PSCI2701 PDF - Public Affairs Analysis

Chapter 1: Summary: Definition and Overview Public affairs is defined as a new field of study that integrates elements from older disciplines while addressing broader concerns beyond traditional academic boundaries. It is characterised more as a set of skills applicable in various academic and profe...

Chapter 1: Summary: Definition and Overview Public affairs is defined as a new field of study that integrates elements from older disciplines while addressing broader concerns beyond traditional academic boundaries. It is characterised more as a set of skills applicable in various academic and professional contexts rather than a rigid academic discipline. The field focuses on activities that reflect or shape human interactions aimed at acquiring or preserving valued resources, acknowledging that almost all human activities have public implications, though not all should be subject to government intervention. The scope of public affairs analysis includes various activities such as analysing behaviour and opinions, designing communication campaigns, formulating public policy initiatives, and producing research reports to enhance understanding of public issues. These activities can serve both applied and academic purposes, with practitioners often engaging in both roles. Analysis and Participation: Dimensions and Roles Public life encompasses both analytic and participatory roles, with individuals varying in their levels of engagement and understanding of public issues. The spectrum of participation ranges from passive citizens with limited knowledge to informed individuals actively involved in public affairs. This categorization highlights the diverse ways people engage with public life, from casual voters to generalist leaders who must navigate multiple public issues. The generalist public affairs consulting community plays a unique role, providing advice and analysis across various public issues, often merging analysis with participation. This community is not confined to a single issue, requiring a broad understanding of multiple topics. Current Problems in Analysis and an Alternative Approach The text critiques the current state of public affairs analysis, suggesting that many analysts have become entrenched in dogmatic views, akin to secular theologians. This has led to a lack of respect for academic institutions and a tendency to promote specific beliefs rather than objective analysis. Good public affairs analysis should be characterised by sound empirical theory, structured methods, and a multidimensional understanding of public good, aiming to aid all participants in public life rather than promote particular causes. Analysts face ethical challenges when their work is influenced by clients with specific agendas. It is essential for analysts to provide a comprehensive understanding of issues, highlighting various perspectives, even if clients may choose to ignore certain findings. The text emphasises the need for public affairs analysts to advocate for changes in public institutions to foster a more understanding-driven approach to public decision-making. Key Points: 1. Definition of Public Affairs: Public affairs is a new field of study that combines elements from various disciplines and focuses on the analysis and participation in public life. It emphasises the importance of understanding the public implications of human activities, which can extend beyond governmental concerns. 2. Role of Analysts: Public affairs analysts are expected to engage in sound empirical theory and structured methods to analyse public issues. Their work should aim to provide a comprehensive understanding of public phenomena rather than merely rationalising personal beliefs or biases. 3. Spectrum of Participation: The chapter outlines a spectrum of participation in public life, ranging from passive citizens with limited knowledge to highly informed individuals actively engaged in public affairs. This spectrum highlights the varying degrees of analysis and participation among individuals. 4. Current Challenges: The chapter discusses the challenges faced by public affairs analysts, including the tendency for some to adopt dogmatic views similar to secular theologians. This has led to a lack of respect for academic institutions and a need for analysts to provide objective insights rather than promoting specific agendas. 5. Ethical Responsibilities: Public affairs professionals have an obligation to promote a redefinition of roles within public life and to advocate for changes in public institutions that foster understanding rather than adversarial conflict. This commitment to change should not lead to discriminatory practices within the field. 6. Activities of Public Affairs Analysts: Typical activities in public affairs analysis include studying elite and mass behaviour, designing communication campaigns, formulating public policy initiatives, and producing research reports to enhance public understanding of issues. 7. Need for Change: The chapter emphasises the importance of modifying public institutions to create a framework that prioritises understanding and collaboration over conflict, which is essential for effective public affairs analysis. Answers to the questions: What are the similarities and differences between typical quantitative research and non-quantitative historical/contextual research? Similarities: 1. Focus on Public Issues: Both typical quantitative research and non-quantitative historical/contextual research aim to analyse public issues and provide insights that can aid in understanding choices, decisions, and problems within public life. 2. Analytic Intent: Both approaches involve an analytic dimension, where researchers seek to understand and interpret data or historical contexts to inform public affairs. Differences: 1. Methodological Approach: Quantitative research typically relies on structured methods and statistical analysis to derive conclusions, while non-quantitative historical/contextual research may involve more narrative and interpretive techniques, focusing on the context and meaning behind events rather than numerical data. 2. Nature of Data: Quantitative research often deals with numerical data that can be measured and analysed statistically, whereas non-quantitative research may utilise qualitative data, such as interviews, historical documents, and case studies, to provide a richer understanding of public phenomena. Chapter 2: Summary: The quantitative approach in public affairs is characterised by systematic stages of inquiry, resembling the scientific method, and is defined as a structured method for analysing information. It involves several key stages: defining a research problem and hypothesis, selecting a research design, addressing measurement issues, collecting data, processing data, analysing data, and reporting results. While these stages may not always be followed in a strict order, they provide a framework for organising quantitative research effectively. Research Types: Confirmatory vs. Exploratory Quantitative research can be categorised into confirmatory and exploratory types. Confirmatory research focuses on testing specific hypotheses, while exploratory research seeks to identify interesting patterns or relationships within data without predefined questions. Both types can utilise the same stages of analysis, but exploratory research tends to be less structured. Purposes of Quantitative Analysis The primary purposes of quantitative analysis in public affairs include explanation, prediction, understanding, and description. Additionally, it allows researchers to systematically summarise large amounts of information, define complex relationships, and make informed decisions about accuracy and precision. Quantitative methods also facilitate the assessment of assumptions underlying the analysis, contributing to the validity of the findings. Limitations and Evolution of Quantitative Methods Quantitative analysis is not infallible; results and techniques are subject to modification and evolution over time. The methods used are conventions that may change as new techniques emerge. This malleability raises questions about the objectivity of quantitative methods, which are better understood as intersubjective conventions rather than unchanging truths. Historical Context and Non-Experimental Designs Most public affairs analysis, whether quantitative or qualitative, is rooted in historical context. Researchers often rely on non-experimental designs, analysing data that has already occurred rather than manipulating variables in controlled experiments. This reliance on historical data presents both strengths, such as authenticity, and weaknesses, including potential biases and assumptions about future patterns based on past data. Alternatives to Historical Analysis Analysts may consider alternatives to relying solely on historical data, such as conducting experiments or engaging in normative debates about desirable outcomes. These approaches can provide insights that are not constrained by past variations, allowing for a broader exploration of public affairs issues. Key Points: 1. Definition and Framework: The quantitative approach in public affairs is characterised by systematic stages of inquiry, resembling the scientific method. It can be defined as a structured method for analysing information, involving stages such as defining a research problem, selecting a research design, and analysing data. 2. Research Types: Quantitative research can be categorised into confirmatory and exploratory types. Confirmatory research tests specific hypotheses, while exploratory research identifies patterns without predefined questions. 3. Purposes of Quantitative Analysis: The main purposes include explanation, prediction, understanding, and description. Quantitative analysis allows researchers to summarise large amounts of information, define complex relationships, and make informed decisions about accuracy and precision. 4. Historical Context: Most public affairs analysis is rooted in historical context, relying on non-experimental designs. This reliance presents both strengths, such as authenticity, and weaknesses, including potential biases. 5. Limitations of Historical Analysis: Analysing historically generated information can be flawed due to biases and assumptions about future patterns based on past data. Researchers must be aware of these limitations when interpreting results.  6. Malleability of Results: The results and techniques in quantitative analysis are subject to modification over time. This raises questions about the objectivity of quantitative methods, which are better understood as intersubjective conventions rather than unchanging truths. 7. Stages of Quantitative Research: The stages of quantitative research include defining the research problem, selecting a research design, addressing measurement issues, collecting data, analysing data, and reporting results. These stages may not always be followed in a strict order. 8. Importance of Analytic Techniques: Identifying appropriate analytic techniques early in the research process is crucial. This allows researchers to address analysis issues effectively, even before data collection. Answers to the questions: List the seven stages of the research process in order. 1. Defining a research problem and related hypothesis. 2. Defining the type of research design or research strategy that will be used to test hypotheses. 3. Considering some of the estimation and measurement issues that will be important in subsequent analysis and determining the appropriate analytic techniques. 4. Determining and implementing an appropriate data collection technique. 5. Performing preliminary data entry and data processing often involving the use of computers. 6. Analysing data with a view to testing original hypotheses. 7. Reporting on the results of analysis for the purposes of generalisation and/or application. Chapter 3: Summary: Chapter 3 discusses three foundational elements essential for forming research problems in public affairs: public good, policy design, and net benefits of policy. Effective analysis in public affairs requires a framework for comparing options, which is challenging without a clear definition of the public good and a policy design framework. These elements are crucial not only for applied research but also for theoretical studies in public affairs, as they help in understanding the implications of various conceptions of the good and policy design. Models of the Public Good The chapter outlines different models of the public good, emphasising that discussions about public issues are often rooted in specific definitions of what constitutes the good. The author stresses the need for public affairs analysts to be cautious about defining the general public good, as such definitions often reflect the interests of specific groups rather than a universally beneficial outcome. Models of Policy Design The chapter further explores the concept of policy design, which involves creating frameworks for structured activities in public affairs. It distinguishes between formal policy statements and the actual actions taken by governments, emphasising that policy can also include proposals not yet adopted. A model of policy design allows analysts to compare various options and understand their implications, moving beyond traditional classifications of policy instruments to a more nuanced understanding of policy formulation and its outcomes. Policy Net Benefit The chapter concludes by discussing the importance of evaluating the net benefits of policy options, which involves assessing the balance of benefits and costs associated with different policies. Key indicators for evaluating net benefits include immediate and secondary benefits and costs, as well as their distribution across the community. The author highlights that while immediate benefits often accrue to those involved in policy design, the long-term impacts may vary, necessitating a careful analysis of how costs and benefits are distributed. Looking Ahead The chapter sets the stage for future research by suggesting that variations in policy design elements should be linked to variations in net benefits. It emphasises the role of belief systems in shaping definitions of the good and the importance of considering diverse perspectives in public affairs analysis. Key Points: 1. Foundational Elements: The chapter identifies three interrelated precursors to forming research problems in public affairs: public good, policy design, and net benefits of policy. These elements are essential for any useful analysis that informs public affairs and allows for the comparison of options. 2. Models of Public Good: Different models of public good are discussed, including individualist definitions that focus on personal preferences and functionalist or transcendent collective definitions that emphasise normative standards. Hybrid positions that combine both individual and collective components are also acknowledged. 3. Policy Design Framework: A clear policy design framework is necessary for constructing and comparing policy options. This framework should encompass both adopted and  proposed policies, allowing for a structured analysis of how different policy configurations can lead to varying outcomes. 4. Net Benefits Evaluation: The chapter emphasises the importance of evaluating the net benefits of policy options, which involves assessing the balance of immediate and secondary benefits and costs. It highlights the potential imbalance between those who benefit from policies and those who bear the costs. 5. Complexity of Public Good: The definition of public good is complex and often reflects the interests of specific groups rather than a universally beneficial outcome. Analysts must be cautious in their definitions to avoid biases. 6. Belief Systems Influence: Various belief systems can shape the definitions of costs, benefits, and policy elements, influencing how net benefits are perceived and evaluated. This underscores the need for a thoughtful and analytic approach to public affairs. 7. Importance of Comparison: The chapter concludes that meaningful development and comparison of policy options are challenging without a clear understanding of what constitutes the public good. This comparison is crucial for coherent debates about public policy implications. Answers to the questions: Three general types of models of the public good were mentioned in readings. What are they? 1. Individualist Definitions: This model assumes that public good arises from the interplay among individual preferences, often influenced by group or institutional affiliations. 2. Functionalist or Transcendent Collective Definitions: This perspective posits that public good exists beyond individual preferences and is guided by normative standards, which may be ethical, moral, or religious in nature. 3. Hybrid Positions: These models attempt to combine individual and collective components, suggesting that public good can include both individual rights and community entitlements. Chapter 4: Summary: The generation of research problems is the initial phase of quantitative analysis, encompassing three major aspects: the logical framework of analysis, the connection between quantitative logic and good theory in public affairs, and practical guidelines for producing research problems. The chapter addresses these aspects, starting with abstract considerations and moving to practical applications. Basic Logical Components of Research Problem Generation Research problems often stem from generalisations, which can be empirical (summarising observed patterns) or hypothetical (stating expected patterns). Generalisations consist of concepts (definitions or categories) and relationships (patterns between concepts). For example, a generalisation might state that as a person's age increases, their preference for short-term government expenditure also increases. Research problems can be generated through deduction (deriving new generalisations from existing ones) or induction (forming generalisations from specific observations). Both processes may coexist in the development of research problems. Systems of Generalizations: Theories and Related Topics Theories can be viewed as systems of interrelated empirical generalisations. They range from weak classifications to strong theoretical systems that can systematically derive hypotheses. Theories are essential for generating research problems and can also evolve from the analysis of those problems. Explanation, Prediction, and Causality Explanation in research involves reducing specific facts or generalisations from higher-order theories, while prediction uses established generalisations to forecast unknown outcomes. Causality is a critical aspect of explanation, requiring covariance, timing, substantive linkage, and non-spurious relationships. Empirical Theory in Public Affairs Good empirical theory in public affairs is vital for understanding findings and their implications. It should consider multiple levels of analysis, avoid simplistic causal assumptions, and be potentially testable. Theories should also address public good, policy design, and the distribution of benefits. Practical Guidelines for Generating Research Problems and Hypotheses The generation of research problems typically involves: 1. Using existing generalisations or theories to formulate a research problem. 2. Refining the problem into a preliminary hypothesis. 3. Reviewing existing research to inform the hypothesis. 4. Adjusting the hypothesis based on findings from the literature. Initial Conceptualization of a Research Problem Research problems should clearly define dependent and independent variables and the expected relationships between them. Visual representations can aid in clarifying these relationships. Key Points: 1. Phases of Research Problem Generation: The generation of research problems is the first phase of quantitative analysis, involving a logical framework, the connection between quantitative logic and good theory, and practical guidelines for producing research problems. 2. Generalisations: Research problems often arise from generalisations, which can be empirical (observed patterns) or hypothetical (expected patterns). These generalisations consist of concepts (variables) and relationships (patterns between concepts) that need to be clearly defined. 3. Hypothesis Development: The translation of research problems into specific hypotheses is crucial. This process involves refining preliminary hypotheses based on existing literature and research findings. 4. Causality and Relationships: Understanding the nature of relationships between variables is essential. This includes establishing covariance, timing of cause and effect, and substantive causal linkage. 5. Operationalization: Concepts must be operationalized, meaning they should be linked to measurable variables. This involves defining concepts at various levels, from general definitions to specific components. 6. Complexity of Causality: Research should not rely on simplistic views of causality. Instead, it should consider complex mechanisms and the temporal dimensions of relationships. 7. Practical Guidelines: Researchers should be aware of practical constraints that may affect their analysis and results. Acknowledging these constraints is important for maintaining the integrity of the research. 8. Visual Representation: Constructing visual representations of the variables and their expected relationships can help clarify the research problem and the direction of relationships. Answers to the questions: The two major components of generalisations are: 1. Concepts can be thought of as variables.  2. Relations describe the connections or patterns between these variables. This distinction is crucial when defining a research problem, as it helps analysts determine which variables are dependent and which are independent, thereby clarifying the nature of the relationships being studied. Understanding these components allows researchers to formulate more precise and testable hypotheses, ultimately leading to a better analysis of the phenomena under investigation. Name four types of systems of generalisations starting with the weakest and proceeding to the strongest. Here, weakness and strength are judged in terms of capacity to support the generation of research problems and explanations. 1. Ad Hoc Classificatory Systems: These are the weakest types of theories, consisting of arbitrary categorizations of information without detailed definitions or consideration of relationships between concepts. 2. Working Theories: These are stronger than ad hoc systems and provide a more structured approach, but they still lack the comprehensive nature of more developed theoretical frameworks. 3. Theoretical Frameworks: These systems combine various concepts and relationships to provide a more robust explanation of phenomena, allowing for systematic derivation of hypotheses. 4. Theoretical Systems: This is the strongest type, offering a complete explanation of a general kind of empirical phenomenon, where hypotheses can be systematically derived for important problems. What are the seven general types of concepts or variables that often form part of public affairs analysis? List them starting with those that are most likely to be dependent, and continue in order of their increasing likelihood of being independent. 1. Outputs and Outcomes of Political and Governmental Activity: These reflect the resources devoted to policy purposes or changes resulting from the application of such resources, and they are typically the focus of explanation or dependent variables in policy analysis. 2. Political Behaviour and Choice: This includes the voting, decisions, and demands of participants in political and policy activity, which are often influenced by other variables. 3. Stable Values: These are closely related to attitudes and preferences but reflect more general organising principles that structure specific attitudes and preferences about actors, issues, and topics.  4. Socio-Economic Factor: These factors can influence political behaviour and choices, acting as intervening variables between more general influences and specific outcomes. 5. Public Opinion: This variable is often analysed in relation to political behaviour and can be influenced by socio-economic factors and stable values. 6. Normative Preconceptions: These are established disciplinary habits that may influence the framing of research problems and the interpretation of findings. 7. Conceptual Frameworks: These provide the overarching structure for understanding the relationships between various variables and are often used to guide empirical analysis. What are the four major types of generalisations? 1. Empirical Generalisations: These are based on observed patterns and summarise existing data. They are often referred to as "laws" in the context of the philosophy of science. 2. Hypothetical Generalisations: These are statements that propose expected relationships or patterns that have yet to be tested against actual data. They are often formulated as hypotheses. 3. Universal Generalizations: These assert that a relationship holds true for all entities within a specified group. For example, a universal generalisation might state that "if families live in poverty, they will always undergo major structural change.”. 4. Statistical Generalisations: These indicate that a relationship is expected to hold true for a portion of the entities rather than all. An example would be stating that "if families live in poverty, they will have a fifty percent chance of undergoing major structural change.”. Three dimensions for classifying current public affairs theories are described in pages 52-56 of the main text. What are they? 1. Level of Analysis: This dimension emphasises the importance of different levels of analysis, such as micro (individual phenomena), meso (organisational/institutional phenomena), and macro (societal/large system level phenomena). 2. Causal Perspective: This dimension focuses on the causal relationships between variables, highlighting the need for theories to explain how certain factors influence outcomes in public affairs.  3. Theoretical Frameworks: This dimension involves the organisation of concepts and categories within a broader structure of generalisations or propositions, which helps in summarising, predicting, and explaining political information. What are the four simple properties of a good hypothesis according to the main text? 1. Clarity: The hypothesis should be clearly defined and understandable, ensuring that its meaning is unambiguous. 2. Specificity: It should specify the relationships between the concepts involved, detailing what is being tested. 3. Testability: The hypothesis must be constructed in a way that allows it to be tested with available methods and data. 4. Value-Free Construction: It should be formulated in a manner that is free from normative biases, focusing on empirical observation rather than subjective values. List the five major levels of operationalization starting with the most general or abstract and proceeding in order of decreasing abstractness or increasing closeness to actual observation. 1. Conceptual Level: This is the most abstract level, providing a general definition of a concept important to the researcher. For example, defining "individual potential power" as "all resources associated with an individual that enable that individual to directly influence the behaviour of others.” 2. Conceptual Components: These are the components included within the terms defined at the conceptual level, which are still fairly abstract. 3. Operational Definitions: This level involves specifying how the concepts will be measured or observed in practice, moving closer to actual data collection. 4. Measurement Implementation: This level focuses on the actual implementation of the measurement processes based on the operational definitions, which involves collecting data. 5. Empirical Observation: This is the most concrete level, where the data collected through measurement is analysed to draw conclusions about the concepts being studied. Chapter 5: Summary: Research design is a crucial aspect of research methods that involves making strategic decisions to test hypotheses or examine research problems. Traditional approaches often focus on classical experimental designs, which are considered the most robust. However, it is equally important to understand weaker designs, as they are frequently used in public affairs and related fields. This chapter outlines the fundamental elements of research design, including validity and threats to validity, and discusses both strong (experimental) and weaker (non-experimental) designs. Basic Elements of Research Design When developing a research design, researchers should consider four fundamental questions: 1. Should concepts and variables be studied over time or at a single point? 2. Is direct manipulation of independent variables necessary? 3. How should groups exposed to different levels of independent variables be compared? 4. Should subjects be randomly assigned to different levels of independent variables?. These questions relate to the type of hypothesis being tested, particularly concerning causality and the need for control and comparison between groups. Validity and Threats to Validity Validity in research design refers to the extent to which a design allows for accurate testing of hypotheses. Internal validity is concerned with the accuracy of conclusions drawn from the study, while external validity pertains to the generalizability of findings to a larger population. Campbell and Stanley identified several threats to internal validity, including history, maturation, experimental mortality, instrumentation, testing effects, regression artefacts, and interactions with selection. Experimental Designs True experimental designs are characterised by randomization, treatment groups, and control groups, making them strong in terms of internal validity. The classic experimental design involves pre-treatment and post-treatment measurements, allowing for comparison of changes between groups. Various advanced experimental designs exist, including multiple treatments, repeated measures, and blocking, which enhance the robustness of findings. Non-Experimental Designs Non-experimental designs, while generally weaker in internal validity, are commonly used in political and policy research. These include quasi-experiments, pre-experiments, and correlational designs. Non-equivalent control group designs are a common type of non-experimental design, where groups are not randomly assigned, leading to potential biases. Other designs in this category may involve treatment without a control group, which can complicate the interpretation of results. Key Points: 1. General Considerations: The chapter begins by addressing fundamental questions that must be considered when developing a research design, regardless of whether it is strong or weak. It emphasises the importance of validity and the threats to that validity in research designs. 2. Validity: Validity is a central theme, with a focus on internal validity (accuracy of conclusions) and external validity (generalizability of findings). The chapter discusses various threats to internal validity, such as history, maturation, and experimental mortality. 3. Experimental Designs: True experimental designs are highlighted for their strengths, particularly their reliance on randomization, control groups, and manipulated treatment variables. The classic experimental design is introduced as a robust model for testing hypotheses. 4. Non-Experimental Designs: The chapter also covers non-experimental designs, which are often weaker in terms of internal validity but are commonly used in political and policy research. These designs include quasi-experiments and correlational studies, which may lack random assignment and control groups. 5. Research Questions: Researchers should consider four key questions when developing a research design: the timing of data collection, the necessity of manipulating independent variables, the comparison of groups, and the random assignment of subjects. 6. Strengths and Weaknesses: While true experimental designs offer significant advantages in internal validity, they may face challenges regarding external validity, such as representativeness and the artificiality of experimental settings. These concerns, while valid, are sometimes overstated. 7. Hybrid Designs: The chapter discusses hybrid designs that combine elements of panel studies and comparative cross-sections, allowing for the analysis of change over time while addressing some limitations of traditional designs. 8. Statistical Considerations: The importance of statistical methods in addressing representativeness and validity is emphasised, particularly in relation to inferential statistics and the need for stratification in sampling. 9. Comprehensive Understanding: The chapter advocates for a comprehensive understanding of both strong and weak designs, as weaker designs are often more practical and widely used in public affairs and related fields. Answers to the questions: What are the seven major threats to the internal validity of a research design? 1. History - Unintended events that occur during a study which alter subject responses on important dependent variables. 2. Maturation - Changes in subjects over time that may influence outcomes. 3. Experimental Mortality (Attrition) - Loss of subjects during the study, which can affect the results if it varies across groups. 4. Instrumentation - Changes in measurement tools or procedures that can affect the consistency of data collection. 5. Testing Effects - The influence of prior testing on subsequent measurements, which can alter how subjects respond. 6. Regression Artefacts - Statistical regression to the mean affecting results, particularly in non-experimental designs. 7. Interactions with Selection - Biases introduced by the selection of subjects, which can combine with other threats to validity. What are the four major components of research design as discussed in class and in the main text. 1. Research Questions: Clearly defining the questions that the research aims to answer is fundamental to guiding the entire research process. 2. Hypotheses: Formulating testable hypotheses that can be evaluated through the research design is essential for establishing the direction of the study. 3. Variables: Identifying and operationalizing the independent and dependent variables involved in the research is crucial for understanding the relationships being studied. 4. Methodology: Selecting appropriate methods for data collection and analysis, including whether to use experimental or non-experimental designs, is a key component of the research design process. Other stuff: Given a set of variables or concepts hypothetically involved in relationships, be prepared to draw a diagram that portrays the most sensible way of viewing the independence and dependence of such concepts or variables. This should be done consistent with material presented in the book and lectures. Review the symbolism for major kinds of research designs. Be able to identify them and understand their major weaknesses and strengths. Be prepared to name some of the more commonly used designs and to discuss their strengths and weaknesses in detail. Be prepared to write a short essay describing and defending a research design that would be appropriate for a hypothesis that will be given to you.