Advanced research methods in social and organizational psychology
Lecture 1 (2x gekeken)
Spreadsheet online tonight -> fill it out!
This is for finding a partner for the assignment
Exam is based on lectures
You need to know and understand what’s on the slides.
Open book exam
Refreshing vocabulary
• 1. True experiment:
22. A design in which participants are assigned randomly to treatments
• 2. Quasi experiment:
5. A design that resembles that of an experiment in that discrete groups are used, but Pp’s
aren’t randomly assigned to treatments nor are treatments randomly determined for groups
• 3. “Between groups" vs. "within groups" design:
6. A treatment between conditions and a measurement referring to differences within the
particpants at different times
• 4. Construct variable:
9. A theoretical variable that has ‘reality status” such as competition, attractiveness,
negative mood
• 5. Operational variable:
10. The researchers’ precise operational definition (i.e., measurement) of a theoretical
construct
6. Independent variable:
12. The variable presumed to cause a change in the dependent variable
• 7. Dependent variable:
2. The variable presumed to be affected by the independent variable
• 8. Hypothesis:
13. A statement of a proposed relation between constructs
• 9. Theory:
4. A well-established principle that has been developed to explain some aspect of the natural
world.
• 10. Construct validity:
19. The degree to which the operational definition accurately measures the construct of
interest
11. Convergent validity:
1. Overlap among variables presumed to measure the same construct
• 12. Discriminant validity:
14. The extent to which it is possible to discriminate between dissimilar constructs
• 13. Random assignment:
15. The process by which subjects receive an equal chance of being assigned to a particular
condition
• 14. Manipulation check:
16. A measured variable designed to assess whether the manipulation worked and tapped a
desired construct
,• 15. Demand characteristic:
17. Aspect of the experiment encouraging the participant to respond according to situational
constraints
16. Reliability:
18. The extent to which a construct is measured without error or bias
• 17. Subject expectancies:
7. A demand characteristic whereby subjects think they know the experimenters’ interests
and act accordingly
• 18. Double-blind procedure:
3. A procedure in which neither experimenter nor pp knows to which condition the pp is
assigned
• 19. Order effects:
21. The effects on behavior of presenting two or more treatments to the same pps.
• 20. Counter-balancing:
8. A technique for controlling order effects by which each condition is presented first to each
participant an equal number of times: present each condition in each possible order
21. Moderation (interaction)
11. The effect of an independent variable on a dependent variable depends on the level of a
third variable (i.e., the moderator)
• 22. Mediation:
20. The effect of an independent variable on a dependent variable is explained by the
change in a third variable (i.e., the mediator)
Moderation & Mediation
Combination = Conditional Process Analysis
- What model you develop depends on the question you want to address & underlying
theory. What is your theoretical model?
- If you want to explain under what conditions something is occurring (e.g. in certain
organizational environments or for certain people)
o Moderation model fits best
- If you want to explain WHY something is happening (i.e. explaining an underlying
mechanism)
o Mediation model fits best
- Combinations (i.e. conditional process analysis) also possible
Example: You are interested in the relationship between emotional job demands and
emotional exhaustion at the work place.
Are there differences between man and woman?
Moderation model
Example: you are interested in why emotional demands relate to emotional exhaustion
Mediation model
You can combine these two models, for example: I want to focus on the relationship
between emotional demands on emotional exhaustion and this is mediated by unfavourable
coping strategies.
This process is stronger for woman then for men
,Combinations: focus on both
Moderation
It tells you about the boundary condition. It can explain under what conditions is the
relationship strong or less strong. It can influence the strength or direction of the
relationship.
Can increase our understanding of the relations between predictors and outcomes
- Moderator variable changes the strength or direction of the relationship between x
and y
For example: I looked into the effects of shared leadership on team creativity for low vs. high
backing-up behavior of team members
Backing up: moderator
- Choice of a moderator and nature of the interaction should be based on theory.
Example slides:
Job Demands-Resources model:
Inherent in this model: true moderation effects.
Various job demands positive relate to strain symptoms at work. Job resources offer
that relationship. If you have certain resources at work (social support, high level of
autonomy) that might help you to decrease this positive relationship between job
demands and strain.
Second moderation hypotheses: Job resources (autonomy, social support) are positive
related to motivation at work. This is when job demands are high.
Moderation: JD-R Model
Investigate relationship between job demands and motivational outcomes.
Misbehavior: independent variable
Dedication: motivational indicator -> outcome
Motivation at work decreases when they have less social support. Teachers who have a lot
of support the relationship is less strong. has the function to buffer!
Moderation
It can also influence the direction of the relationship between the independent and
dependent variable.
People who are high in self-esteem they are more likely to speak up at work, but only when
they are high in self esteem and low in self monitoring. When they are high in self
monitoring this relationship is negative.
Crossed moderation (the lines are crossed).
Linear regression model with one interaction term (2-way interaction)
Test the hypotheses:
- Hierarchical Multiple Regression
, o Preferred to examine moderator effects when either predictor (independent
variable) or moderator (or both) are measured on a continuous scale or are
categorical
Formula: independent variable interacts with the moderator. The formula consist of Y (DV)=
Intercept (starting point of your slope) + two components (indicators of the dependent
variable). Set=moderator
Third term: interaction term
You want to know if the interaction term differ significantly from zero. This is how you test
your moderation hypothesis.
Second approach: Whether adding the interaction term to your regression, explains the
additional variance in your dependent variable. Is this addition greater than zero?
If the additional variance is great than zero -> indicator of whether your interaction term is
of well use.
R2: used as an effect size indicator in the regression analyses, it varies between 0 and 1. The
higher, the more variance can be explained by the interaction term.
Two components: Beta1 and Beta2: independent variabele
Third: represents the interaction term -> the heart of the moderation hypothesis
In the end you want to know if the interaction term differs significantly from zero
How do you calculate the interaction term?
- Some software (e.g. R; Process Macro) -> automatically
- Other software (e.g. SPSS): new variable should be calculated
First: center both X and Z around the perspective sample means
Are there any potential control variables that might influence your independent and or your
dependent variable that you might not be interested in. By adding control variables you
would like to rule out alternative explanations for your hypothesis effect.
Adding control variables is important: When you do not have a lot of control over your
variables. So not-randomized, experimental design.
Steps
- Compute cross-product of X and Z
- Regress Y on control variable(s)
o Ruling out alternative explanations for a hypothesized effect
o In research designs that cannot prevent the influence of confounding
variables
- Regress Y on X and Z
- Regress Y on X*Z
Lecture 1 (2x gekeken)
Spreadsheet online tonight -> fill it out!
This is for finding a partner for the assignment
Exam is based on lectures
You need to know and understand what’s on the slides.
Open book exam
Refreshing vocabulary
• 1. True experiment:
22. A design in which participants are assigned randomly to treatments
• 2. Quasi experiment:
5. A design that resembles that of an experiment in that discrete groups are used, but Pp’s
aren’t randomly assigned to treatments nor are treatments randomly determined for groups
• 3. “Between groups" vs. "within groups" design:
6. A treatment between conditions and a measurement referring to differences within the
particpants at different times
• 4. Construct variable:
9. A theoretical variable that has ‘reality status” such as competition, attractiveness,
negative mood
• 5. Operational variable:
10. The researchers’ precise operational definition (i.e., measurement) of a theoretical
construct
6. Independent variable:
12. The variable presumed to cause a change in the dependent variable
• 7. Dependent variable:
2. The variable presumed to be affected by the independent variable
• 8. Hypothesis:
13. A statement of a proposed relation between constructs
• 9. Theory:
4. A well-established principle that has been developed to explain some aspect of the natural
world.
• 10. Construct validity:
19. The degree to which the operational definition accurately measures the construct of
interest
11. Convergent validity:
1. Overlap among variables presumed to measure the same construct
• 12. Discriminant validity:
14. The extent to which it is possible to discriminate between dissimilar constructs
• 13. Random assignment:
15. The process by which subjects receive an equal chance of being assigned to a particular
condition
• 14. Manipulation check:
16. A measured variable designed to assess whether the manipulation worked and tapped a
desired construct
,• 15. Demand characteristic:
17. Aspect of the experiment encouraging the participant to respond according to situational
constraints
16. Reliability:
18. The extent to which a construct is measured without error or bias
• 17. Subject expectancies:
7. A demand characteristic whereby subjects think they know the experimenters’ interests
and act accordingly
• 18. Double-blind procedure:
3. A procedure in which neither experimenter nor pp knows to which condition the pp is
assigned
• 19. Order effects:
21. The effects on behavior of presenting two or more treatments to the same pps.
• 20. Counter-balancing:
8. A technique for controlling order effects by which each condition is presented first to each
participant an equal number of times: present each condition in each possible order
21. Moderation (interaction)
11. The effect of an independent variable on a dependent variable depends on the level of a
third variable (i.e., the moderator)
• 22. Mediation:
20. The effect of an independent variable on a dependent variable is explained by the
change in a third variable (i.e., the mediator)
Moderation & Mediation
Combination = Conditional Process Analysis
- What model you develop depends on the question you want to address & underlying
theory. What is your theoretical model?
- If you want to explain under what conditions something is occurring (e.g. in certain
organizational environments or for certain people)
o Moderation model fits best
- If you want to explain WHY something is happening (i.e. explaining an underlying
mechanism)
o Mediation model fits best
- Combinations (i.e. conditional process analysis) also possible
Example: You are interested in the relationship between emotional job demands and
emotional exhaustion at the work place.
Are there differences between man and woman?
Moderation model
Example: you are interested in why emotional demands relate to emotional exhaustion
Mediation model
You can combine these two models, for example: I want to focus on the relationship
between emotional demands on emotional exhaustion and this is mediated by unfavourable
coping strategies.
This process is stronger for woman then for men
,Combinations: focus on both
Moderation
It tells you about the boundary condition. It can explain under what conditions is the
relationship strong or less strong. It can influence the strength or direction of the
relationship.
Can increase our understanding of the relations between predictors and outcomes
- Moderator variable changes the strength or direction of the relationship between x
and y
For example: I looked into the effects of shared leadership on team creativity for low vs. high
backing-up behavior of team members
Backing up: moderator
- Choice of a moderator and nature of the interaction should be based on theory.
Example slides:
Job Demands-Resources model:
Inherent in this model: true moderation effects.
Various job demands positive relate to strain symptoms at work. Job resources offer
that relationship. If you have certain resources at work (social support, high level of
autonomy) that might help you to decrease this positive relationship between job
demands and strain.
Second moderation hypotheses: Job resources (autonomy, social support) are positive
related to motivation at work. This is when job demands are high.
Moderation: JD-R Model
Investigate relationship between job demands and motivational outcomes.
Misbehavior: independent variable
Dedication: motivational indicator -> outcome
Motivation at work decreases when they have less social support. Teachers who have a lot
of support the relationship is less strong. has the function to buffer!
Moderation
It can also influence the direction of the relationship between the independent and
dependent variable.
People who are high in self-esteem they are more likely to speak up at work, but only when
they are high in self esteem and low in self monitoring. When they are high in self
monitoring this relationship is negative.
Crossed moderation (the lines are crossed).
Linear regression model with one interaction term (2-way interaction)
Test the hypotheses:
- Hierarchical Multiple Regression
, o Preferred to examine moderator effects when either predictor (independent
variable) or moderator (or both) are measured on a continuous scale or are
categorical
Formula: independent variable interacts with the moderator. The formula consist of Y (DV)=
Intercept (starting point of your slope) + two components (indicators of the dependent
variable). Set=moderator
Third term: interaction term
You want to know if the interaction term differ significantly from zero. This is how you test
your moderation hypothesis.
Second approach: Whether adding the interaction term to your regression, explains the
additional variance in your dependent variable. Is this addition greater than zero?
If the additional variance is great than zero -> indicator of whether your interaction term is
of well use.
R2: used as an effect size indicator in the regression analyses, it varies between 0 and 1. The
higher, the more variance can be explained by the interaction term.
Two components: Beta1 and Beta2: independent variabele
Third: represents the interaction term -> the heart of the moderation hypothesis
In the end you want to know if the interaction term differs significantly from zero
How do you calculate the interaction term?
- Some software (e.g. R; Process Macro) -> automatically
- Other software (e.g. SPSS): new variable should be calculated
First: center both X and Z around the perspective sample means
Are there any potential control variables that might influence your independent and or your
dependent variable that you might not be interested in. By adding control variables you
would like to rule out alternative explanations for your hypothesis effect.
Adding control variables is important: When you do not have a lot of control over your
variables. So not-randomized, experimental design.
Steps
- Compute cross-product of X and Z
- Regress Y on control variable(s)
o Ruling out alternative explanations for a hypothesized effect
o In research designs that cannot prevent the influence of confounding
variables
- Regress Y on X and Z
- Regress Y on X*Z
