Experimental Psychology: Exam 1
Variance and how it's measured
A descriptive statistic that indicates the amount of observed variability in participants behavior, tells us how much scores cluster around the mean.
A theory?
A formalized set of propositions that attempt to specify the interrelationships among a set of concepts. (usually is a statement about the causal relation between 2 or more variables)
What is the difference between a hypothesis and a theory?
A hypothesis attempts to answer questions by putting forth a plausible explanation that has yet to be rigorously tested. A theory, on the other hand, has already undergone extensive testing by various scientists and is generally accepted as being an accurate explanation of an observation.
Hypothesis?
A proposition that follows logically from a theory; also, a predicition regarding the outcome of a study.
Authority
Accept because it comes from respected source (e.g., "the boss says its true so it must be so." Problem: authorities can be wrong
Tenacity
Accept ideas as true because they have been accepted for so long. (e.g., it has always been that way)
Crobach's alpha coefficient
An index of interitem reliability. Equivalent to the average of all possible split-half reliabilities (an index of interitem reliability)
Science
Combines rationalism and empiricism.
The difference between convergent and discriminant validity
Convergent is documenting the validity of a measure by showing that it correlates appropriately with measures of related constructs and discriminant is documenting the validity of a measure by showing that it does not correlate with measures of conceptually unrelated constructs.
Difference between descriptive stats and inferential stats
Descriptive summarizes the data, (averages, percentages, etc) and inferential draws conclusions (t-tests, f-tests, etc)
What are the properties of the number system that are important when measuring psychological variables?
Do not always have all four properties: identify, magnitude, equal interval, true zero (not always there in psych): nominal, ordinal, interval, ratio.
Criterion-related validity
The extent to which a measure allows a researcher to distinguish among respondents on the basis of some behavioral criterion
Effect size
The proportion of total variance that is systematic
Falsifiability?
The requirement that a hypothesis must be capable of being falsified.
Why does science need both rationalism and empiricism? Why is neither approach sufficient by itself?
Theory is empty without empirical support (empiricism), data is not interpretable without theory (rationalism). Science is a cyclical process, theorize-->observe-->(circle)
Rationalism
Through reasoning and by using "rules of logic" (it makes sense to me" Problem: things are not always as logic would dictate, especially when studying human behavior"
Variability
Variations in behavior .
What is parsimony and why is it important to science?
When faced with two competing theories that do an equally good job of explaining the observations, we should prefer the simpler one. It's important because it steers scientists away from unnecessary concepts.
Physiological measures and 4 measures
a measure of bodily activity; in behavioral research, physiological measures generally are used to assess processes within the nervous system. Measures: neuro-imaging, measures of autonomic nervous system activity, blood and salvia assays, precise measurement of overt reactions.
Participant observation
a method of data collection in which researchers engage in the same activities as the participants they are observing
Measurement
assigning numbers that represent values of the variable, it is an attempt to match the properties of an real number system with properties of the variable of interest.
Double-barrled question
attempt to get at multiple issues at once
Difference between concurrent and predictive validity
concurrent is a form of criterion related validity that reflects the extent to which a measure allows a researcher to distinguish between respondents at the time the measure is taken. Predictive is the form of criterion related validity that reflects the extent to which a measure allows a researcher to distinguish between respondents at some time in the future.
Operational Definitions
defines a concept by specifying precisely how the concept is measured or induced in a particular study.
Observational methods
direct observation of human behavior
Basic Research
is conducted to understand psychological processes without regard for whether or not the knowledge is immediately applicable.
Applied Research
is to find solutions for certain problems rather than to enhance general knowledge about psychological processes.
Self-report measures
measure on which participants provide info about themselves, on a questionnaire or in an interview, for example.
How is reliability related to systematic and total variance?
measures must be reliable in order to be valid..? finish
Reliability
refers to the consistency or dependability of a measuring technique.
Face validity and its problems
refers to the extent to which a measure appears to measure what it's supposed to measure. Problems: subjective andinaccurate because its looks valid doesnt mean it is valid. Researchers may want measures low in face validity (to disquise true purpose of the study)
Difference between systematic variance and error variance
systematic variability in behavior that is caused by the variables of interest and error is variability in behavior that is unrelated to variables of interest
Construct validity
the degree to which a measure of a particular construct correlates as expected with measures of other constructs.
3 Goals of Science
to describe, predict and explain behavior.
Empiricism
"I observed it to be true" Make systematic observations, gained through observation (our sense) Problems: our senses can be wrong, observation shows what but not how
How are reliability and validity related?
"reliability is a necessary but not a sufficient condition for validity" Measures must be reliable in order to be valid. But reliable measures can still be invalid assessments.
Five propositions about variability
1. All behavioral research attempts to explain behavioral variability. 2. Research questions are all questions about behavioral variability. 3. research studies must be designed to answer questions about variability. 4. Measurement of behavior involves assessment of variability. 5. Statistical analyses are used to describe and account for observed variability in behavioral data.
What are the seven phases to the research process?
1. Idea generating phase: ideas coming from many places 2. Problem-Definition Phase: vague ideas are formalized into testable research questions. 3. Procedure-Design Phase: must translate your ideas into testable hypotheses. 4. Observation Phase 5. Data-Analysis Phase 6. Interpretation Phase- making sense out of the results 7. Communication Phase-must report in APA style, must describe in enough detail so that someone else can replicate the study.
What are the three criteria for Scientific Investigation? Why is each one important?
1. Systematic Empiricism: relying on observations to draw conclusions about the world (senses) 2.Public Verification: research much be conducted so that findings of one researcher can be observed, replicated and verified by others. (it's important because perceptions are biased, senses are subject able) 3. Solvable problems: Questions must be answerable given current knowledge and research techniques (has to be falsifiable, to be able to be proved wrong)
What are the two jobs of a scientist?
1. to DISCOVER new phenomena- relics on empiricism, we often get our research ideas by observing those around us. 2. To EXPLAIN these phenomena- relics on rationalism and requires knowledge through reasoning and by using "rules of logic"
How do we improve reliability?
?
What is the difference between induction and deduction? What role does each one play in the scientific process?
Inductive reasoning goes from specific to general, is data driven and starts with observation to theory. Abstracting a hypothesis from a collections of facts. "post hoc theory". Deductive reasoning goes from general to specific and is theory driven. A process of reasoning from a general proposition (the theory) to specific implications of that proposition (the hypothesis).
Why are most scientists not content to stop at the first two goals?
Just because people believe it, doesn't make it so. Findings are often counterintuitive. Scientists must question their own common sense assumptions.
Intuition
Knowledge just comes to you without thought or experience. (e.g., "it feels true") Problem: our perceptions are biased.
difference between naturalistic observation and contrived observation
Naturalistic observes behavior in real-world setting where as contrived is a setting arranged for observing and recording behavior.
Properties of the 4 scales of measurement (nominal, ordinal, interval and ratio)
Nominal: each number reflects a category, e.g., eye color; ordinal: Rank order, # reflects the order but not the amount, e.g., olympic medals, class rankings. Produces ordered data. Interval: Equal intervals, provides info about absolute differences between stimuli, e.g., temperature, SAT/GRE. Most psychological variables are interval scale. Ratio: interval scale w/o true zero, e.g., time, distance, frequency. Produces score data.
Difference between undisguised observations an disguised observations
Nondisguised is participants know they are being observed and disguised is obvering behavior without participants knowledge. People behave differently when being observed.
Differences between observational, physiological and self-report measures
Observational involves direct observation of behavior (observe non-verbal flirting behavior at a bar). Physiological records internal processes not directly observable (heart rate). Self-Report measures are replies people give on questionnaires or interviews.
How do the observed score, true score and measurement error relate to each other?
Observed score is the true score plus the measurement error and the true score is the score that the participant would have obtained if our measure were perfect and we were able to measure whatever we were measuring without error.
Why is proof impossible? If we cannot prove, then what do we do in science?
Our observation can confirm or disconfirm a hypothesis but we can never prove a theory is true. Why? Even if 100 observations support your theory, the next one you make could prove it wrong. Scientists never say prove. Disproof does not necessarily mean the death of a theory. Often, we revise theories and then test again.
Difference between post hoc theory and priori theory? Which one is preferred?
Post hoc are explanations that are made after the fact, (provides little evidence of its accuracy or usefulness). Researchers can often explain findings post hoc that they would not have predicted in advance of conducting the study. Priori is a convincing test of a theory, specific predictions about what will occur in a study, this is the preferred method.
Validity
Refers to the extent to which a measurement procedure actually measures what it is intended to measure rather than measuring something else (or nothing at all).
Archival Research
Research that uses data collected prior to the time the research was designed. Researchers analyze data pulled from existing records, such as census data, court records, etc.
Social desirability, acquiescence and nay-saying
Social desireability is where participants respond in a socially acceptable manner, can lower validity. Acquiescence is where some people show a tendency to agree with statements regardless of the content whereas other tent to express disagreement (naysaying)
What is the difference between test-retest, interitem and interrater reliability?
Test-retest reliability refers to the consistency of participants responses on a measure over time. Interitem reliability assesses the degree of consistency among the items on a scale. Interrater reliability is the degree to which the observations of two independent raters or observers agree; also called interjudge or inter observer reliability.
Item-total correlation
The correlation between respondents scores on one item on a scale and the sum of their responses on the remaining items; an index of interitem reliability.
Measurement Error
The deviation of a participants observed score from his or her true score.