Summary for Research Methods
Variables
Independent variable: A variable that is being examined to determine whether it influences
another variable.
Dependent variable: A variable that is being examined to determine whether it is influenced
by another variable.
Quantitative variables: values express an amount; discrete or continuous (age)
Qualitative variables: values express categories (2 or more levels) ( 1=male; 2=female)
Confounding variable: A variable that influences the dependent variable and is associated
with the independent variable, although it is not a consequence of it. Z influences X and Y is
a special form of confounding variable where Z influences both IV and DV
Intermediator variable: A variable that is influenced by the independent variable and that
influences the dependent variable. IV→Intermediator variable→DV
Alternative explanation: X→Y or Y→X; Z---X and Z→Y
Positive, negative or curvilinear relationships. Or no relationship.
Designs
Within-subjects design: The design of a study examining different conditions with the same
individuals. Each participant experiences all conditions. There are several problems that can
occur with the within-subjects design:
1. Order effect→the order of presenting the treatments affects the dependent variable
2. Practice effect→performance on the second task might improve merely because of
the practice gained on the first task
3. Fatigue effect→participant becomes bored, tired or distracted
4. Carryover effect→the impact of the first treatment can carry over to influence the
response to the second treatment
Between-subjects design: The design of a study examining different conditions with
different individuals. Each participant receives one condition.
Mixed design: The design of a study that combines a between-subjects design with a within-
subjects design.
→ Independent variable 1: Condition (A vs. B) Independent variable 2: Time (before vs. after)
Dependent variable: Measurement
→ Independent variable: Condition (A vs. B); Dependent variable: Improvement = Post-test –
Pre-test
Randomization
Randomization of groups: The practice of splitting the sample into multiple groups based on
chance. A different condition is administered to each group. The participants are assigned to
the various conditions ‘at random’.
Intra-group counterbalancing: A ‘balanced’ list of orders is created. Each participant
receives one of the orders on the list at random. So in the group all conditions are
administered in exactly the same phase, on average. Each participant receives the condition
one time.
Intra-group randomization of order: The order is determined at random for each participant
anew. In the group, all conditions are administered in approximately the same phase, on
average. Each participant receives the condition one time.
, Intra-subject counterbalancing: Each participant receives the conditions in multiple,
sophisticated orders. So for each participant, all conditions are administered in exactly the
same phase, on average. Each participant receives the condition multiple times.
Intra-subject randomization of order: Each participant receives each condition very often in
a random order. For each participant, all conditions are administered in approximately the
same phase, on average. Each participant receives the condition multiple times.
Research Designs
Non-experimental study: Research in which the independent variable is not manipulated,
but is only measured.
True experiment: A study that demonstrates that the independent variable influences (or
does not influence) the dependent variable. These things are needed:
-manipulation of the independent variable;
-randomization of groups (between-subjects design) or
-randomization of order / counterbalancing (within-subjects design);
-elimination of all potentially confounding variables;
-post-measurement of the dependent variable;
-comparison of the conditions regarding the mean of the dependent variable.
Quasi-experiment: A study in which it is established (albeit not with 100% certainty) that the
independent variable influences the dependent variable. Requirements:
-manipulation of the independent variable
-both treatment group and control group, albeit non-randomized
and
-both a pre-test and a post-test of the dependent variable
-comparison of the treatment group to the control group with regard to
progress (post-test minus pre-test)
Pre-experiment: A study suggesting that the independent variable influences the dependent
variable. Requirements:
-manipulation of the independent variable
-treatment group and control group, albeit non-randomized
or (but not both)
-both a pre-test and a post-test of the dependent variable
-comparison of the treatment group to the control group or of the pre-
test to the post-test
Case studies/N=1: One participant. Non-experimental study. Intra-subject randomization of
order→
1. Reversal design: A (baseline period); B (treatment period); A (baseline period)
2. Multiple baseline designs: the effectiveness of the treatment is demonstrated when a
behavior changes only after the manipulation is introduced; such change must be
observed under multiple circumstanced to rule out the possibility that other events
were responsible
Creating balanced Latin square: Each condition is presented once at each time and each
condition is used with equal frequency before each other condition and with equal
frequency after each other condition.
First row: A B L C L-1 D L-2 E L-3 F etc. and L represents the last condition in your study.
Field experiment: In a natural setting. Researcher loses ability to control many aspects of the
situation.
Pre-experiments:
Variables
Independent variable: A variable that is being examined to determine whether it influences
another variable.
Dependent variable: A variable that is being examined to determine whether it is influenced
by another variable.
Quantitative variables: values express an amount; discrete or continuous (age)
Qualitative variables: values express categories (2 or more levels) ( 1=male; 2=female)
Confounding variable: A variable that influences the dependent variable and is associated
with the independent variable, although it is not a consequence of it. Z influences X and Y is
a special form of confounding variable where Z influences both IV and DV
Intermediator variable: A variable that is influenced by the independent variable and that
influences the dependent variable. IV→Intermediator variable→DV
Alternative explanation: X→Y or Y→X; Z---X and Z→Y
Positive, negative or curvilinear relationships. Or no relationship.
Designs
Within-subjects design: The design of a study examining different conditions with the same
individuals. Each participant experiences all conditions. There are several problems that can
occur with the within-subjects design:
1. Order effect→the order of presenting the treatments affects the dependent variable
2. Practice effect→performance on the second task might improve merely because of
the practice gained on the first task
3. Fatigue effect→participant becomes bored, tired or distracted
4. Carryover effect→the impact of the first treatment can carry over to influence the
response to the second treatment
Between-subjects design: The design of a study examining different conditions with
different individuals. Each participant receives one condition.
Mixed design: The design of a study that combines a between-subjects design with a within-
subjects design.
→ Independent variable 1: Condition (A vs. B) Independent variable 2: Time (before vs. after)
Dependent variable: Measurement
→ Independent variable: Condition (A vs. B); Dependent variable: Improvement = Post-test –
Pre-test
Randomization
Randomization of groups: The practice of splitting the sample into multiple groups based on
chance. A different condition is administered to each group. The participants are assigned to
the various conditions ‘at random’.
Intra-group counterbalancing: A ‘balanced’ list of orders is created. Each participant
receives one of the orders on the list at random. So in the group all conditions are
administered in exactly the same phase, on average. Each participant receives the condition
one time.
Intra-group randomization of order: The order is determined at random for each participant
anew. In the group, all conditions are administered in approximately the same phase, on
average. Each participant receives the condition one time.
, Intra-subject counterbalancing: Each participant receives the conditions in multiple,
sophisticated orders. So for each participant, all conditions are administered in exactly the
same phase, on average. Each participant receives the condition multiple times.
Intra-subject randomization of order: Each participant receives each condition very often in
a random order. For each participant, all conditions are administered in approximately the
same phase, on average. Each participant receives the condition multiple times.
Research Designs
Non-experimental study: Research in which the independent variable is not manipulated,
but is only measured.
True experiment: A study that demonstrates that the independent variable influences (or
does not influence) the dependent variable. These things are needed:
-manipulation of the independent variable;
-randomization of groups (between-subjects design) or
-randomization of order / counterbalancing (within-subjects design);
-elimination of all potentially confounding variables;
-post-measurement of the dependent variable;
-comparison of the conditions regarding the mean of the dependent variable.
Quasi-experiment: A study in which it is established (albeit not with 100% certainty) that the
independent variable influences the dependent variable. Requirements:
-manipulation of the independent variable
-both treatment group and control group, albeit non-randomized
and
-both a pre-test and a post-test of the dependent variable
-comparison of the treatment group to the control group with regard to
progress (post-test minus pre-test)
Pre-experiment: A study suggesting that the independent variable influences the dependent
variable. Requirements:
-manipulation of the independent variable
-treatment group and control group, albeit non-randomized
or (but not both)
-both a pre-test and a post-test of the dependent variable
-comparison of the treatment group to the control group or of the pre-
test to the post-test
Case studies/N=1: One participant. Non-experimental study. Intra-subject randomization of
order→
1. Reversal design: A (baseline period); B (treatment period); A (baseline period)
2. Multiple baseline designs: the effectiveness of the treatment is demonstrated when a
behavior changes only after the manipulation is introduced; such change must be
observed under multiple circumstanced to rule out the possibility that other events
were responsible
Creating balanced Latin square: Each condition is presented once at each time and each
condition is used with equal frequency before each other condition and with equal
frequency after each other condition.
First row: A B L C L-1 D L-2 E L-3 F etc. and L represents the last condition in your study.
Field experiment: In a natural setting. Researcher loses ability to control many aspects of the
situation.
Pre-experiments:
