I describe a series of seven experiments investigating how undergraduate students' comprehension of 2x2 `interaction' bar and line graphs widely used to present data from two-way factorial research designs is affected by both the graph format and the nature of the interaction with them. The first four experiments investigate how different Gestalt principles of perceptual organization operate in the two graph formats and demonstrate the effects of these principles (both positive and negative) on graph comprehension. In particular, Gestalt principles are shown to hinder significantly students’ comprehension of data presented in line graphs compared to bar graphs and that the patterns of errors displayed by students are systematic. The analysis also informs the development of two modified line graphs, one of which improves data interpretation significantly to the level of the bar graphs. The final three experiments investigate more deeply how the processes involved in different types of interaction with graphs affect users’ comprehension of the data depicted. In the first four experiments, participants attempted to understand the graphs while thinking aloud. However, a subsequent study (Experiment 5) demonstrated that writing an interpretation produced significantly higher levels of comprehension for line graphs than when thinking aloud. The final two experiments sought to identify the cause of this difference by isolating demands specific to the verbal protocol condition. The results of this research show that (a) in certain circumstances the Gestalt principles of perceptual organization that operate in different graph formats can significantly affect the interpretation of data depicted in them but that (b) these effects can be attenuated by the nature of the interaction. The implications of this research are that identifying an appropriate method of interaction as well as ensuring appropriate display design ensures that the majority of users will be able to interpret these graphs appropriately and so recommendations can be made for graph use in educational settings.

In an effort to document and monitor the scale and scope of recent conflicts (1998–2004) in Democratic Republic of Congo (DRC), the International Rescue Committee (IRC), in conjunction with some of the world’s leading epidemiologists, conducted a series of five surveys in the country over a seven years’ period (2000–2007). Estimates of conflict-related mortality generated from the IRC’s surveys range from 3.3 million between years 1998 and 2002, to 5.4 million excess deaths for the period between 1998 and 2007. Reflecting on the IRC’s work, the current study combines four different data sources – 1984 DRC Population Census, 1995 and 2001 DRC Multiple Indicator Cluster Survey (MICS) and the 2007 DRC Demographic and Health Survey (DHS) – to derive demographic estimates and assess the extent of population loss associated with the conflict period between 1995 and 2007. Both statistical and demographic techniques are relied upon for this purpose. Findings from this study do not warrant estimates produced by the IRC. The IRC approach may have overestimated the scale of excess deaths associated with the 1998 – 2007 armed conflict period. Because the approach used in the current study is mainly based on selected assumptions, a level of uncertainty is expected to be associated with the derived estimates. For this reason, sensitivity analyses have been conducted to define a range of plausible estimates representing the excess population loss.

This thesis examines how bank market structure affects industry performance in developing countries. A high degree of bank concentration would be associated with tight constraints and high borrowing costs, while it has also been argued that, it would be easier for firms to access credit if the banking system is concentrated. Foreign banks are seen to promote financial development and spur economic growth; while critics suggest that a larger foreign bank presence in developing countries is associated with less credit to the private sector. Also, government ownership of banks is responsible lower economic and slow financial development, while others argue that government banks promote long-run growth. The implications of bank market structure on the real economy are examined using cross-country, cross-industry panel data from developing countries, along with a variety of econometric techniques, and standard measures of industry performance. The research aims to ascertain whether bank market structure in developing countries influences financing for firms differently as a result of industry-specific characteristics. It also examines if institutional characteristics helps in explaining industrial performance in the short-run. As a follow-up to one of the findings, the research examines if banks would prefer to fund innovative firms in a liberalized environment by exploring the impact of financial development on the export structure. The main empirical findings are as follows: first, it may not be possible to identify robust or consistent findings concerning the effects of good institutions; secondly, it might not necessarily be the case that financial development specifically benefits firms based on specific industry characteristics; and finally, the research finds that banking sector development reduces export sophistication and increases export concentration. This may suggest that banking sector development enforces specialization according to existing comparative advantage.

Psychiatric distress, smoking and excessive alcohol consumption are common health problems which often occur together and are patterned by socioeconomic position. Smoking and drinking behaviours and mental health problems tend to develop over the transition from adolescence to adulthood, so this thesis aimed to investigate the mechanisms by which socioeconomic factors influence their co-development during this stage of life as young people make transitions into adult social roles. Data were primarily taken from three UK cohort studies (two nationwide birth cohorts respectively born in 1958 and 1970, and a cohort of adolescents from Glasgow who were also born in the early 1970s), so it was possible to examine whether mechanisms were context-dependent. Additional data from the youth sub-sample of the British Household Panel Study allowed investigation of socioeconomic inequalities in early adolescent smoking development in more recent history (1994-2008). A combination of person and variable centred analysis techniques (latent class analysis, structural equation modelling, propensity weighting, and event history analysis) were employed to investigate the role of socioeconomic background and transitions to adulthood in development of smoking, drinking and psychiatric distress in adolescence and early adulthood. Inverse probability weighting and multiple imputation were employed to account for missing data. A strong association was identified between socioeconomic disadvantage and adolescent smoking, despite recent increases in tobacco control in the UK. Smoking appeared to be an important mechanism, or at least a marker for other mechanisms, linking socioeconomic disadvantage to heavier drinking, psychiatric distress, and early school-leaving. Aside from smoking, there were other mechanisms leading to heavy drinking and psychiatric distress. For psychiatric distress, these were still mainly associated with socioeconomic disadvantage, especially in early adulthood, whereas for drinking there were mechanisms associated with socioeconomic advantage. Participation in tertiary education appeared to be an important example of such a mechanism, linking socioeconomic advantage to heavier drinking in early adulthood. Remaining in education was strongly linked to delaying other adulthood transitions, but different patterns of early transitions exhibited different associations with smoking, drinking and distress in different contexts. Tackling inequalities in smoking may help reduce inequalities in drinking and distress in adolescence and early adulthood, and policies increasing access to tertiary education should consider the deleterious effects on drinking behaviours.

Structural inference for Bayesian networks is useful in situations where the underlying relationship between the variables under study is not well understood. This is often the case in social science settings in which, whilst there are numerous theories about interdependence between factors, there is rarely a consensus view that would form a solid base upon which inference could be performed. However, there are now many social science datasets available with sample sizes large enough to allow a more exploratory structural approach, and this is the approach we investigate in this thesis. In the first part of the thesis, we apply Bayesian model selection to address a key question in empirical economics: why do some people take unnecessary risks with their lives? We investigate this question in the setting of road safety, and demonstrate that less satisfied individuals wear seatbelts less frequently. Bayesian model selection over restricted structures is a useful tool for exploratory analysis, but fuller structural inference is more appealing, especially when there is a considerable quantity of data available, but scant prior information. However, robust structural inference remains an open problem. Surprisingly, it is especially challenging for large n problems, which are sometimes encountered in social science. In the second part of this thesis we develop a new approach that addresses this problem|a Gibbs sampler for structural inference, which we show gives robust results in many settings in which existing methods do not. In the final part of the thesis we use the sampler to investigate depression in adolescents in the US, using data from the Add Health survey. The result stresses the importance of adolescents not getting medical help even when they feel they should, an aspect that has been discussed previously, but not emphasised.

This thesis provides an additional perspective of the Merger Paradox, namely that mergers and acquisitions (M&A) continue to be transacted when historically their results seem to be disappointing overall. The thesis shows that when a theoretically sound basis (related to the Resource Based View and expressed as twelve design principles) is used to design a performance measurement framework, then there is no association between a firm's post-acquisition performance and the scale of a firm's previous acquisitions; the thesis then shows, by contrast, that there is a positive association between firms with an above-average level of past acquisitions (by value) and higher financial performance. This divergence provides both a motive and an ability to continue to undertake M&A, despite a lack of association of acquisitions with longer-term operational performance and very strong evidence of diseconomy of scale in the most crucial business process, for the case examined, which is the research and development (R&D) process in the research-based pharmaceutical sector. Additionally, the thesis examines the relative merits of Return on Sales and Return on Assets as financial metrics of performance, and establishes statistically significant differences in the measurement of performance by these two metrics. The thesis also establishes a contrast between the findings at the level of the firm and at the level of the sector, namely acquisitions considered in aggregate are associated with gains at the sector level, even though this association was not observed when acquisition was considered at the level of the acquiring firm. The thesis provides a new application of Data Envelopment Analysis and establishes a scale efficiency relationship for the pharmaceutical R&D process. A further empirical contribution is the examination of the statistical distribution of acquisitions in the pharmaceutical sector and confirmation of the consistency of that distribution with a power-law.

Various models explaining micro knowledge-generating behaviour (in particular exporting and innovating) in the economics literature are underpinned by the overlapping assumption that these activities are largely determined by the resources/capabilities possessed by firms. Despite their perceived importance, there is a dearth of evidence on how these heterogeneous resources and firm-specific capabilities can be incorporated into economics models to quantify their roles in determining microeconomic behaviour. Therefore this thesis attempts to bridge this gap in the literature by integrating the resource-based view (RBV) as a new IO theory into the microeconomics literature and empirically utilising micro level data to investigate the significance of such resources/capacities in determining exporting and innovation activities, moderating their inter-relationships as well as conditioning their impacts on the firm’s performance. These heterogeneous resources have been proxied using firm size, productivity, capital intensity, intangible assets, various dimensions to absorptive capacity, the deployment of R&D sourcing strategies and so on. Using establishment-level data covering all UK market-based sectors in 2000, the findings show that all these factors have a large impact upon the propensity and/or intensity of establishments’ exporting and/or R&D activities, with an especially noticeable role in breaking down entry barriers to undertaking such activities. Given the significant impact of exports on knowledge-creating R&D activity, the thesis subsequently investigates and confirms additional learning effect of exporting as embodied in the firm-level exports-productivity relationship using a nationally representative panel dataset covering both manufacturing and services sectors in the UK, for the 1996-2004 period. Lastly, this thesis also attempts to provide an initial inspection of the contribution of innovation (proxied by R&D stock) to productivity using plant-level panel data for Northern Ireland. Based on the estimation of a ‘knowledge production function’ separately for various manufacturing industries, the overall long-run results show that R&D stock does have a positive impact upon productivity.

This thesis seeks to explore everyday aspects of time. Traditionally, the psychological study of time has been defined as the processes by which a person adapts to and represents temporal properties in order to synchronise external events. On the other hand, a good understanding of time is also vital when it comes to occupational and social organisation. How should time be considered across psychology remains an open question. While time perception is an established field in cognitive psychology, previous research has often focussed on either the perception of very short time intervals (milliseconds), or psychobiological effects of celestial time cycles (e.g. sleep/wake cycle or seasonal affective disorder). However, there remain several other aspects of time that while categorically different are no less important for example, 'mental time travel' or chronesthesia is the ability to mentally project into the future or past. While these phenomena are well acknowledged, it is only in the last few decades that research has started to document other 'higher level' cognitive processes that exist beyond traditional psychophysical constructs. By combining a range of experimental and secondary data analysis methodologies, this thesis examines the relationship between everyday units of time and systematic changes in behaviour across socially derived time cycles (i.e. the calendar week and the working day). It also considers the effects of individual differences on aspects of interpersonal organisation (e.g. punctuality and watch wearing). The main findings indicate that research into psychological time can and should go beyond minutes and seconds as present-day cognitive models are inadequate when it comes to accounting for everyday time processing errors. In addition, understanding the mechanisms behind higher-level timing processes may only become apparent if the topic makes a concentrated effort to become integrated with day-to-day cognition and behaviour. The results also have several applied implications including practical recommendations for optimising appointment systems in the National Health Service. Finally, these findings are discussed in relation to the ongoing debate regarding where psychological time research should focus future efforts if it is to maintain its current momentum from a theoretical and applied perspective.

This thesis makes use of recently collected UK Community Innovation Survey data to investigate 3 areas that allow to characterise and thus understand more clearly the innovation process in the UK. Firstly strategies of innovation used by firms are identified. Next the determinants of innovation, that is factors driving innovation inputs and outputs, are estimated. Thirdly this work examines the effectiveness of financial public support towards innovation. This also allows to establish which firms are more likely to be in receipt of public support and thus whether government innovation policy is in line with its objectives. Furthermore in this thesis a measure of absorptive capacity for the CIS is created, to see whether this proxy contributes in explaining innovative activities and the receipt of public support towards innovation. Similarly a measure of appropriability is generated for use as an explanatory variable in the estimation of the determinants of innovation. Both of these measures permit to find out if their latent variables have nonlinear effects in explaining propensity and extent of innovative spending. All these aspects have not received attention in previous literature, in large part due to the novelty of the data used. Besides the empirical evidence gained on the above, the addition to the literature of this thesis lies in examining several CIS survey rounds together. For one this serves as a robustness check for the conducted applications and on the other hand it allows investigating the comparability of the survey rounds. For this work the CIS 4, the CIS 5 and the CIS 6 are used as they are the most similar and comparable samples of UK businesses to date. Nevertheless it was found that differences in terms of design, wording and exclusion of responses to some question sets in the different surveys impedes their use for trend analysis and panel data analysis. Something the data collecting agencies need to address in the future. Despite these issues the conducted investigation has provided useful insights into innovation as it takes place in the UK. The first empirical chapter has been able to identify two major modes of innovation as captured by the survey. A ‘traditional’ or ‘linear’ strategy aimed at introducing product and process innovations, relying on innovative activities such as R&D and also making use of sources of information, more strongly from market sources then from science sources. Secondly a ‘dynamic’ or ‘systemic’ strategy also involving innovative activities such as R&D but more strongly making use of knowledge sources from science as well as relying on cooperation. The interpretation of this “blue skies strategy” which is not directly linked to achieving technological outputs is that it generates knowledge that helps to keep abreast of market developments and to be ready to spot opportunities in line with the literature on dynamic capabilities thus the identified strategies allow for a plausible interpretation congruent with innovation theory. In this chapter the aforementioned measure of appropriation and absorptive capacity were also successfully generated. These were then shown to play a significant role in explaining innovative activities in the subsequent empirical chapter, both exhibiting decreasing returns to scale. Following the CDM methodology this work has confirmed that knowledge capital as proxied by predicted R&D spending intensity is as important in generating service innovations as it is in stimulating goods innovations for the UK. The results also show that absorptive capacity not only indirectly impacts the likehood of introducing service innovations through its effect on knowledge capital as for goods innovations but also directly. This suggests that services once conceived further have to be tailored to individual customer’s needs. Hence absorptive capacity is specifically important in a developed economy dominated by service sector industries. At the same time the fit of the models confirmed that the CIS could do better at explaining service and process innovations by soliciting more information that are likely to cause these types of innovation. Finally further support for the innovation productivity nexus has been found. The last empirical chapter then established that absorptive capacity is also an important factor explaining the likehood of firms to be in receipt of financial public support towards innovation. This chapter further concluded that the financial public support towards innovation in the UK has in the recent past been effective at stimulating innovative performance besides just R&D spending. The government’s objective of supporting start-ups, that potentially face difficulties in financing their innovative activities, as well as supporting cooperation, vital for the dissemination of knowledge in the economy, is met according to the results. However SMEs could not be shown to be statistically more likely to be in receipt of public support despite facing the same problems as start-ups, though at least they are not less likely to be in receipt of public support then large firms. This finding stipulates that policy objectives are not achieved with regard to specifically targeting SMEs.

The 16 publications included in this thesis are the results of a programme of research between 1993 and 2009 into the labour market and labour market-related impacts of the large-scale spatially concentrated losses of industrial jobs in Great Britain from the 1970s to the 1990s. The British conventional wisdom has been that labour market recovery was relatively quick, and that the effects were not particularly profound. Continuing labour market distress was mainly ascribed to labour ‘supply-side’ factors rather than to locally deficient labour demand. The research challenges these views. It draws particularly on the British Keynesian tradition, and on authors such as J. F. Kain, John Kasarda and William Julius Wilson from the USA, which experienced similar job losses around a decade earlier. Issues covered include the statistical measurement and spatial variation of unemployment and related economic disadvantage, unemployment disguised as sickness, long-term unemployment, migration and lone parenthood, and there is also analysis of policies on employment and social inclusion. The research shows that ‘Travel-to-Work Areas’ (TTWAs) do not correctly identify the employment ‘fields’ of residents of areas of high unemployment. They have biased errors due to imbalance between commuting inflows and outflows, and obscure the main variation in unemployment on the urban-rural dimension. Three papers on Incapacity Benefit (IB) analyse the dynamics of change in the stock of claimants, investigating the roles played by health status and labour market conditions. The most recent of these papers examines whether the striking fall in IB claims in Glasgow and other former industrial areas in 2003-08 was the result of official interventions or of improving labour market conditions, concluding that it was mainly the latter. A key ‘supply-side’ assumption was that being unemployed in itself makes people less ‘employable’ – the theory of ‘state-dependence’. The paper on long-term unemployment radically challenges this interpretation. It points out that the literature on the relationship between long-term and short-term unemployment has generally failed to consider the appropriate time-lags or the behaviour of the standard measure of long-term unemployment. It shows that the phenomenon which the theory of state-dependence purports to explain does not occur to any significant extent. Outmigration and housing abandonment are significant effects of local job loss. The paper on housing abandonment demonstrates a statistical relationship across England in 1997 between social housing surplus and ‘real unemployment’, while a further paper challenges the view that there was no longer a deficiency of demand for labour in Glasgow and the Clyde Valley in the 1990s by investigating migration patterns. It demonstrates that net flows between individual Scottish areas and the rest of the UK were to a substantial extent determined by changes in labour demand. A new finding is that little adjustment to employment change occurs through migration within Scotland. The large increase in lone parenthood in Great Britain since the 1960s has been strongly correlated across areas with male worklessness. The US literature suggested that this relationship is causal, and this thesis is investigated in two papers. The earlier of these was the first comprehensive published application of this interpretation to the modern British case. A further paper concludes that falling male employment accounted for around half the rise in lone parenthood in Great Britain in 1971-2001. Two of the papers present a comprehensive picture of the geographical distributions of the different groups of disadvantaged people in the labour market, showing that they all conform to a similar pattern which in turn is related to deficient labour demand.