-
Explains phenomena and can be falsified, involves entities unsee. Explains something.
Theory
-
Statement about the relationship between observable variables.
Hypothesis
-
If P is true than Q is true, P is true therefore Q is true.
Modus ponens
-
If P is true than Q is true, Q is not true, therefore P is not true.
Modus Tollens
-
If P is true than Q is true, Q is true, therefore P is true.
Affirming the consequent
-
propositional logic that is valid but not useful to science.
Modus Ponens
-
Propositional logic that is both valid and useful to science.
Modus Tollens
-
Propositional logic that is neither valid nor useful to science.
affirming the consequent
-
Values of the IV that the researcher chooses to use in an experiment.
levels of the independent variable
-
you can have causation
Advantage of the experimental approach
-
Scale of measurement that occurs when different entities receive different scales.
identity
-
Scale of measurement that occurs when the ordering of values reflects the ordering of the trait being measured.
Magnitude
-
Scale of measurement that occurs when a difference of 1 on a scale represents the same amount of the trait being measured everywhere on the scale.
Equal intervals
-
Scale of measurement that occurs when 0 on the scale represents a complete absence of the trait being measured.
Absolute zero
-
Measurement scale that has only the identity property.
Nominal scale
-
football jerseys, zip codes
nominal scale
-
scale in which no arithmetic operations are meaningful
nominal scale, ordinal scale
-
Measurement scale that has only the identity property and the magnitude property.
ordinal scale
-
basketball rankings, class rankings
ordinal scale
-
Measurement scale that has only the identity,magnitude and equal interval properties.
interval scale
-
Fahrenheit, Celsius,
interval scale
-
scale that can be added and subtracted with meaning
interval scale
-
Measurement scale that has the identity, magnitude, equal intervals, and absolute zero.
ratio scale
-
weight in pounds, # correct on test, Calvin temp. age
ratio scale
-
scale where all arithmetic operations are allowed
ratio scale
-
Occurs when all members in a population have equal opportunity of being chosen for the experiment.
random selection
-
occurs when every member in an experiment has equal chance to be assigned to each level of the IV or to each treatment order.
random assignment
-
An experiment in which the IV is manipulated, but there is no random assignment.
Quasi Experiment
-
graph showing the number of times each score occurred in an experiment
frequency distribution
-
a distribution that is symmetrical and bell shaped.
Normal distribution
-
distribution that has extreme high scores and in which the mean is higher than the median
Positively skewed distribution, the mean is higher than the median.
-
a distribution that has extreme low scores and in which the mean is lower than the mean
negatively skewed distribution
-
formula for sum of squares
E(x-Xbar)^2
-
performance with the DV changes due to experience
practice effects
-
change in behavior due to becoming aware of the manipulation
sensitization effects
-
happens when the effects of one treatment still persist when another in introduced
carry-over effects
-
method of assigning subjects to orders of IV so across subject practice effects are balanced
counterbalancing
-
function of a Latin square
counterbalancing practice effects
-
extent to which the experiment provides a valid test of the relationship between the IV and DV
Internal validity
-
probability of making a type I error
alpha
-
probability of making a type II error
beta
-
Probability that a given experiment will find an effect of the IV on the DV if an effect exist.
statistical power
-
Can't prove that the IV has no effect on the DV.
Cant prove the null hypothesis
-
When extreme values of a variable fall closer to the mean when re-tested
Regression to the mean
-
how is the logic of an experiment ruined by confounds
It is impossible to know whether or not the effect on the DV was due to the IV or due to confound.
-
Extent to which results apply outside of the research.
External validity
-
aspects of a study that indicate to the subject how they are suppose to respond
Demand characteristics
-
Subject changes behavior due to unintentional cues from the experimenter.
Rosenthal effect/ experimenter expectancy effect
-
Subject changes behavior due to the expectancy that a change should occur.
Placebo Effect
-
what do single blind and double blind procedures help overcome.
Single- placebo, double- Rosenthal and placebo
-
Subject changes behavior due to the knowledge that they are being observed.
Hawthorne effect
-
The effect on the DV is due to the novelty of the IV and not the inherent quality of the IV itself.
Novelty effect
-
A type of naturalistic observation in which the researcher become part of the group being observed.
Participant Observation
-
Measures the variability in the DV due to error and the IV in an ANOVA.
MS treat
-
Measures the variability in the DV due to error in an ANOVA.
MSwithin
-
Measures the whether or not the IV has an effect on the DV in an ANOVA
F
-
A within subject experiment in which performance is measured both before and after IV manipulation.
time series design
-
Time series design in which performance is measured only once before and once after.
Single Group, Pretest -Posttest design
-
A time series design in which several measurements of the DV are made both before and after IV manipulation.
interrupted time series design
-
A design in which a pretest and a post test are given to two non-randomly assigned groups of subjects. The difference between the pretest and post test score serves as the DV.
non-equivalent before and after design
-
Design in which the researcher uses archival data to study an event that occurred in the past.
Ex post Facto design
-
A design in which the behavior is reported before treatment. Then the treatment is stopped to see if the behavior reverts.
Reversal design
-
Single subject research in which the behavior is recorded before treatment, treatment is administered and behavior measured, another treatment is administered and behavior measured again. Repeated multiple times. One behavior multiple treatments.
multielement design
-
Single subject research in which several behaviors are monitored. Treatment to one is administered, does behavior change relative to others? Than repeated for other behaviors. Multiple behaviors, one treatment.
Multiple baseline design
-
Statistic used to indicate the strength and direction of linear relationship between two variables.
Correlation Coefficient
-
Possible values of a correlation coefficient
+1.0 to -1.0
-
When one variable tends to increase so does the other, or vice verse
positive relationship
-
When one variable tends to increase the other tends to decrease, and vice versa.
Negative relationship
-
Formula for proportion of variance accounted for
r^2
-
proportion of variability that can be predicted from knowing the value of another.
proportion of variance accounted for
-
Floor or ceiling effects on one variable in a correlation.
range restriction
-
Sample the first people you see without regard to characteristics.
convenience sampling
-
Sample the first people you see based off of characteristics or pre-specified conditions.
Quota sampling
-
Randomly select a % of people from a list of population.
simple random sampling
-
sampling where you divide population into sublist, but have random sample mechanism
stratified random sampling
-
sampling where you have stages to choose populations, 2 or more random samplings
- multistage sampling
- Ex- first by city, than by name
-
best use of survey according to prof.
To compare changes and responses over time and across different demographics.
-
What does the margin of error in survey research mean?
That there is a 95% chance that if conducted on total population, in the same way as the sample, 95% of results would be within the sample.
-
Always Pg. 1, author's names, authors affiliation
title page
-
summary of the article, on pg 2, includes problem being investigated, subjects,research method, findings, conclusion
abstract
-
pg 3, introduce problem, review previous research/literature, state purpose,
introduction
-
recipe for conducting the research, divided into 3 sections: participants, apparatus or equipment, procedure, how to replicate
the method
-
the data collected, graphs/tables, means and variability, results of stat test,
results
-
relates data you collected to the research question. How data bear upon question posed in the intro, the theoretical implications, any shortcomings, and future research
discussion
-
the status of an article that has been accepted for publication, but which has not yet appeared in the journal.
in press
-
a typeset copy of the article that is sent to the author for final approval
galleyproofs
-
How does the Nuremberg code of ethics differ from the APA code?
The APA code differs in that it allows for certain forms of deceit and coercion to be applied to potential research subjects that the Nuremberg does not.
-
Factors determining statistical power?
- Correlation between the IV levels
- Alpha Level
- Variability in the DV
- Effect Size
- Sample Size
-
Why do an ANOVA instead of multiple t-test?
- Multiple t-test take more work
- multiple t-test inflate the type I error rate
-
Why correlation can not be used to determine causation?
- Can not tell which is the cause and which is the effect (directionality problem)
- There could be a third variable having an effect( third variable problem)
-
What are the mathematical limitation of correlations?
- poor at capturing non-linear relationships
- they are greatly affected by extreme scores
- can be lowered by range restriction
-
disadvantages on non parametric statistics?
- tend to be overly conservative
- no non parametric tests exist for some research designs
-
Mathematical relationship between t and F distributions
at any alpha level: (t(x)^2 or t^2(x) = f(1,x) for any X
-
require assumption about the distribution of the DV, that it is normal, in order to obtain correct p-values.
parametric statistics
-
Class of tests that do not assume the DV is normally distributed.
Non parametric Statistical tests
-
Advantages of non parametric statistics
- less restrictive assumptions
- often easier to calculate than parametric tests
-
Factors to consider when evaluating polls?
- question wording
- question order
- lack of full disclosure
- bias in choosing the sample
- ??
-
problems with all single subject experimental designs
- experimenter bias
- placebo effects
- generalization
- likely fact I and type II errors
-
Ethical principles when conducting psychological research?
- subjects should agree to participate
- no coercion
- confidential
- gains to science outweigh any harm
- all reports accurate
-
Experiments need to have these 2 things?
- random assignment
- manipulation of the IV
-
naturalistic observation
research technique in which the researcher simply observes and describes behavior.
-
correlational aproach
research technique in which the researcher determines whether variables are related without manipulating the variables
-
Independent variable
variable for which the researcher chooses values
-
dependent variable
variable the researcher measures to determine the effects of the IV
-
Between subjects design
research design in which each subject is assigned to one level of the IV
-
within subject design
research design in which each subject is assigned to all levels of the IV
-
Type I error
Finding an effect of the IV on the DV when in reality no such effect exist.
-
Type II error
Failing to find an effect of the IV on the DV when in reality there is an effect
-
confound
a nuisance variable that varies reliably with the IV
-
factorial design
A design in which there is more than one independent variable and each level of every independent variable is present at every level of the other independent variables.
-
interaction
occurs when the effect of one independent variable on the dependent variable in a factorial design changes depending on the levels of another independent variable.
-
nuisance variables
any variables other than the IV that have an effect on the DV
-
floor effects
occurs when the values of the DV are so low that they are unlikely to be effected by the IV
-
ceiling effects
occurs when the values of the DV are so high that they are unlikely to be effected by the IV
-
Main effect
the independent effect of one IV on the DV in a factorial design.
-
Response acquiescence effect
the tendency for subjects to respond yes to question that they have thought little about
-
random factor
an IV whose levels were chosen randomly from a population of possible values
-
fixed factor
an IV whose levels were chosen non-randomly
-
Research in which the researcher is directly asking questions
survey research
-
Can you prove a scientific theory false? true?
Yes, no
|
|
