Chapter 2, 3 & 15 (page 463-470) Flashcards
Social Research - Approaches and Fundamentals
Deductive reasoning
From general info to specific
Inductive reasoning
From observations to making generable info & looking back / \
Hypotheses
An expected but unconfirmed relationship between two or more variables
Explanatory variables
Variables that are the object of study, all other variables are extraneous
Dependent variables (y)
The variable that is being explained and predicted / outcome
Independent variables (x)
The variables that do the influencing and exploring / cause
Bivariate hypothesis
Expected relationship between just two variables (are all possible variables and has a total effect)
Multivariate hypothesis
Expected relationship betwee a dependent variable Y and multiple variables X
Different multivariate hypothesis:
- Multiple causality
- Mediation (partial mediation)
- Moderating effect
- Spurious relationship
Multiple causality
Relative importance of independent variables
Mediation
An indirect effect (X1 –> X2 –> Y)
Partial mediation
An direct + indirect effect
Moderating effect
Intensifier or suppressor
Spurious relationship
An observed relationship between X1 and Y is spurious because they share a common cause X2
Common cause
Antecedent
Explanatory hypothesis
Explanation
Elaboration
Our understanding of a bivariate relationship by introducing a ‘third’ variable in contingency tables (cross/tabulations). Applies the moderation, mediation and spuriousness.
Simpson’s paradox
When you are comparing two variables there seems to be a relationship, but when you are splitting the data into two groups on a third variable, it is the other way around. = So, you can draw two opposite conclusions from the same data depending about how you split things up.
Causality, 3 necessary conditions:
- Association (variables must have a statistical association)
- Direct relationship (influence should be from cause to effect)
- Nonspuriousness (no extranous variables are allowed to explain the relationship between the variables of interest).
Units of analysis
The researched entities (objects/events)
Nested data (multilevel)
Combining data from different units of observation in which individual cases constitute elements of larger groups
External aggregation
Data from previous research
Internal aggregation
Aggregating individual level data (data you collect)
Logical fallacies
Drawing conclusions at one level while analyzing findings at another level
Two types of Fallacies
Ecological fallacy & Atomistic fallacy
Ecological fallacy
Drawing conclusions at the individual level while analyzing group level data
Atomistic fallacy
Analysis of individual level data is used to draw conclusions on aggregate level data