LECTURE 4 Flashcards
The dimension that the researcher intentionally
manipulates
➔ It is “independent” in a sense that its values are created by
the experimenter and are not affected by anything else that
happens in the experiment
➔ List of potential IVs are endless (e.g., physical environment,
aspects of a given task, psychological states, intervention)
Independent variable
▪️ Refers to the situation wherein an extraneous variable changes systematically across different conditions of an experimen
▪️ It sabotages an experiment since we might not be able to
tell whether the changes in DV is caused by our IV or by the
confounding
▪️ One of the most important activities in research
Confounding
CONSTRUCT VALIDITY
▪️ ____ - test results should not highly
correlate with scores on tests measuring other constructs
Discriminant validity
CONSTRUCT VALIDITY
▪️ ___ - test results should highly correlate
with scores on tests measuring the same or similar construct
Convergent validity
▪️ It deals with the transition from theory to research application
Construct validity
CRITERION VALIDITY
___ reflects whether scores on a
measuring device correlate with scores obtained from
another method of measuring the same concept
Concurrent validity
CRITERION VALIDITY
▪️ ____ - refers to the ability of the instrument to predict future behavior or performance based on the score obtained from it
Predictive validity
▪️ Represents how the content of our measure fairly reflect the
content of the quality we are measuring
▪️ Consensus on the content validity can be obtained through
various experts who will rate each item of the instrument
▪️ might be reduced if the language of the instrument is difficult to understand for people who will take it.
Content validity
▪️ A measured operational definition has ___ if it
appears to measure what it intends to measure
▪️ is considered the least stringent type of validity
Face validity
▪️___ in experiments refers to the principle of actually studying the variables that we intend to study
▪️ An operational definition can be reliable, yet invalid
▪️ ____ - providing evidence of validity of an experimental procedure
Validity
Manipulation check
▪️ The extent to which different parts of a questionnaire, test, or other instruments attain consistent results; measures the internal consistency
▪️ It is most often used if the multiple-item instrument
measures a single construct variable.
Interitem reliability
▪️ Refers to the reliability obtained through comparison of scores of people who have been measured twice with the same instrument given that there is a reasonable interval
▪️ Reliable measures should produce very similar scores each
time the person is measured
Test-retest reliability
▪️ If there is little agreement, it is more likely that the
measuring procedure is not reliable
Interrater reliability
▪️ Refers to the agreement between measurements of the same responses from different observers
▪️ Typically, this method is used in content analysis
Interrater reliability
▪️ An experimental operational definition is considered reliable if each time we create the conditions, we get similar consequences.
Reliability
defines a variable in terms of
observable operations, procedures, and measurements in the context of an experiment
Operational definition
▪️ Occurs when a selection threat combines with another threat
▪️ If subjects were not randomly assigned to treatment conditions, any one of other threats may have affected some experimental groups but not others.
Selection interaction
▪️ Refers to the occurrence of subjects dropping out of the experiment
▪️ It should be always considered whenever
the attrition rate of a particular treatment is high compared to others - something about the treatment could be making them
Subject mortality
▪️ Occurs whenever the researcher does not assign subjects randomly to the different conditions of an experiment
▪️ Random assignment is done to prevent systematic differences between subjects of one treatment condition to the subjects of another treatment.
▪️ Example: Subjects in the “Tutoring Group” are not randomly assigned;
instead, they are selected by their experimenter based on their perceived
need for tutoring.
Selection
Example:
▪️ Subjects who had the highest performance in the math exam obtained lower scores after the pretest.
▪️ Students who had extremely low motivation scores presented a slight increase in motivation after the pretest.
Statistical regression
▪️ Also known as the regression toward the mean
▪️ Statistically, extreme scores tend to have less test-retest reliability than do moderate scores
▪️ Scores at both low and high extremes typically get closer to
the mean even without treatment at all.
Statistical regression
▪️ Occurs when some features of the measuring instrument itself changes during the experiment
Example:
▪️ Photocopy of questionnaires were clearer in posttest than the pre-test.
▪️ Weighing scale malfunctioned due to repeated use.
Instrumentation
▪️ Refers to the effects on the DV produced by a previous administration of the same test or other measuring
instrument
▪️ Example: Subjects performed better on the post- training memory test not because the spelling mnemonics technique (IV) employed, but simply because they became more familiar with the test (format, or the specific items).
Testing
▪️ Refers to any internal (physical or psychological) changes in
subjects that might have affected the scores on the DV
▪️ Experimental studies with the within-subjects design are prone to this classic threat (e.g., boredom and fatigue)
▪️ Studies that take months or even years to finish are also
prone to maturation (e.g., learning)
Maturation
▪️ Refers to the history of the experiment, it involves outside events that happened before the experimentation which could have caused the experimental effects.
▪️ Example: Subjects were sleep deprived before experimentation due to their scheduled midterm examination the next day.
History
▪️ Proposed by the psychologist ____, these are the eight kinds of extraneous variables that can threaten the internal validity of an experiment
Donald Campbell
▪️ All quasi-experimental studies contain one or more classic
threats, and it is the reason why they are not considered as _____ - they are potentially confounded
True experiments
