Cours 1

• Research Question


  A question that can be answered with evidence, which improves our understanding of the world
• Characteristics of a good research question

  • Answerability: There must be potential evidence that could provide a believable answer
  • Informative: The answer should help improve understanding or explain a theory about the world
• Theory


  An explanation of why or how something happens
• Hypothesis


  A specific, testable prediction derived from a theory
• Why start with a research question?

  • Avoids data mining: Starting with a question focuses research on informative, theory-driven investigations
  • Improves theory: Research questions derived from theories help deepen our understanding of the world
• Where do research questions come from?

  • Curiosity: Natural wonder about how the world works
  • Theory: Propositions about how something in the world functions
  • Observation: Noticing patterns or anomalies in data or the environment
  • Opportunity: Unique datasets or unusual events can inspire distinctive research questions
• How to evaluate a good research question

  • Consider potential results: Think about whether possible answers can inform or challenge the theory
  • Consider feasibility: Assess if necessary data is accessible and the question can be realistically answered
  • Consider scale: Ensure the question's complexity matches available resources and time
  • Consider design: Identify a reasonable method to investigate the question
  • Keep it simple: Focus on specific, manageable aspects to avoid overcomplicating
• Types of research besides empirical research

  • Theoretical research
  • Applied research
  • Exploratory research
  • Historical research
  • Review research
• Empirical research


  A type of research that involves collecting data through observation, experimentation, or other direct or indirect forms of measurement
• Types of empirical research

  • Experimental research
  • Observational research
  • Survey research
  • Case study research
• Conducting empirical research


  • Carefully design studies, collect data, and analyze it using appropriate statistical methods
  • Present results in a clear and concise manner, with appropriate citations to support the findings
• Differences between empirical research and everyday data

  • Systematic procedure and documentation vs unsystematic collection of information
  • Purpose of decision making and knowledge gain vs orientation and sense of purpose
  • Use of uniformly defined terms and scientific hypotheses vs use of non-transparent colloquial language and everyday assumptions
  • Statistical evaluation procedures and verification of validity criteria vs decision based on subjective probabilities and unfounded generalization
  • Permanent review and criticism in professional circles vs belief in your own theory
• Research procedure


  • Exploration phase - Develop the question
  • Theoretical phase: Link research to relevant theories
  • Planning phase: Variables, experimental design, sampling, measurement instruments, planning statistical analysis
  • Investigation phase: Develop a coherent chain of reasoning
  • Evaluation phase: Replicate and generalize findings
  • Decision phase: Make research available for critique and publish findings
• Theory


  Describes, explains and predicts facts. Consists of a network of proven hypotheses or recognized empirical "regularities"
• Criteria for evaluating theories

  • Logical consistency
  • Falsifiability
  • Parsimony
  • Proven value
• Hypothesis


  Describes a presumed relationship or difference between characteristics or the change of characteristics

  
• Theories form a hypothesis network from the individual characteristics and hypotheses
• Theory


  Valid as long as it is not disproved
• Law


  Always, everywhere valid. Deterministic statement type, applicable in natural sciences
• Ethical issues in empirical research

  • Weighing of benefits: scientific progress vs. human dignity
  • Personal responsibility
  • Duty to inform
  • Voluntary participation
  • Avoidance of psychological or physical harm
  • Anonymity of results
• Criteria for a scientific hypothesis

  • Testable: Can be tested through empirical methods
  • Falsifiable: Can be proven false if not supported by evidence
  • Empirical: Based on empirical evidence or observations
  • Specific: Clear in terms of what it predicts or explains
  • Parsimonious: As simple and straightforward as possible
• Examples of scientific and non-scientific hypotheses

  • When people are frustrated, they react aggressively (scientific)
  • There are children who cry (not scientific)
  • When people watch television, they are satisfying their television needs (not scientific)
  • The stronger the fatigue, the weaker the concentration (scientific)
  • Heavy cigarette use can lead to a heart attack (scientific)
  • Students have a lot of free time (not scientific)
  • There is a correlation between intelligence and shoe size (not scientific)
  • Eating vegetables is healthy (not scientific)
• Directional hypothesis


  Alternative hypothesis contains less than ("<") or greater than (">") sign, testing for positive or negative effect
• Non-directional hypothesis


  Alternative hypothesis contains not equal ("≠") sign, testing for some effect without specifying direction

  
• Variable


  A bunch of observations of the same measurement
• Variable


  In the context of empirical research, a bunch of observations of the same measurement
• Examples of variables


  • The monthly incomes of 433 Romanians
  • The number of business mergers in Bucharest in each year from 2017-2024
  • The psychological "extraversion" score from interviews with 744 children
  • The color of 532 flowers
• Successfully describing a variable means being able to take those observations and clearly explain what was observed without making someone look through all the individual observations
• Variables


  • Allow researchers to systematically study the relationship between different factors and outcomes
  • By manipulating or measuring variables, researchers can test hypotheses and develop theories about human behavior and cognition
  • Characteristics values vary, otherwise they are just a constant
  • Characteristics values can be recorded by rule-based assignment of numbers, i.e., measurement
• Common types of variables

  • Continuous
  • Count
  • Ordinal
  • Categorical
  • Binary
  • Qualitative
• Continuous variable


  A variable that could take any value (perhaps within some range)
• Count variable


  A variable that counts something, cannot be negative or fractional
• Ordinal variable


  A variable where some values are "more" and others are "less," but there's not necessarily a rule as to how much more "more" is
• Categorical variable


  A variable recording which category an observation is in, none of the options is "more" than the others
• Binary variable


  A categorical variable that only takes two values, often "yes" and "no"
• Qualitative variable


  A variable that is not numeric in nature, but also not categorical
• Independent variable


  The variable that the researcher manipulates in order to see how it affects other variables
• Dependent variable


  The variable that the researcher measures to see how it is affected by the independent variable
• Control variable


  Variables that are held constant or controlled in order to prevent them from affecting the outcome of the study
• Mediating variable


  Variables that explain how or why the independent variable affects the dependent variable