Video Lecture 2 (Week 1), Measurement and Methods. Guy Moors.
Summary and notes
Concepts, variables and hypotheses (chapter 3)
Concepts (or constructs) = general/abstract description of a social phenomenon e.g. ethnocentrism
Variable = empirical manifestation of a concept, e.g. scale that measures ethnocentrism. About how
you can measure/observe it.
Hypotheses = an expected relationship between 2 or more variables that can be researched/tested.
About observations you make. E.g. women are on average less ethnocentric than men. Hypotheses
bring concepts together in a different way.
Bivariate hypothesis (type that is most often used in research) : expected relationship between two
variables (=total effect)
When presented in a diagram: X ---- Y
Legend: X = independent variable (‘cause’) It’s value is independent of other variables in a study.
Independent variables are also called: explanatory variables, predictor variables, right-hand-side
variables (since they appear on the right-hand side of a regression equation).
Y = dependant variables (‘outcome’) is the effect. Its value depends on changes in the
independent variable. Dependent variables are also called: response variables, outcome variables,
left-hand-side variables (they appear on the left-hand side of a regression equation).
= direction of effect (from independent on dependent).
Recognizing independent variables:
- Is the variable manipulated, controlled or used as a subject grouping method by the
researcher?
- Does this variable come before the other variable in time?
- Is the researcher trying to understand whether or how this variable affects another variable?
Recognizing dependent variables:
- Is this variable measured as an outcome of the study?
- Is the variable dependent on another variable in the study?
- Does this variable get measured only after other variables are altered?
Examples: “The higher the emotional intelligence of a person, the higher the amount of money a
person gives to good causes.” Variables in this hypothesis are emotional intelligence, amount of
money. “the higher” signifies the direction of a relationship. In this case; positively. The wording
reflects metric measurement (scale).
Reflective exercise: the more a manager uses a laissez-faire management style the more likely this
will lead to a reduced work performance amongst employees. The higher the satisfaction level with
ones personal life the higher the chances they will have a stable relationship. Important to express
what you mean when writing a hypothesis!!
Use a regression analysis when your two variables are metric. This way you can present individual
observations in the regression model. When both Y and X are metric -> method to analyse =
regression -> estimates how much units change in Y is observed with one unit increase in X. E.g. Y is
metric and X is categorical -> estimates differences in mean score in Y for each category in X.
,What is a regression model?
A regression model provides a function that describes the relationship between one or more
independent variables and a response, dependent, or target variable. E.g. the relationship between
height and weight may be described by a linear regression model.
Multivariate hypothesis: expected relationship between a dependant variable Y and multiple
independent variables X.
Types of multivariate hypotheses:
1. Relative importance
- When you make a prediction about the relative importance of independent variables
(multiple causality) (causing the same outcome)
o
- ++ means that we expect X1 to be a more important variable than X2.
- Example of relative importance:
o
o Education would be considered more important (positively) while unemployement
benefit is considered to be negatively
2. Mediation:
= interpretation of a relationship. It tries to determine a relationship between an independent
variable and a dependent variable, by referring to a variable that is in between the two. It’s a chain of
effects (an indirect effect). Interpretation: strictly used in a situation where mediation is involved.
= the effect of the independent variable (X1) on the dependent (Y) is indirect trough its effect on the
intervening or mediating variable (X2) that in turn has an effect on the dependent (Y).
= indirect effect X1 ------> X2 -------> Y
o E.g.
o
o Accompanying hypothesis: “The older a person the less likely the chance of re-
entering the labour market. This effect is fully mediated by the opportunity to re-
enter since the older a person is the less opportunities there are and by consequence
the lower the chances of re-entering.”
o ! Look at the sign of effects (+ or -) in the diagram. The + sign is consistent with the
hypothesis !
, A positive relationship can be worded in 2 different ways (however, these are the same hypotheses):
1. “the more opportunities to re-enter the labour market the higher the likelihood of re-
entering”
2. “the less opportunities to re-enter the labour market the lower the likelihood of re-entering”
Partial mediation = direct + indirect effect.
When we talk about mediation, we basically say that there is no direct effect of the independent
variable on the dependent variable. You would still expect a direct effect alongside the indirect
effect.
o
3. A moderating effect (Moderation)
= interaction hypothesis. Indicating that you expect that the relationship between an independent
variable X1and the dependent variable Y, depends on the moderator (X2). Different from the
mediator, is that X2 is not in between X1 and Y. But how X2 has an impact on how X1 impacts Y. It’s a
conditional effect. (intensifier)
= effect of X1 on Y is conditional on the moderator (X2); or: the effect of X1 on Y is different
depending on the value of the moderator X2.
= conditional effect [intensifier (+) or suppressor (-) effect]
- Example:
- Hypothesis: “the higher the willingness to work, the higher the likelihood of re-entering the
labour market. This effect is intensified by the level of opportunity to re-enter.”
o
4. Spurious relationship
= common cause (antecedent).
= explanatory hypothesis (=explanation).
= an observed relationship between X1 and Y is spurious because they share a common cause: X2
o
- Example:
Summary and notes
Concepts, variables and hypotheses (chapter 3)
Concepts (or constructs) = general/abstract description of a social phenomenon e.g. ethnocentrism
Variable = empirical manifestation of a concept, e.g. scale that measures ethnocentrism. About how
you can measure/observe it.
Hypotheses = an expected relationship between 2 or more variables that can be researched/tested.
About observations you make. E.g. women are on average less ethnocentric than men. Hypotheses
bring concepts together in a different way.
Bivariate hypothesis (type that is most often used in research) : expected relationship between two
variables (=total effect)
When presented in a diagram: X ---- Y
Legend: X = independent variable (‘cause’) It’s value is independent of other variables in a study.
Independent variables are also called: explanatory variables, predictor variables, right-hand-side
variables (since they appear on the right-hand side of a regression equation).
Y = dependant variables (‘outcome’) is the effect. Its value depends on changes in the
independent variable. Dependent variables are also called: response variables, outcome variables,
left-hand-side variables (they appear on the left-hand side of a regression equation).
= direction of effect (from independent on dependent).
Recognizing independent variables:
- Is the variable manipulated, controlled or used as a subject grouping method by the
researcher?
- Does this variable come before the other variable in time?
- Is the researcher trying to understand whether or how this variable affects another variable?
Recognizing dependent variables:
- Is this variable measured as an outcome of the study?
- Is the variable dependent on another variable in the study?
- Does this variable get measured only after other variables are altered?
Examples: “The higher the emotional intelligence of a person, the higher the amount of money a
person gives to good causes.” Variables in this hypothesis are emotional intelligence, amount of
money. “the higher” signifies the direction of a relationship. In this case; positively. The wording
reflects metric measurement (scale).
Reflective exercise: the more a manager uses a laissez-faire management style the more likely this
will lead to a reduced work performance amongst employees. The higher the satisfaction level with
ones personal life the higher the chances they will have a stable relationship. Important to express
what you mean when writing a hypothesis!!
Use a regression analysis when your two variables are metric. This way you can present individual
observations in the regression model. When both Y and X are metric -> method to analyse =
regression -> estimates how much units change in Y is observed with one unit increase in X. E.g. Y is
metric and X is categorical -> estimates differences in mean score in Y for each category in X.
,What is a regression model?
A regression model provides a function that describes the relationship between one or more
independent variables and a response, dependent, or target variable. E.g. the relationship between
height and weight may be described by a linear regression model.
Multivariate hypothesis: expected relationship between a dependant variable Y and multiple
independent variables X.
Types of multivariate hypotheses:
1. Relative importance
- When you make a prediction about the relative importance of independent variables
(multiple causality) (causing the same outcome)
o
- ++ means that we expect X1 to be a more important variable than X2.
- Example of relative importance:
o
o Education would be considered more important (positively) while unemployement
benefit is considered to be negatively
2. Mediation:
= interpretation of a relationship. It tries to determine a relationship between an independent
variable and a dependent variable, by referring to a variable that is in between the two. It’s a chain of
effects (an indirect effect). Interpretation: strictly used in a situation where mediation is involved.
= the effect of the independent variable (X1) on the dependent (Y) is indirect trough its effect on the
intervening or mediating variable (X2) that in turn has an effect on the dependent (Y).
= indirect effect X1 ------> X2 -------> Y
o E.g.
o
o Accompanying hypothesis: “The older a person the less likely the chance of re-
entering the labour market. This effect is fully mediated by the opportunity to re-
enter since the older a person is the less opportunities there are and by consequence
the lower the chances of re-entering.”
o ! Look at the sign of effects (+ or -) in the diagram. The + sign is consistent with the
hypothesis !
, A positive relationship can be worded in 2 different ways (however, these are the same hypotheses):
1. “the more opportunities to re-enter the labour market the higher the likelihood of re-
entering”
2. “the less opportunities to re-enter the labour market the lower the likelihood of re-entering”
Partial mediation = direct + indirect effect.
When we talk about mediation, we basically say that there is no direct effect of the independent
variable on the dependent variable. You would still expect a direct effect alongside the indirect
effect.
o
3. A moderating effect (Moderation)
= interaction hypothesis. Indicating that you expect that the relationship between an independent
variable X1and the dependent variable Y, depends on the moderator (X2). Different from the
mediator, is that X2 is not in between X1 and Y. But how X2 has an impact on how X1 impacts Y. It’s a
conditional effect. (intensifier)
= effect of X1 on Y is conditional on the moderator (X2); or: the effect of X1 on Y is different
depending on the value of the moderator X2.
= conditional effect [intensifier (+) or suppressor (-) effect]
- Example:
- Hypothesis: “the higher the willingness to work, the higher the likelihood of re-entering the
labour market. This effect is intensified by the level of opportunity to re-enter.”
o
4. Spurious relationship
= common cause (antecedent).
= explanatory hypothesis (=explanation).
= an observed relationship between X1 and Y is spurious because they share a common cause: X2
o
- Example:
