Lecture 1
Methods
Book: B.C. Straits & R.A. Singleton - Social Research, Approaches and Fundamentals (6th edition).
Origins of research topics
• Personal interest of the researcher
• The structure and state of the scientific discipline
• Social problems
• Social premiums
• Practical considerations
Chapter 2
Scientific questions
Questions that can be answered by making observations that identify the conditions under which
certain events occur. Still, to qualify as scientific knowledge, the answers to such questions must take
a particular form, a form that meets the requirements of description, explanation, prediction, and
understanding.
Product: logical structure of knowledge that tells us how and why (laws, principles and theories).
Process: the methods and logic of inquiry through which the knowledge is created, tested and
refined.
Theory <- Proposition -> Hypotheses
Theory: provides an explanation for a proposition or set of propositions.
Proposition is a general statement regarding a regularity in the behaviour or opinion of subjects.
Hypothesis: an expected relationship between 2 or more variables that can tested/observed.
Example
Proposition Hypothesis
When someone performs a task well, the Athletes will perform better the more spectators
performance will be better with the presence of there are.
others than without.
,Observation is the end of deduction and the start of induction.
Concept is a general/abstract description of a social phenomenon (e.g. weight, social status,
extraversion).
↓ operationalization
Variable is an empirical manifestation of a concept = measurable (with a scale).
A variable refers to the measure itself; a concept to the definition of what is measured.
Chapter 3
Causality; 3 conditions
1. Association = statistical relationship between the variables.
Often weak relationships are observed, due to measurement errors and multicausality.
2. Direction of the relationship
Independent variable influences the dependent variable (and not the other way around)
3. Nonspuriousness = no extra variables or antecedents are allowed to explain the relationship
between the variables of interest. Antecedents are often called 'control variables'. The more
control variables in a model, the more likely the relationship is not spurious if still observed
when the control variables are included.
Multiple causality: all models including more than 2 variables.
Unit of observation: who or what is observed.
Unit of analysis: who or what defines the cases for which you want to make predictions.
Example:
• Unit of observation = direct manager or leader of the work team
• Unit of analysis = work team
Nested data = multilevel data
Combining data from different units of observation in shich individual cases constitute elements of
larger groups (aggregates). Example: European Values Study; survey research among representative
samples across European countries.
Sources of information at group level:
• National & regional statistics
• Data from previous research (on gruop level)
• Aggregating (grouping) individual level data
Logical fallacy
Drawing conclusions at one level while analyzing findings at another level.
Ecological fallacy
Drawing conclusions at the individual level while analyzing group level data.
Atomistic fallacy
Drawing conclusions on aggregate level from analyses of individual level data.
, Types of hypotheses
Bivariate hypothesis = expected relationship between two variables.
X -> Y
Multivariate hypothesis = expected relationship between a dependent variable Y and multiple
independent variables.
X1 -- ++ -->
Y
X2 -- + -->
Mediation is the effect of the independent variable (X3) on the dependent (Y) is indirect through its
effect on the intervening of mediating variable (X4) that in turn has an effect on the dependet (Y).
X3 -- - --> X4 -- + --> Y
X3 = age / grades
X4 = opportunity to re-enter the labor market / self-esteem
Y = re-entering the labor market / happiness
"The older a person the less likely the chance of re-entering the labor market. This effect is fully
mediated by the oppurtunity to re-enter, since the older a person is the less opportunities there are
and by consequence the lower the likelihood of re-entering."
Moderation: the effect of X5 on Y is affected by X4; the effect of X5 on Y is different depending on the
value of the moderator X4; the effect of X5 on Y is conditional on the moderator (X4).
Methods
Book: B.C. Straits & R.A. Singleton - Social Research, Approaches and Fundamentals (6th edition).
Origins of research topics
• Personal interest of the researcher
• The structure and state of the scientific discipline
• Social problems
• Social premiums
• Practical considerations
Chapter 2
Scientific questions
Questions that can be answered by making observations that identify the conditions under which
certain events occur. Still, to qualify as scientific knowledge, the answers to such questions must take
a particular form, a form that meets the requirements of description, explanation, prediction, and
understanding.
Product: logical structure of knowledge that tells us how and why (laws, principles and theories).
Process: the methods and logic of inquiry through which the knowledge is created, tested and
refined.
Theory <- Proposition -> Hypotheses
Theory: provides an explanation for a proposition or set of propositions.
Proposition is a general statement regarding a regularity in the behaviour or opinion of subjects.
Hypothesis: an expected relationship between 2 or more variables that can tested/observed.
Example
Proposition Hypothesis
When someone performs a task well, the Athletes will perform better the more spectators
performance will be better with the presence of there are.
others than without.
,Observation is the end of deduction and the start of induction.
Concept is a general/abstract description of a social phenomenon (e.g. weight, social status,
extraversion).
↓ operationalization
Variable is an empirical manifestation of a concept = measurable (with a scale).
A variable refers to the measure itself; a concept to the definition of what is measured.
Chapter 3
Causality; 3 conditions
1. Association = statistical relationship between the variables.
Often weak relationships are observed, due to measurement errors and multicausality.
2. Direction of the relationship
Independent variable influences the dependent variable (and not the other way around)
3. Nonspuriousness = no extra variables or antecedents are allowed to explain the relationship
between the variables of interest. Antecedents are often called 'control variables'. The more
control variables in a model, the more likely the relationship is not spurious if still observed
when the control variables are included.
Multiple causality: all models including more than 2 variables.
Unit of observation: who or what is observed.
Unit of analysis: who or what defines the cases for which you want to make predictions.
Example:
• Unit of observation = direct manager or leader of the work team
• Unit of analysis = work team
Nested data = multilevel data
Combining data from different units of observation in shich individual cases constitute elements of
larger groups (aggregates). Example: European Values Study; survey research among representative
samples across European countries.
Sources of information at group level:
• National & regional statistics
• Data from previous research (on gruop level)
• Aggregating (grouping) individual level data
Logical fallacy
Drawing conclusions at one level while analyzing findings at another level.
Ecological fallacy
Drawing conclusions at the individual level while analyzing group level data.
Atomistic fallacy
Drawing conclusions on aggregate level from analyses of individual level data.
, Types of hypotheses
Bivariate hypothesis = expected relationship between two variables.
X -> Y
Multivariate hypothesis = expected relationship between a dependent variable Y and multiple
independent variables.
X1 -- ++ -->
Y
X2 -- + -->
Mediation is the effect of the independent variable (X3) on the dependent (Y) is indirect through its
effect on the intervening of mediating variable (X4) that in turn has an effect on the dependet (Y).
X3 -- - --> X4 -- + --> Y
X3 = age / grades
X4 = opportunity to re-enter the labor market / self-esteem
Y = re-entering the labor market / happiness
"The older a person the less likely the chance of re-entering the labor market. This effect is fully
mediated by the oppurtunity to re-enter, since the older a person is the less opportunities there are
and by consequence the lower the likelihood of re-entering."
Moderation: the effect of X5 on Y is affected by X4; the effect of X5 on Y is different depending on the
value of the moderator X4; the effect of X5 on Y is conditional on the moderator (X4).
