WHAT IS IS PROCESS TRACING AND WHEN CAN WE USE IT IT? PVM-Y302 - - PowerPoint PPT Presentation

WHAT IS IS PROCESS TRACING AND WHEN CAN WE USE IT IT?

PVM-Y302 Advanced research methods: Process tracing Session 1, 1 April 2019

OUTLINE

• What is process tracing?
• What can we do with it?
• Evidence and types of prior knowledge
• Variants of process tracing

WHAT IS IS PROCESS TRACING?

• ‘Systematic examination of diagnostic evidence selected and analyzed

in light of research questions and hypotheses posed by the investigator’ (Beach & Pedersen, 2013)

• Qualitative within-case analysis – close engagement with prior

knowledge

• Goal of analysis: descriptive or explanatory (causal claims)
• Empirical observation -> diagnostic evidence
• Diagnostic evidence: if mechanism M exists, what observable

evidence would it leave in a case?

WHAT CAN WE DO WIT ITH PT?

• Describe new political and social phenomena systematically

(e.g. color revolutions, the use of social media in political protest)

• Hypothesis testing and assessing causal mechanisms
• Insight into causal mechanisms
• Enrich case knowledge
• Use together with quantitative methods to get a more

complex picture

TYPES OF PRIOR KNOWLEDGE

• Conceptual frameworks - related concepts + ideas about how they

can be operationalized

• Recurring empirical regularities - established patterns in the

relationship of two or more phenomena (‘if a then b’ - causal or descriptive approach)

• Theory 1: builds on recurring regularities by connecting them as a set
• f insights (verified, interconnected hypotheses)
• Theory 2: empirical regularities + explanations as to why they occur -

this may also be called an explanatory model

VARIANTS OF PROCESS TRACING

• Theory-centric:
• Theory testing
• Theory building
• Case-centric: outcome explaining
• More common in PT studies

THEORY TESTING PT

• Purpose: is a certain causal mechanism present and does it function as

theorized?

• Starting point: we know where both X and Y are
• Outcome of analysis: was the causal mechanism present in a single case?

Did it function as expected?

• What are we tracing: Single generalizable mechanism
• Steps:
• Conceptualize causal mechanisms and context
• Collect empirical evidence to be used for causal inferences
• Update confidence about (partial) presence/absence of predicted mechanism

THEORY BUIL ILDING PT

• Purpose: what is the causal mechanism between X and Y?
• Starting point: Empirical knowledge that needs structured analysis
• Outcome of analysis: Detect plausible hypothetical causal mechanisms about how

X is linked to Y

• What are we tracing: Single generalizable mechanism
• When to use it:
• Known correlation between X and Y but no known mechanisms or theory
• Known outcome but unclear causes
• Steps:
• Search empirical material for possible diagnostic evidence of the causal

mechanism

• Infer evidence from step 1 reflects an underlying causal mechanism
• Repeat steps 1-2 as needed

OUTCOME EXPLAINING PT

• Purpose: what mechanism accounts for a certain outcome?
• Starting point: case knowledge + theoretical background of

mechanisms

• Outcome of analysis: minimally sufficient explanation (most

important aspects with no redundant parts)

• What are we tracing: Case-specific (composite) mechanism
• Steps:
• Determine ‘facts’ of the case (empirical narrative)
• Update sufficiency of explanation for outcome
• Repeat steps 1-2 as needed

PT VARIANTS: DIF IFFERENCES

• Theory-centric or case-centric?
• Test or build theorized causal mechanisms?
• Focus on systematic mechanisms expected to be present in a

population, or case-specific mechanisms?

• What types of inferences are made?
• Theory testing - about the presence/absence of a mechanism
• Explaining outcome - about the sufficiency of the explanation

ROUND-UP UP

• PT: qualitative in-case analysis, engage with prior knowledge
• Goals: descriptive or causal analysis
• Turning observations/prior knowledge into diagnostic

evidence

• If a mechanism M exists - what observable evidence

would it leave in a case?

• Variants: theory testing, theory building, outcome explaining