. The focus of this work in on the changing patterns of criminal behaviour - both through the life course and across time. Part of this work will attempt to disentangle age period and cohort effects. Another part will focus on the development of spatio-temporal models for reported crime.

Lancaster-Warwick-Stirling homepage

In this research strand we analyse complex multivariate data on childhood development, and investigate the relationship between domain general skills like language or executive functioning and domain specific skills like 'theory of mind'... more

In this research strand we analyse complex multivariate data on childhood development, and investigate the relationship between domain general skills like language or executive functioning and domain specific skills like 'theory of mind'. This is a complex multivariate problem which requires a graphical modelling approach through "directed acyclic graphs".

Lancaster-Warwick-Stirling homepage

In this research strand we investigate issues of changing gender roles and changing preferences to post compulsory education over time, using data from the British Household Panel Survey and the Youth Cohort Surveys... more

In this research strand we investigate issues of changing gender roles and changing preferences to post compulsory education over time, using data from the British Household Panel Survey and the Youth Cohort Surveys. Here, problems of attrition and missing data in repeated ordinal measurements over time will be addressed.

Lancaster-Warwick-Stirling homepage

Goldstein and Noden introduced a new model-based approach to the measurement of diversity by considering a multilevel model where the main focus of interest was the modelling of variation... more

Goldstein and Noden introduced a new model-based approach to the measurement of diversity by considering a multilevel model where the main focus of interest was the modelling of variation. In the case of schooling, eligibility for free school meals has been the focus of much interest. These authors used the proportion of such children in a school as their response variable in a 3-level model which explicitly included school and local education authority (LEA) effects. Thus the between-school and between-LEA variances are model parameters. In essence these parameters capture the diversity among schools and LEAs and functions of the estimates of them will correspond to different indexes that have been put forward in the literature. There are several advantages to such a modelling approach.

Extensions to this work will be to further data sets and also in the following methodological directions. Firstly, it is straightforward to extend the modelling to cross-classified models, so that in the context of schooling we can take into account neighbourhood effects and study the effects of different stages of schooling. There is current work taking place on the measurement of school catchment areas and this can be used when exploring neighbourhood effects. Likewise, pupil mobility across schools and neighbourhoods can be handled using multiple membership models.

Researchers: Harvey Goldstein, Kelvyn Jones, Simon Burgess, Rich Harris

LEMMA II homepage

In this strand we will explore three substantive questions: (i) The impact of families on pupil achievement, (ii) School competition and pupils' learning progress, and (iii) Parental selection into schools and neighbourhoods (catchment areas) driving the school competition processes... more

In this strand we will explore three substantive questions: (i) The impact of families on pupil achievement, (ii) School competition and pupils' learning progress, and (iii) Parental selection into schools and neighbourhoods (catchment areas) driving the school competition processes.

(i) The impact of families on pupil achievement.

To our knowledge, there has been no multilevel analysis of datasets which contain both multiple children per family and multiple children per school, thereby allowing estimation of separate variance components for children, schools and families. In an analysis of US data a series of separate sibling, peer, neighbourhood and schoolmate correlations has been modelled to adolescent achievement data and the largest correlation has been found to be between siblings (0.5). A full multilevel analysis of such data will properly partition these effects reducing their size. However, the sibling (family) variance component in a multilevel model of pupil attainment may well remain the most important source of variation in the system. Developmental psychological theory predicts that when families are placed under stress there will be fragmentation and differential failure of children within the family.

We will explore this hypothesis by allowing the within-family variance to be a function of socio-economic stress.

(ii) School competition and pupils' learning progress.

Using PLASC we will develop modelling frameworks and methodology to assess the evidence for school competition affecting pupils' learning progress after conditioning on the individual and contextual effects of differential school intakes. If there is a residual effect of school competition on pupils' learning progress the school-level random effects in a multilevel model (pupils within schools) will be correlated. The between-school correlations will be modelled initially in terms of a proximity function based on de facto catchment areas. For example, only schools which have overlapping catchments could be allowed to covary. The dependence between pairs of overlapping schools can then be further explored by elaborating the function for between-school covariance to include, for example, differences in school characteristics.

(iii) Parental selection into schools and neighbourhoods (catchment areas) driving the school competition processes. One hypothesis is that the main mechanism for school competition effects is parental selection into schools and neighbourhoods. To test this hypothesis we extend the school competition model in (ii) by including catchment area directly as a random classification in the model. Pupils are therefore multiple members of catchment areas and catchment areas are cross-classified with schools. If there are effects of parental selection into schools and neighbourhoods on pupils' learning progress we will find a non-zero correlation between school and neighbourhood random effects. We will model the school-neighbourhood correlation in terms of adjacency indicators and attractor mechanisms based on historical characteristics of schools and neighbourhoods. If the driving force for school competition effects is parental selection into schools and neighbourhoods, we will see a weakening of any between-school dependencies established in model (ii) and significant dependency between schools and catchment areas.

Researchers: Jon Rasbash, Fiona Steele, Harvey Goldstein, Simon Burgess and Jo-Anne Baird

LEMMA II homepage

It is well established that risky behaviours in families cluster together and that the causal relationships between these behaviours is not well understood... more

It is well established that risky behaviours in families cluster together and that the causal relationships between these behaviours is not well understood. This project will use data from the Avon Brothers and Sisters Study, which has very detailed entire family (including non-target siblings) longitudinal information on 200 of the ALSPAC families.

We will examine the directionality of relationships between parental depression, child behaviour, marital difficulties and family type (single parent, nuclear, step) in a multiprocess model. Family data provide a particularly challenging set of issues for multiprocess models because of the complex structure of families: individuals are multiple members of dyads and dyads are nested within families. The multivariate response variables are of different types (normal and multinomial) and defined at different levels. We will begin by exploring reciprocal causation between pairs of risk variables and treat the remaining set of risk variables as exogenous to the system. We can then proceed to exploring triplets of endogenous variables, thus building a realistically complex model for the transmission of risks within families. We will tackle issues of endogeneity caused by selection and reciprocal and lagged effects.

Researchers: Jon Rasbash, Fiona Steele, Jenny Jenkins, Tom O'Connor, Jonathan Evans, Carol Propper, Frank Windmeijer

LEMMA II homepage

LEMMA II previous work implements procedures that are based on methodological extensions that allow multivariate mixtures of normal, ordered or unordered categorical responses that can be defined at any level of a data hierarchy... more

LEMMA II previous work implements procedures that are based on methodological extensions that allow multivariate mixtures of normal, ordered or unordered categorical responses that can be defined at any level of a data hierarchy. The 2-level model is considered in detail and a major application is to multiple imputation for missing data. We use latent variable ideas to create an underlying set of latent multivariate normal responses: one normal response for each binary or ordered response variable and a set of normal responses for each multicategory response variable. This reduces the analysis to a multivariate normal model that allows us to apply standard algorithmic steps in the estimation.

In multiple imputation there are two models. One is the scientific model of interest (MOI) and the other is the imputation model (IM). The basic idea is that all the variables that are present in the MOI form a set of response variables in the IM which is then fitted, within a multilevel structure, with intercepts in the fixed part of the model. For a set of multivariate Normal responses this is straightforward and in addition, if any responses are missing, they will be randomly imputed within an MCMC analysis. For original non-normal variables, these imputed values are then transformed back to the original (ordered or unordered) scales so that the imputed 'complete' datasets will have all variables on their original scales.

Researchers: Harvey Goldstein, James Carpenter

LEMMA II homepage

The classic random effects multilevel model assumes independence between units of the same classification (e.g. school effects are independent) and independence between units of different classifications (e.g. school and neighbourhood effects are independent)... more

The classic random effects multilevel model assumes independence between units of the same classification (e.g. school effects are independent) and independence between units of different classifications (e.g. school and neighbourhood effects are independent). The research questions to be addressed in the 'Realistic Models for School Effects' strand require both these assumptions to be relaxed. By definition when modelling school competition, school effects cannot be independent, and also when exploring parental selection mechanisms into schools and neighbourhoods, school and neighbourhood effects cannot be independent.

We propose to develop methodology to handle both these cases. Non-independence between units of the same classification, where the classifications are at higher levels, has been handled conditional autoregressive models. We propose an alternative approach that readily extends to the second, more general, case of non-independence between random classifications. Under LEMMA we have developed models that allow for correlation between crossed random classifications where the classifications share units. Rasbash et al. fit a social relations model where bi-directional relationship measures can be decomposed into actor, partner, dyad and family random classifications. Correlations are modelled between the actor and partner random classifications, which are different mappings from the bi-directional relationship scores (level 1 units) to the same set of individuals. We propose to extend this work to allow correlations between classifications that do not share units.

Researchers: Harvey Goldstein, William Browne, Paul Clarke, Fiona Steele, Jon Rasbash

LEMMA II homepage

The Data Integration stream will evaluate and document procedures for CAQDAS-based methodological integration by (i) employing selected qualitative software packages to conduct secondary analysis of qualitative data on the social factors in response to natural environmental risk arising from climate change, (ii) comparing findings from these procedures to the statistical analysis of the quantitative data in these datasets... more

The Data Integration stream will evaluate and document procedures for CAQDAS-based methodological integration by (i) employing selected qualitative software packages to conduct secondary analysis of qualitative data on the social factors in response to natural environmental risk arising from climate change, (ii) comparing findings from these procedures to the statistical analysis of the quantitative data in these datasets. QSR NVivo, MAXqda and ATLAS.ti variously provide means of importing quantitative data and linking with qualitative datasets, converting qualitative codes into quantitative variables and allowing their export to statistical packages. We will also include the hybrid software suite that includes QDA Miner 3 which starts from a different epistemological starting points, as it offers traditional CAQDAS functions which can be used alongside enhanced quantitative approaches to the analysis of large datasets (e.g. multidimensional scaling, heatmaps, dendrograms, proximity plots). Such tools answer a wider range of research needs, often associated with policy research, public/media/academic discourse, or analysis of Internet and e-mail data.

Duration: September 2008 - 2010 Researchers: Graham Hughes, Thomas Koenig, Jane Fielding, Ann Lewins

QUIC homepage

The second MICS project relates to multi-stream visual data. Social science increasingly uses visual data, and a new networked video conferencing technology called 'Access Grid' allows people at many locations to participate in 'virtual fieldwork' or teaching sessions convened by a host site... more

The second MICS project relates to multi-stream visual data. Social science increasingly uses visual data, and a new networked video conferencing technology called 'Access Grid' allows people at many locations to participate in 'virtual fieldwork' or teaching sessions convened by a host site. This project will refine and document procedures developed in the last two years for the use of the Access Grid in primary data collection and advanced pedagogy. It will build on an ESRC e-Social Science project that delivered the world's first 'virtual fieldwork' via the Access Grid, and on an institution-funded pilot project delivering advanced software training via the Access Grid. Thus this projects aims to document how to analyse AG multi-stream visual data using CAQDAS, and deliver training via AG. The substantive testbed application will be to conduct virtual fieldwork involving staff of the Environment Agency and/or National Probation Service.

Duration: September 2008 - 2010 Researchers: Nigel Fielding, Christina Silver, Thomas Koenig, Ray Lee

QUIC homepage

The geo-referencing project will apply and evaluate CAQDAS tools that offer GIS-type functionality, via geo-referencing a crime risk assessment methodology which explores the social environmental risk arising from crime/disorder... more

The geo-referencing project will apply and evaluate CAQDAS tools that offer GIS-type functionality, via geo-referencing a crime risk assessment methodology which explores the social environmental risk arising from crime/disorder. Geo-referencing qualitative software will enable users to add a spatial dimension to qualitative data analysis. Currently users of GIS and of CAQDAS do not much intersect, yet the gains in being able to code, annotate and analytically manipulate visual representations of physical space with CAQDAS functionality are attractive. Using environmental scan methodology developed to support police/community crime audits, this stream will evaluate the affordances of GIS-type CAQDAS functionality and develop an exemplar study for the TCB component of the node's work.

Duration: September 2008 - 2010 Researchers: Jane Fielding, Nigel Fielding, Graham Hughes, Christina Silver

QUIC homepage

This project will combine empirical research and software training to develop a set of interactive web-based protocols for different practical and methodological purposes... more

This project will combine empirical research and software training to develop a set of interactive web-based protocols for different practical and methodological purposes. Alongside the existing one-day training programme, we will be running a new series of two-day training workshops in the use of leading qualitative software at which participants work with their own data. A sample of participants from each workshop will be invited to partake in a longitudinal project in which we track their use of software. Their experiences will contribute to the design of the online protocols for choosing, planning and using CAQDAS packages.

We aim to develop two sets of protocols. Firstly generic protocols to facilitate novice software users in three key areas: i) making informed and critical choices between packages; ii) planning software use and setting up projects efficiently; iii) and using software to facilitate different methodological approaches to qualitative data analysis. Secondly, software specific protocols which focus on more advanced aspects of analysis and software use derived from the MICS projects and focusing on aspects of data management and team working.

Duration: February 2009 - 2011 Researchers: Ann Lewins, Christina Silver

QUIC homepage

The analysis conducted for the data integration, visual analysis and GIS-CAQDAS projects will be tracked and documented as the research proceeds such that analytic tasks performed using particular software tools can be systematically compared and evaluated. .. more

The analysis conducted for the data integration, visual analysis and GIS-CAQDAS projects will be tracked and documented as the research proceeds such that analytic tasks performed using particular software tools can be systematically compared and evaluated. These projects will be archived and used to develop exemplar software projects available from the CAQDAS website which can be used as self-learning and/or guided teaching aids. The exemplar projects will be accompanied by explanatory materials concerning analytic, practical and technical procedures.

Duration: February 2009 -2011 Researchers: Ann Lewins, Christina Silver, Graham Hughes, Thomas Koenig

QUIC homepage

In this project we are investigating personal associations that are 'critical' in people's lives. Relationships with friends, acquaintances, or colleagues, may be significant because they are close and supportive. But equally they may be difficult or even 'toxic', and important relationships may be lost or ebb and flow over time.Our emphasis on critical associations, and the methods we are deploying to explore them, are designed to capture not only the positive and supportive aspects of significant personal associations, but also these more negative elements... more

In this project we are investigating personal associations that are 'critical' in people's lives. Relationships with friends, acquaintances, or colleagues, may be significant because they are close and supportive. But equally they may be difficult or even 'toxic', and important relationships may be lost or ebb and flow over time.Our emphasis on critical associations, and the methods we are deploying to explore them, are designed to capture not only the positive and supportive aspects of significant personal associations, but also these more negative elements.

Methodological approaches include:

• Era Memory Workshops
• Mass Observation Directive
• Ethnographic interviewing

Critical Associations project homepage

In this project we are exploring dynamics between and within generations, and how generation itself is experienced. Our focus is onhow older people live out their relationships with other generations, and whether and how they themselves identify with a generation... more

In this project we are exploring dynamics between and within generations, and how generation itself is experienced. Our focus is onhow older people live out their relationships with other generations, and whether and how they themselves identify with a generation.

Unlike most research in this area, we are not confining our investigations to familial geneartional relationships, because we think that older people's interactions with non-family members of their own and other generations are highly significant to their quality of life and experiences of ageing.

We are combining qualitative and quantitative approaches, using secondary analysis of data from a major national survey, the English Longitudinal Study of Ageing (ELSA) and a new, linked, qualitative study with sub samples of ELSA participants and their younger friends, acquaintances and relatives.

Our methodological approach involves four linked elements:

• Secondary analysis of ELSA data including: mapping of inter/generational family and friendship networks, their positive and negative qualities and the nature of exchanges they involve; cluster analysis
• Linked qualitative study of inter/generational engagements in a sub-sample of up to 35 ELSA respondents using a range of methods including
• biographical and retrospective interview methods
• photo, sensory and object elicitation
• participatory/creative mapping of relationships
• Further qualitative study of up to 25 significant "younger generations" of our ELSA sample.
• Socio-cultural/historical contextual analysis of changing inter/generational "climates"

Inter/generational Dynamics project homepage

The last 60 years of research into repeated interaction, explored in games such as the Prisoner's Dilemma, has produced many theories. Some are based around an idealised capacity for individuals to maximise their payoff, assuming access to complete knowledge... more

The last 60 years of research into repeated interaction, explored in games such as the Prisoner's Dilemma, has produced many theories. Some are based around an idealised capacity for individuals to maximise their payoff, assuming access to complete knowledge. More recent ideas stem from the idea of a limited rationality, where bounded rules operate with limited memories. Although unbounded rationality appears very unlikely, whether or not real players follow bounded rule-sets remains an open question.

Whatever the theory, and regardless of its apparent plausibility, its scientific merit rests on whether or not it accounts for data taken from human interactions. We undertake an in-depth investigation into historically important approaches in the literature and compare it to available data. In this way, we shift focus away from a question that pervades much of the current literature, i.e., 'which rule-based strategies do real players employ?' and we ask 'how do the unfolding dynamics of rule-based strategies compare with the unfolding dynamics of real players?' In this way we can begin to re-evaluate the influential, computational approach taken over the last 30 years, and how it might or might not relate to how real people behave. If players use such rules, which particular forms are representative? If such rules do not reproduce data, then which alternative representations should we explore?

These are difficult issues, but are fundamental to progress in the area of the problem of cooperation. We therefore develop a review of existing approaches, both modern and more classical, with the criterion of relevance to real data in mind. We aim to motivate research in the area of repeated interaction by refocusing efforts along the lines originally provided by the pioneering researchers of the 1950's and 1960's. Their methods have been largely ignored by modern research, and we attempt to readdress their role within models of repeated human interactions.

Richard Holden

SIMIAN homepage

"Radical innovation" refers to innovations that are not quantitative extensions of existing technologies (bigger, faster, cheaper ones) but rather qualitatively new. Their value, meaning and effects are unforeseeable... more

"Radical innovation" refers to innovations that are not quantitative extensions of existing technologies (bigger, faster, cheaper ones) but rather qualitatively new. Their value, meaning and effects are unforeseeable. They are most noticeable by their impact on existing technologies, which they do not so much surpass at previous tasks as render them irrelevant or obsolete. Genuine novelty emerges suddenly from the interplay of interdependent, interacting parts.

How is such radical innovation produced? Through relatively small-scale processes:

• Tinkering - trying out combinations of existing practices and technologies
• Brokerage - communicating with people in other groups, organisations or communities
• Reflection and reaction - challenging parts of the current structure, perhaps motivated by the need to differentiate oneself following conflict with its existing leaders.

The fact that novel combinations of known components, imported ideas in novel contexts, and departures into novel territory can have radical, unforeseeable effects - both positive and negative - is what makes "radical innovation" worth understanding.

Why do we need computer simulation models to study this concept? Other forms of research cannot deal with "radical innovation". Qualitative approaches such as ethnographic studies can capture complexity in micro-events, but they are difficult to scale up. Quantitative methods using statistical analysis can deal with large scale, but struggle with complex interdependent parts, non-linearity, and feedback loops. Simulation modelling can support the aims of qualitative researchers and extend their findings, using some of the formalisation and rigour of quantitative methods. It alone seems well placed to represent the phenomena of genuinely novel innovations.

Christopher Watts

SIMIAN homepage

Simulation has been a ground breaking methodological innovation. It is often labelled the 'virtual laboratory of the social sciences' as it allows for the controlled variation of parameters, identification of mechanisms and testing of theories similar to laboratory experiments in the natural sciences... more

Simulation has been a ground breaking methodological innovation. It is often labelled the 'virtual laboratory of the social sciences' as it allows for the controlled variation of parameters, identification of mechanisms and testing of theories similar to laboratory experiments in the natural sciences. The major research focus of this strand of SIMIAN is to show how this innovative methodology can shed light on the 'Paradox of Agency': How are social norms possible without losing agent autonomy?

Nowhere is the Paradox of Agency more apparent than in the analysis of criminal behaviour. The purpose of this research strand is to identify and where possible fill gaps in the modelling of criminal behaviour. A major focus is the different modes of explanation of criminal behaviour as environmentally conditioned activity (external) versus criminal behaviour as individually chosen activity (internal).

This resulting book will enrich social science and philosophical debates on normative behaviour by critically assessing existing models of criminal behaviour and contributing a new internal perspective. It will thus fulfill the twofold task of assessing simulation as a method for social science research and enriching the methodology with a newly developed internal model of criminal behaviour focused on moral reasoning.

Corinna Elsenbroich

SIMIAN homepage