3. Literature, research design, and the steps of Research

The Purposes of Social Research

Exploratory

Investigating new/poorly understood phenomena — little idea of what the phenomenon looks like in society & how it could be measured.
Goal: collect a lot + diverse information about the phenomenon → based on this = descriptive/explanatory research.

• Inductive/deductive

• Questionnaire-based data, qualitative interviews, content analysis, focus groups, etc.

• Complex explanations based on quantitative and or qualitative data

Descriptive

Systematically describe social phenomena — measure phenomena on how widespread + diverse it is within a society.

• Neither deductive nor inductive

• Demographic/administrative data collection, participant observation

• Descriptive statistics, thick descriptions

Explanatory

Identify casual relationships/mechanisms → goal → document + explain why a given penomena takes different forms in different groups in society.

• Inductive

• Non-intrusive observaiton

• Thick descriptions

• Objective causes

• Positivist paradigm

• Methodological collectivism

Interpretive

• Subjective causes, attribution of meaning

• Action theory, then phenomological paradigm

• Methodological individualism

Research plan = A real plan

• What do we want to study?

• Why do we want to study it?

• Why is it worth studying?

• Why is this topic significant?

• What is the goal of research?

• How do we plan to carry out the study?

Steps of a research plan…

1. Resarch topic

2. Formulating research problem

   1. What exactly to investigate?

3. Research questions & hypothesis

4. Conceptualization & operationalization

   1. Defining key concepts + measure — how?

5. Choosing research method

   1. Determined by topic & objective

   2. Should the study be qualitative or quantitative?

   3. How will the data be collected?

   4. What type of analysis will be used?

6. Sampling

   1. Selecting cases/participants

   2. How do we plan to sample from the population?

   3. How will we access the participants?

   4. How do we persuade them to participate?

7. Instrument design & testing

8. Data collecting

9. Data process & cleaning

10. Data analysis

11. Dissemination

    1. Publishing & communicating findings

Defining topic of Research

• Are you interested about it?

• What’s the scientific contribution? (theoretical, practical, conceptual, methodological)

• Are we testing hypotheses/answering research questions?

• Does it extend, refine, test existing theories?

• Does it apply concepts in new contexts?

• Does it resolve theoretical contradictions?

• Does it introduce new measurement strategies?
  
  

We also need to know the scope of it:

• Why is this research worth doing?

• How can the results be generalized to?

• What’s the temporal & geogrephical scope?

Why use literature?

Positivist approach:

• More credibility

• Shows what researcher adds

• Differentiates own results from others
  
  

Structuralist approach:

• Engages in scientific discourse

• Positions research in academic fields

• Acknowledges other agents as experts

A good research question is…

• Interesting — for the researcher & academia

• Relevant — contributes to existing knowledge

• Feasible — realistic within time & budge constraints

• Ethical — what do we imply by researching this topic?

• Concise, clear, specific

• Answerable — possibly supported by sub-questions

• Has dependant + independent variable, specific time & population

• Broad — if it’s answerable with “yes” or “no” → it’s not a good question!

• Not too broad either — not too narrow, but narrow enough to research and answer the question easily

What is a hypothesis?

• Clear, testable statement predicting the relationship between two or more variables

• Connects the theory with the empirical test

• Written as:

  • “If x, then y” or “There’s a relationship between x and y” or “There’s a relationship between x and y, because…” ← one sentence!!

• Every research questions needs at least one hypothesis! — to answer a research question

Measurement — what are the key variables related to the study?

Conceptualization: how would you define, what do you want to measure? — defining precisely what we mean by each concept.

• The first step in social science research is the specification and precise definition of the concepts used

• It’s goal is to determine which indicators will be used by the researcher to measure th ephenomenon in question

• We may have a given concept that cannot be directly measured (e.g.: empathy, trust)

• Such a concept can be measured using indicators or makers

Operationalization: how do you want to measure it? — defining specific procedures or measurement tools through which our key concepts can be empirically examined.

• Defining the specific research procedures and steps — survey questions or indicators?

• E.g.: in survey research, this refers to the developmentof the actual questionnaire items: how detailed the measurement will be, and what types of questions will be used?

Conceptualization: Trust

• Trust can be conceptualized as a psychological state in which an individual is willing to accept vulnerability, because they have positive expectations about another person’s or institution’s intentions and behaviour. Trust therefore becomes particularly impotant in situations involving uncertainty, risk, dependence.
  
  

• Trust may refer to:

  • Interpersonal trust: confidence in particular individuals (friends, colleagues, acquaintances)
    
    

  • Generalized trust: belief that most people (incl. strangers) can generally be trusted
    
    

  • Institutional trust: confidence that institutions (government, universities, courts, police) will act competently, fairly, predictably
    
    

  • Particularized trust: trust directed toward people with whom an individual has close personal/group-based relationships
    
    

  • Percieved trustworthiness — commonly understood with three central dimensions: ability, benevolence, integrity

Operationalization: Trust

• How could you measure it with survey questions or indicators?

  • “Do you think most people can be trusted"?” (scale 1-5)

• The process of degning specific procedures/measurement tools through our key concepts can be empirically examined
  
  

• Example: Religiosity, age

• Hypothesis: Older individuals are more religiosu than younger individuals

• Religiosity: “How often do you attend religious services?”

• Attributes: daily, several times a week, weekly, several times a month, monthly, several times a year, only on major religious holidays, or never

• Age: “How old are you?”

• Attributes: exact age measured on a continous scale from the minimum to the maximum possible values

Data Processing and Analysis

Data processing:

• Data entry

• Data cleaning

• Weighting

Analysis:

• What type of analysis will be conducted?

• Is the research descriptive, or does it aim to explain casual relationships?

Measurement Qualities

• Precision: the degree of specificity in measurement — it reflects how finely we distuingish between the attributes of a variable

  • E.g.: Pest County → Budapest → District 13 (it reflects how finely we distuingish between the attributes of a variable)
    
    

• Reliability: a measurement is repeated, it should yield the same result — it’s stable, representative, equivalent, reliable, consistent

  • Stability: repeated measurements yield consistent results

  • Representativeness: results are consistent across different samples - population

  • Equivalence: different indicators produce the same outcome

  • Reliability: consistency, same result each time
    
    

• Internal validity (accuracy, measures what it’s supposed to measure): — how accurately a measurement reflects the actual meaning of the concept it’s intended to measure

  • Types of validity:

    • Face validity: does the measurement appear to reflect the concept?

    • Criterion validity: how well does the measure correlate with an external criterion?

    • Construct validity: does the measure relate to other virables as expected theoretically?

    • Content validity: does it cover all dimensions of the intended concept/phenomenon?
      
      

• External validity (generalizability): how well the findings of a study can be generalized to a broader population — reliable and internally valid
  (external validity may be questionable!!)

Errors in the Process of Knowing

Inaccurate Observation:

• Scientific observation is a conscious acitivity — simple or complex measurement tools helps avoid observational errors
  E.g.: “What was the speaker wearing?” - guesswork vs. accurate recall
  
  

Overgeneralization:

• We detect a pattern and assume that a few similar events confirm a general rule — we draw broad conclusions from a limited number of observations
  Solutions:

  • Replication (repeat the study)

  • Sufficiently large and representative sample
    
    

Selective Observation:

• We tend to notice only the events that confirm our assumptions/prior expectations
  E.g.: “The exception proves the rule”
  
  

Reductionism:

• Overly narrow explanations, or choosing an inappropriate unit of analysis — the researcher treats some units as disproportionately important compared to others.

Triangulation

• Triangulation refers to the method of using multiple sources/independent paths to reach the same conclusion → strengthens the validity of the claim

  • Used in land surveying/location positions (e.g.: navigation)

• In social research, it refers to combining multiple qualitative methods

• Its scope is broader, it includes combining data, researchers, theories, methods

Types of Triangulation

Data and Research triangulation

• Across time, space, or participants

• Multiple researchers examining the same data

Theory triangulation

• Interpreting data suing different theoretical frameworks

Methodological triangulation

• within-method/between-method combinations

Research Ethics

• Researchers bear responsibility — how we generate knowledge, interpret it, communicate it to the public

• Ethical dilemmas may arise at any stage of the research process, even the choice of topic involves ethic

• Key ethical concerns:

  • Who is included/excluded in the production of knowledge?

  • What kinds of issues deem “worthy” of research?

  • Research takes place in non-democratic contexts — it’s not an interaction between equals

  • Quantitative designs are typically less democratic than qualitative designs

  • Practical, technical, ethnical principles of conducting research often come into conflict

Whom do we owe ethical responsibility?

• Participants

  • Participation must be voluntary

  • Participants must be informed in advance

  • No harm should come to them

  • Guarantee of anonymity and confidentiality

  • Respect for the right to privacy

• The profession and colleagues

• The broader public

• Funders/sponsors of the research

Two approaches to Handling Ethical Dilemmas

Rule-based approach:

• Predefined protocls/codes of conduct

• Followed without weighing consequences

  • Advantages:

    • transparency

      • minimizes the risk of “cutting corners”

  • Disatvantages:

    • rigid and inflexible

Judgment-based approach:

• Emphasizes principles and guidelines rather than strict rules

• Consequences are weighed in context

  • Advantages:

    • accommodates the uniqueness of each research situation

  • Disatvantages:

    • risk of subjectivity

    • time-consuming