Scientific Method in Social Sciences

• Techniques for quantitative social research applied to political science, focusing on social research and quantitative methodology are being reviewed

The Scientific Method in Social Sciences

• The scientific method in the social sciences is reviewed

Origins of the Scientific Method

• Originating philosophies, deductivism and inductivism, are being reviewed

Deductivism

• Deductivism involves Aristotelian logic and syllogisms
• It applies logical methods, such as "If A = B and B = C, then A = C"

Inductivism

• Inductivism is based on empiricism
• It is where theories are developed based on observations and experience
• Inductivism's approach moves from data to theory

Modern Scientific Method

• Karl Popper suggests the hypothetico-deductive method
• A key definition is falsifiability, meaning truth can only be demonstrated until it is disproven
• Scientific objectivity is not solely about the method
• Beliefs, values, and techniques are shared by a scientific community via paradigms (Kuhn 1978)
• Social sciences study humans as both objects and subjects
• Values guide questions, but not the procedure

Key Aspects of the Scientific Method

• Science uses a circular procedure of induction, empirical generalizations, theory, deduction, and hypothesis
• This involves observation, data collection, interpretation, and analysis

Understanding Theory

• A theory has established propositions within a scientific field and includes concepts, propositions, and axioms/laws

Main Functions of Theories

• Used to explain known generalizations and supports current knowledge
• Allows for theoretical exploration, to propose hypotheses on empirical connections
• Aids in planning new studies and form the basis for new insights

About Hypotheses

• Hypotheses are predictions about relationships which are derived from theory and tested empirically
• Theory leads to hypotheses
• Includes concepts linked to variables and indicators
• Specifies the unit of observation
• Hypotheses include two types: Descriptive and Explanative

Descriptive

• Descriptive type points out a percentage increase in party affiliation

Explanative

• Explanative indicates how a variable behaves when another changes

Research through Observation

• Observation involves data collection, selection of relevant variables, and adherence to methodology

Empirical Generalization

• Data analysis informs about observed relations and quantitative analyses provide estimated results with confidence
• Outcomes for hypotheses are confirmation or weakening/reformulation of the theory
• Empirical relations need a theoretical basis

Research Process and Design

• The research process involves the design of scientific investigation

Research Process Overview

• Involves:
• A topic of interest driven by the researcher's motivation
• Relation to existing theories
• Problem definition
• Research question formulation
• Hypothesis creation
• Data analysis to either accept or reject

Topic of Interest

• Researchers motivation/interests are included

From Topic to Problem

• Not all topics can turn into problems
• The problem must be clearly delimited
• Objectives include exploratory, descriptive, explicative, predictive, evaluative research
• Theoretical considerations use literature review, identifies research gaps, new research designs, and avoidance of errors

Defending a Research Problem

• It should be original, and address a gap in literature
• The study needs to be feasible, with available data, sufficient resources, and time
• Relevance includes theoretical aspects of general knowledge and application to socially relevant themes

Relevancy

• It applies to specific cases, improves methodology, and may shift a paradigm

Research Questions

• Questions can be relevant without systematic testing or to look for evidence supporting a side of a debate
• Involves examining underexplored aspects, to challenge frequently-accepted hypotheses, to study unquestioned premises
• Involves applying outside theories, or questioning something’s existence

Examples of Research Questions

• What is the effect of turnout on voting?
• What is the impact of personal characteristics on the probability of voting?
• What circumstances cause new parties to form?
• What are the consequences of implementing a low-emission zone?

Problem Operationalization

• Hypothesis links research question and previous theory
• Requires concept operaionalization, that is broken into dependent variables and independent variables
• The analysis units and population are stated in the hypothesis

Example Hypotheses

• When participation increases, more people vote for the left
• Higher education correlates with higher voter turnout
• Less economic growth leads to more electoral competition
• Rural backgrounds correlate with lower earning
• Traffic restrictions lead to more voting to the opposing party

Research Design

• The design depends on objectives, resources, and time

Research Design Based on Objectives

• Justification and feasibility must be considered

Investigation Strategies

• Strategies include documentary sources/existing data (Basic), case studies (Qualitative), surveys (Quantitative), and triangulation (Mixed-methods)
• Involves sample design and data collection

Research Design Typologies

• Grouped per experimental, temporal, and desired objectives

Research Design (1) Experimental

• Experimental study types in this category are:
• Preexperimental
• Quasi-experimental
• Experimental
• Manipulation, measurement and control of variables occur

Research Design (2) Temporal

• Divided into the following sections:
• Seccional
• Longitudinal
• Tendency
• Cohort
• Panel
• Observations in time and subject selection

Research Design (3) Objectives

• This is categorized by:
• Exploratory
• Descriptive
• Explicative
• Predictive
• Evaluative

Research Design Validity

• Validity encompasses internal, external, construct, and statistical conclusion

Internal

• To dismiss alternate explanations

External

• Results should be generalizable

Construct

• Conceptual operationalization is necessary

Statistical Conclusion

• Requires power, adequacy, and reliability for data analysis

Quantitative Techniques and Statistics

• Quantitative and statistical methods in research are required

Characteristics of Quantitative Techniques

• It measures social phenomenon magnitudes of interest
• Involves measuring data and intensive use of quantitative databases
• Analyzes using statistical methods
• Maximize control of data for other explanations
• Decisions on analysis are made before accessing the data
• Replication of studies is faciliated

Statistics in Social Sciences

• Statistics is a branch focused on collecting, managing, describing, representing, and analyzing data for population study

Division of Statistical Analysis

• Divided into descriptive and inferential statistic

Descriptive Statistics

• Describes and classifies using tables and graphs
• Used to organize, structure, summarize, and present data

Inferential Statistics

• Analyzes/interprets relationships via inductions
• Used to estimate parameters, always subject to uncertainty

Basic Requirements

• It is essential is to have quantitative data

Research Questions

• When defining a question to research, the following applies:
• To select studies with care and to understand population versus sample
• Data can either be collected or preexisting

Quantitative Data Collection

• It involves methods for quantitative data collection

Types of Databases

• Databases are classified according to:
• Coverage degree (exhaustive/partial)
• Observation level (individual/aggregate)
• Periodicity (continuous/periodic/circumstantial)
• Space-time dimension
• Sectional (transversal)
• Time series (longitudinal)
• Panel
• Investigator role (direct/indirect observation)
• Data nature (documentary, census, statistics, surveys, interviews)

Data Classifications

• Data is classified by secondary analysis
• Other aspects include:
• Post-electoral survey
• Gender violence data
• Statistics from INE
• Medical test data
• PIB evolution by region
• Business surverys

Sources of Existing Data

• Common sources for data include:
• Surveys
• Electoral data
• Country data
• Conflict data
• Statistics
• Thematic organizations' data
• Repositories

Data Availability

• There is greater data availability and access
• The journals require open publication of the data
• Data is accessible from academic articles
• New mechanisms exist via AI, geolocation, image processing