Test 3 Flashcards
The process of determining the characteristics and/or quantity of a variable through systematic recording and organization of observations
Measurement
Four levels of measurement
Nominal, Ordinal, Interval, Ratio
Level of measurement that is:
- Qualitative
- Has no order
- Numbers do not carry any numerical value
- Assigned numbers are used to label
Nominal Level
Level of measurement that is:
- Quantitative
- An ordered category or rank
- Can determine if an observation is greater than, less than or equal to other observations
- Does not indicate how much the difference is
- The amount separating the levels is unknown
Ordinal Level
Level of measurement that is:
- Quantitative
- Specifies relative position
- Assumes equal distance between points on the scale
- No true zero
- Impossible to make proportional statements because zero is meaningless
Interval Level
Level of measurement that is:
- Quantitative
- The most specific type of measurement
- Has a true and meaningful zero
- Can be used to measure a type of behavior of a participant
Ratio Level
What are the four scales of the Interval Level?
Likert, Semantic Differential, Guttman/Cumulative, Multi-Item
Scale that measures a participant’s feelings or attitudes toward another person, issue or event
Likert Scale
Scale that measures the meanings participants assign to some stimulus
Semantic Differential
Scale that measures from broad to more specific
Guttman/Cumulative
Scale that cannot be truly measured
Multi-Item Scale
When something expresses accuracy/truthfulness, it has:
Validity
Validity that makes sense on its face
Face Validity
Validity that deals with how a particular measure holds up when compared to some outside criterion
Criterion Validity
How well a measure predicts that something will happen in the future
Predictive Validity
How well a scale measures up against another scale that has been demonstrated to measure exactly the same thing
Concurrent Validity
The extent to which your variables are logical related to other variables; constructs relate in a logical way
Construct validity
When two measures you expect to be related are shown to be positively statistically related
Convergent Validity
To what extent you can replicate the same results
Reliability
Reliability measured when you give a survey to the same participants at different times
Test/Retest Reliability
An indicator of how similarly codes are coding content both in terms of identifying units of analysis and in the contextual labels they ascribe to those units
Inter-Coder Reliability
The order in which the items in a measure are presented affect the ways in which people respond
Alternate Form Reliability
A means of evaluating internal consistency of a scale that compares the total score for a scale with individual item scores for the same scale
Item-Total Reliability
A means of evaluating internal consistency of a scale that compares one randomly selected half of a scale from the other randomly selected half of the same scale
Split-Half Reliability
Four types of questions on a survey
Frequency items, nominal items, ordinal items, Likert scale items
Survey question that:
- Tells the researcher how often someone does something
- Can suggest ranges
Frequency items
Survey question where respondents self- report things that only they can know about themselves
Nominal Self Items
Survey question where respondents report their perceptions of a person or an object
Nominal Items
Survey questions where respondents reveal how they rate or rank related objects
Ordinal Items
Scales on a survey that tell researchers how strongly someone likes or dislikes something
Likert Scale Items
The entire group of interest in a research project
Population
A subset group of the chosen population
Sample
What is used by researchers to control variables?
Experiments
Experiments that:
- Have manipulation of independent variables
- Have random assignment of participants
- Have a control group
Full Experiment
Experiments that:
- lack random assignment and/or a control group
- are designed to rule out history and maturation as validity threats
- uses pretests and post-tests in more complicated ways
Quasi-Experiment
Three Conditions for Establishing Causation
- Independent Variable must precede Dependent Variable
- IV and DV must covary
- Alternative causation must be ruled out to ensure that changes in the DV came from changes in the IV
Three Conditions for Full Experiments
- Independent Variable
- Randomly divided assignment of participants
- Always have a control group
A likelihood of relationship between two variables
Correlation
Implies that one variable causes a change in the direction of the other variable
Causation
Group in an experiment that receives no manipulation
Control Group
Group in an experiment that receives fake manipulation
Placebo Group
Group in an experiment that receives complete manipulation
Experimental group
7 Main Threats to Validity
History, Hawthorne effect, Maturation, Sleeper Effect, Ceiling/Floor Effects, Sensitization, Mortality
A threat to validity issue where something totally unrelated to your study happened at a particular time and may have affected the responses
History
A validity issue dealing with the fact that subjects can change over time, which can affect their responses to the measures a researcher is interested in
Maturation
Validity issue phenomenon that relates to persuasion; delayed increase of the effect of a message that is accompanied by a discounting cue
Sleeper Effect
Validity issue where participants develop a strategy to do better/worse in the experiment
Sensitization
Validity issue where participants may have moved away, decided to drop out or got bored
Mortality
Validity issue where the tendency for participants to change their behavior or responses when they know they are being observed
Hawthorne Effect
Validity issue where participants scores cluster toward the high end of the measurement
Ceiling Effect
Validity issue where participants scores cluster toward the low end of the measurement
Floor Effect