Home / Summaries / Class notes - Methodology: Operationalization, Design and Analyses / variable-independent-design

(Quasi-) Experimental designs

30 important questions on (Quasi-) Experimental designs

What does experimental design involve?

Assignment of participants to different conditions
manipulaton of one or more variables by experimenter
measurement of effects of this manipulation on one or more other variables
the control of all other variables

How can we operationalize an independent variable?

You can measure it
you can manipulate it (induce it), active change.

What kind of variable is the dependent variable?

The variable that the researcher measures for changes in order to determine the effects of manipulating the independent variable

What types of manipulations are there? Of the independent variable

Stimulus manipulation (different groups, different stimuli)
Instructional manipulation (different groups, different instructions)
Contextual manipulation (different groups, differenct situation)

What types of manipulation are there?

Threat manipulation
Feedback manipulation
uncertainty manipulation

What kind of measures are there for de dependent variable

Subjective
Cognitive
Behavioral
neurobiological (MRI)
Physiological (hartslag)

What do you need to concider when you measure the dependent variable ?

If you measure it explicit of implicit

What are Ceiling effects when measuring the dependent variable?

When the Dependent variable reaches a level that canno be exceeded. So while there may be an effect of the IV, that effect can't be seen beceause everybody has hit the maximum.
Een taak is zo makkelijk dat iedereen 100% scoort.

What three conditions are necesarry to claim that variable A causes changes in variable B?

Relationship condition (variable A and variable B must be related)
Temporal antecedence condition (proper time order must be established)
Lack of alternative explanation condition (no confouding variable)

What is a extraneous variable?

This is a third variable that may compete with the independent variable explaining the effects on the dependent variable.

Why are extraneous variables undesirable?

Beceause they add error to an experiment

There are two standard methods for controlling extraneous variables;

Control by randomization
Control by holding constant or matching

What does control by randomization mean?

Allocation of participants to groups in random way
it offers a method for controlling a multitude of variables simultaneously and does not require specific attention to each extraneous variable

What is the goal of randomization?

The goal is to disrupt any systematic relation between extraneous variables and the independent variables and thereby prevent the extraneous variable from becoming a confounding variable

How can the impact of an extraneous variable on the results be eliminated?

By holding it constant.

How can the impact of an extraneous variable on the results be minimized ?

By matching the levels of the variable across treatment conditions. Bijv; dezelfde hoeveelheid mannen als vrouwen.

What is important to take in mind with a manipulation check after the manipulation?

If you do the manipulation check right after the manipulation make sure that you do not inadvertently also then prime the non-experimental or control conditions.

What type of experimental designs are there?

1. Two group designs, vergelijken van twee groepen
2. three-(or more) group designs
3. Factorial designs
4. Parametric designs
5. Matched pairs design
6. Repeated measures designs

How many levels do you have with a parametric design?

Several 'levels' of an independent variable covered with random allocation of participants to groups to get a view of the effect of the independent variable over a range of values

What will you do to the participants with a matches pairs design?

Establishing pairs of participants with similar scores on a variable known to be related to the dependent variable. Random allocation of members to different experimental groups.

What does a repeated measures design mean?

Designs where the same participant is tested under two or more experimenal conditions ( an extreme matched pairs design)

When do you use a simple two group design?

When orders effects are likely
Independent variables are not suitable for repeated measurement
In real life, people wouldn't often receive multiple treatments
People might be expected to be sensitized by pre-testing on a matching variable

When do you use a factorial design?

You are interested in more than one independent variable
Interactions among independent variables may be of concern or when researchers are specifically interested in these interactions (moderation model)

When do you use a parametric design?

The independent variable has a range of values or levels of interest
You wish to investigate the form of nature of the relationship between independent and dependent variables.

When to use a matched pairs design?

You have a matching variable which correlates highly with the dependent variable
obtaining the score on the matching variable is unlikely to influence the treatment effects
individual differences between subjects are likely to mask treatment effetc (effects of independent variable)

When do you use a repeated measures design?

Order effects appear unlikely
The independent variable lend themselves to repeated measurement
In real life, people would be likely to be exposed to different treatments
Individual differences between subjects are likely to mask treatment effects

When do you use a control group?

It is not clear what the 'normal' level of the dependent variable is
You're interested in comparing the treatment to a normal population

What are default's in the mediation design?

Measurement in itself might interfere with how the process leads to the effect of interest.
Mediation analysis is correlational, a third variable explains the relationship
Mediator & DV are very similar ( is there discriminant validity)
Possible interaction with independent variable

How many steps do you have with a experimental causal chain design?

2 (or more) experiments/steps;

IV causes MED
MED causes DV

How does your design look with a moderation of process design?

1 experiment with 2 manipulations, IV and mediator.

The question on the page originate from the summary of the following study material:

Methodology: Operationalization, Design and Analyses

View summary

A unique study and practice tool
Never study anything twice again
Get the grades you hope for
100% sure, 100% understanding

Remember faster, study better. Scientifically proven.