Psych Research Methods Exam 2 Vocab

Front 1

Experiment

Back 1

Systematic research study in which the investigator directly varies some factor(s), holds all others constant, and observes the results of the variation.

Front 2

Field Experiment

Back 2

Experiments that take place in the field.
Field RESEARCH is a broader term for any empirical research outside the lab, including field studies and studies using nonexperimental methods.

Front 3

Types of manipulated independent variables

Back 3

Situational - features in the environment that participants might encounter. Like bystander effects.
Task - Vary type of task. Different types of problems.
Instructional - Telling different groups to perform a task but the HOW is different.

Front 4

Extraneous Variables

Back 4

Variables not of interest to the researcher but that might influence behavior being studied if not controlled properly.

a CONFOUND is an uncontrolled extraneous variable that covaries with the independent variable (varies at the same time)

Front 5

Ceiling and Floor effect

Back 5

Ceiling - Average scores for the groups in a study are so high that no difference can be determined between conditions.
Like making a test so easy everyone - including dum-dumbs - scores well.
Floor - opposite

Makes it difficult to find a difference between conditions

Front 6

Subject variables

Back 6

Ex post facto, natural group, nonmanipulated, quasi-experimental, and participant variables.

Existing characteristics of individuals participating in the study.

In other words, the independent variable is SELECTED, not MANIPULATED

Front 7

Statistical conclusion validity

Back 7

The extent to which the researcher uses statistics properly and draws the appropriate conclusions from statistical analysis.

Front 8

External validity

Back 8

The degree to which research findings generalize beyond the specific context of the experiment being conducted.

Highest external validity means being able to generalize to other populations, environments and times.

Front 9

Ecological validity

Back 9

relevance to everyday cognitive activities of people trying to adapt to their environment.

A type of environmental validity

Ulric Neisser, Cornell

Front 10

Internal validity

Back 10

The degree to which an experiment is methodologically sound and confound-free.

Front 11

Threats to internal validity - Time-based (pre/post-testing)

Back 11

1. History
2. Maturation - People change
3. Regression to the mean. Larger sets of scores tend to diminish the extremity of extremes because as you get more scores, majority will cluster around the mean.
4. Testing - pretesting influences posttesting. Each time a test is performed subjects become more sensitized to the test.
5. Instrumentation - problem when measurement instrument changes.

Front 12

Threats to internal validity - Participant-based

Back 12

1. Subject selection effects - unequivalence of groups. Matching based of pretests, like Brady's monkey experiment.
2. Attrition/Subject mortality - Groups of people are prone to dropping out as the experiment goes on (old people dying). Group becomes less representative.

Front 13

Four types of validity

Back 13

1. Statistical conclusion - proper data analysis
2. Construct - defining variables meaningfully and precisely
3. Internal - free of confounding, methodologically sound
4. External - Generalizability

Front 14

Between vs. Within-subjects designs

Back 14

Between is when each group is exposed to different conditions
Within, also referred to as repeated-measures, is when one group is exposed to all levels of manipulation.

Front 15

Techniques to create equivalent groups

Back 15

1. Random assignment - when every subject has an equal chance of being placed in a group
2. Matching - Participants are grouped based off some characteristic(s) (matching variable). Groups are commonly matched based off age.
3. Controlling Order effects - counterbalancing

Front 16

Counterbalancing techniques

Back 16

Participants tested once per condition in a within-subjects design.
1. Complete - Ensure every sequence is used at least once.
2. Partial/incomplete counterbalancing - Subset of all possible sequences is used
3. Latin Square - A super-anal version of complete counterbalancing. Ensures a) every condition of the study occurs equally often in every sequential position and b) every condition follows and precedes every other condition exactly once.

More than once per condition
1. Reverse counterbalancing - present conditions in order, then the opposite order
2. Block randomization - conditions are in a random order, and they don't reset until every condition has occurred.