degree to which the measure will generate the same result in the same condition
consistency of an experiment or test
validity
degree to which we measure what we purport to measure
the factual accuracy of an observation
reliability vs. validity
validity is almost entirely dependent on reliability
both are necessary
scientific method
1) statement/hypothesis;
2) observations;
3) replicability;
4) model/law
order
processes are not random;
they happen in a recognized pattern or sequence
determinism
all events have a cause;
psychological phenomena have antecedents, or preceeding circumstances that cause an event
empiricism
practice of relying on observation and experimentation
parsimony
the simplest answer is usually the right one;
good if supported by data
naturalistic observation
observations made in natural or native settings
hypothesis
an educated guess
independendent variable
the manipulated variable
dependent variable
the variable whose reaction is being observed
experimental group
group that receives the experimental treatment (some manipulation by the experimenter)
control group
treated exactly like the experimental group except they do not receive the experimental treatment
primacy
tendency to remember things better when they come first in a series
recency
tendency to remember things better when they come last in a series
experimental bias
anything the researcher does that interferes w/ the experimental design
single-blind study
subjects are not aware of which group they're in
double-blind study
both judges and subjects are unaware of what group the subjects are in;
often used in drug research or anything involving subjective judging
within subject design
participants are both the control and the experimental groups (all subjects are in all conditions)
demand characteristics
environmental cues that lead to a subject/participant to respond a particular way (either subtle or not subtle) because they know what the researcher is looking for
correlational studies
studies of the relationship between two variables;
does not establish causality!
factorial design
experimental design looking at two or more independent variables
main effect
number of variables
interaction effect
when the effect of 1 variable is not the same at each level of the other variable.
(intersect=interaction; no intersect=no interaction)
quasi-experimental design
still have experimental and control groups, but no random assignment w/ an open system
small "n" research
less than 30 subjects;
results are less generalizable;
often used in drug trials
case study
a detailed, non-experimental analysis of a person or group
generalization
the extent to which a finding applies to persons other than those that were the subject of the study
controls
must account for other factors by holding for other conditions and subject variables constant
confounds
some other variable(s) that affects the results
paradigms
a model or pattern an investigator uses to organize research
between group design
two separate groups: experimental and control;
each subject is in one condition
matched subject design
when a subject variable is so critical in the experiment, subjects are matched on that variable
repeated measures design
all subjects are exposed to all conditions, all subjects are their own control
placebo design
two control groups are used, w/ one receiving a "placebo" version of the treatment
randomization
participants are assigned by chance to 1 of 2+ conditions
ad-lib matching
matching a sample as best you can
random block design
using matching techniques to try to equate groups on important characteristics
latin square design
counterbalancing procedure;
each condition occurs equally often
(1,2,3,4; 2,3,4,1)
practice effect
improvement over multiple trials
fatigue effect
when subjects become bored or tired and their performance decreases
attrition
loss of participants over a long period of time
problem statement
a precise statement of what knowledge is sought and why it was sought
method
the plan of the research;
how the knowledge was gained
results
a precise statement of the knowledge that was gained
variable
a characteristic that varies between individuals
data
observation made on a variable
measurement
a scheme for the assignment of numbers or symbols to specify different characteristics of a variable
sample
collection of subjects we select as representative of the population
population
the group of subjects to whom the research applies
descriptive statistics
summary descriptors or characteristics of the population/sample (mean, median, & mode)
descriptive research
a research plan undertaken to define the characteristics and/or relationship among variables based on systematic observation of these variables
experimental research
a research plan undertaken to test relationships among variables based on systemic observation of variables that are manipulated by the researcher
"n"
number of participants in a study
"s"
subject
scale
a specific scheme for assigning numbers or symbols to designate characteristics of a variable
nominal scale
qualitative scale;
labels used to differentiate observations
ordinal scale
scale that implies an order (ascending or descending);
ranked, not w/ set intervals
interval scale
scale w/ equality between units;
has an arbitrarily assigned zero points;
can only do +, & -
ratio scale
scale w/ equality between units;
has an absolute zero points;
can do +, -, *, & /
distribution
collection of measurements viewed in terms of the frequency of observations
frequency
way to organize data;
can simplify w/ a graph
bell curve
the shape of a "normal" distribution;
mean of 100, standard distribution of 15;
mean=median=mode
measures of central tendency
indexes that refer to how scores tend to cluster in a particular distribution
mean
the sum of the scores in a distribution divided by the number of scores
(aka the average)
median
the midpoint or midscore in a distribution
mode
the most frequent score in a distribution
indices of dispersion/measures of variability
indexes that describe the dispersion or scatter across the measurement scale;
range, variance, & standard deviation
range
the highest score in a distribution minus the lowest score
variance
the mean of the squared deviation scores about the mean of a distribution
standard deviation
the square root of the mean of the squared deviation scores about the mean of a distribution
(or, the square root of the variance)
statistic
characteristic of a sample
parameter
characteristic of a population
normal distribution
definition of a particular functional relation between deviations about the mean of a distribution and the probability of these different deviations occuring
standard error of the mean
the standard deviation of a distribution of sample means
population mean
inferred by the mean of the sample
null hypothesis
a statement/assumption of no difference
research hypothesis
a statement/assumption of a statistically significant difference
significance
the level of calculated probability was sufficiently low as to serve as grounds for rejection of the null hypothesis
1-tailed test
a directional hypothesis test that incorporates a rejection region in only one tail of the probability curve used for a given statistic
2-tailed test
a nondirectional hypothesis test that incorporates rejection regions in both tails of the probability curve used for a given statistic
t-test
used to calculate the probability of whether a particular difference between sample means would be expected under the terms of the null hypothesis
ANOVA
used when there are 3+ values to compare;
shows if there is a difference somewhere in the values
F-ratio
variance between groups / variance within groups
multivariate analysis
statistical model for testing the influence on multiple dependent variables in a single research design
multifactor analysis
statistical model for testing the consequences of manipulating 2+ independent variables in a single research design
interaction effect
whatever nonerror variation is observed among the individual group when the main-effects variation has been removed
non-parametric statistics
used in small "n" research;
when descriptive alone is ok
Chi-squared
compares an observable set of frequency with an expected set of frequency;