Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
92 Cards in this Set
- Front
- Back
statistics
|
an objective means of interpreting a set of data
|
|
3 things statistics allow us to do
|
describe characteristics (mean, median, mode), test relationships (correlation, regression), test differences (t-tests, ANOVA)
|
|
the value of a population; the value of a sample
|
parameter, statistic
|
|
a subset of the population
|
sample
|
|
4 ways to select a sample
|
random, stratified random, systematic, volunteers
|
|
example of random sample
|
fishbowl method, random number generation
|
|
example of stratified random
|
determine relative characteristics (choose 25 random fr, 25 random so, 25 random jr, and 25 random sr for a total of 100)
|
|
example of systematic
|
choose every 10th person (sampling interval of 10)
|
|
what kind of research is random assignment essential for
|
experimental
|
|
what is the unit of analysis
|
the most basic unit from which data can be produced
|
|
give 3 examples of central tendency scores and 3 examples of variability scores
|
mean, mode, median; standard deviation, variance, range
|
|
how do you measure the median
|
if odd number of scores, it's the middle number; if even number of scores, take the average of the middle two
|
|
when would the median be used over the mean
|
skewed data/outlier(s)
|
|
the degree of difference of each individual score in the group from the central tendency score
|
variability score
|
|
one number that best represents a group of scores
|
central tendency score
|
|
what are the 3 assumptions about the distribution of parametric tests that must be met?
|
population is normally distributed on the variable of interest; homogeneity of variance in the samples selected, the observations are independent
|
|
for positive and negative skewness, which way do the tails head?
|
for positive skewness, the tail is heading positive
|
|
does increasing the alpha level from 0.05 to 0.30 increase or decrease the chance of making a type I error?
|
increases the chance of making a type I error and decreases the chance of detecing a real difference (type II error); alpha is the chance of making a type I error
|
|
what are the two types of kurtosis
|
more peaked, more flat
|
|
what are the two things statistical techniques can determine
|
the reliability (statistical significance), and how meaningful (strength, magnitude, size)
|
|
what does the null hypothesis state
|
there is no difference/significance between the variables
|
|
most statistical analysis are based on what
|
probability; a measure of how likely it is that your results occurred by chance
|
|
what is the difference between a type I and type II error
|
type I: rejecting a true null hypothesis
type II: accepting a false null hypothesis |
|
what happens when you change alpha from .05 to 0.01
|
you decrease the chance of making a type I error and increase the difficulty of detecting a real difference
|
|
what are the values for a small effects size (meaninfulness), a moderate, and a large ES?
|
0.2, 0.5, 0.8
|
|
how is effect size calculated?
|
(M1-M2)/SD
Mean and standard deviation |
|
what happens to the effect size as the standard deviation increases
|
effect size decreases
|
|
what is power
|
the capability of rejecting a false null hypothesis/detecting a real difference
|
|
what is a commonly accepted level of power in behavioral research
|
0.8
|
|
what is the range for power
|
0-1
|
|
as the number of subjects increase, what happens to the difference needed between group for significance
|
the difference between groups decreases as number of subjects increase; people will cheat and add more subjects to get significant results
|
|
what is a statistical technique used to determine the relationship between two or more variables
|
correlation
|
|
what is the coefficient of correlation
|
a quantitative value of the relationship between two or more values; can range from 0-1 in either the positive or negative direction
|
|
what is the line of regression
|
best fit line or line of best fit
|
|
what can you do to alpha to avoid making a type I error
|
decrease the value of alpha
|
|
what are other names for "r", the coefficient of correlation?
|
Pearson product moment coefficient of correlation, interclass correlation, simple correlation, Pearson r
|
|
The Pearson r measures the degree of _______ relationships between two variables.
|
linear
|
|
what three things about a correlation is the researcher interested in?
|
positive or negative
statistical significance how strong/meaningful is it |
|
how do you determine the appropriate degrees of freedom for the table of critical values for r
|
N-2
|
|
what is r^2
|
coefficient of determination; it is a measure of the variance shared by 2 variables being correlated, indicating the portion of the total variance that can be accounted for by the variance in the other measure
|
|
how do you calculate the % variance accounted for?
|
r^2 x 100%
|
|
how do you calculate the unaccounted variance
|
1-r^2
|
|
what is an alpha level
|
the probability of making a type I error
|
|
what is a t-test
|
a computational way of relating differences between treatment means to the amount of variability we would expect to see between any 2 sets of data drawn from the same population
|
|
what are the two types of t-tests and how do you set them up
|
independent t-test: two different groups of subjects
dependent t-test: one group of subjects tested twice |
|
what determines the difference between a one-tailed and a two-tailed t-test
|
where you place your alpha-level
|
|
in a one tailed test, which error are you more likely to make and why
|
type I error (rejecting a true Ho) because it is easier to get into the region of rejection
|
|
is it easier to get into the region of rejection for a one-tailed or a two-tailed t-test
|
one-tailed
|
|
what is the greatest amount of groups you can have for a t-test
|
2
|
|
what is calculated in a dependent t-test that is not calculated in an independent t-test and vice versa?
|
dep. has a % improvement and indep. has an effect size
|
|
what are the three types of ANOVA
|
one-way ANOVA (simple)
factorial ANOVA repeated measures ANOVA |
|
what is used to compare more than 2 mean scores
|
ANOVA
|
|
What is the F ratio
|
the ratio of true variance (between groups) divided by error variance (within groups); used in ANOVA
|
|
what is the true variance and error variance due to?
|
true variance is due to the independent variable while error variance is due to chance and sampling error
|
|
if the independent variable had an effect, then F ratio will be _____
|
large; >1
|
|
what are Scheffe, Tukey WSD, Newnan-Keuls, Duncan, and Fisher LSD
|
types of post hoc tests used for one way ANOVA
|
|
how are the independent variables set up in a one way analysis
|
there is only 1 independent variable, typically with 3 or more levels (ex. 20%, 40% and 60% VO2 max)
|
|
how are the independent variables set up for a factorial ANOVA
|
there are two or more independent variables (2-way, 3-way, etc)
|
|
what are factors, levels, main effects, and interactions for factorial ANOVA
|
factors are independent variables, levels are the specific value of a factor, main effect is the effect of a factor by itself when other factors are held constant, and interaction is the combined effect of several factors
|
|
what is a repeated-measures ANOVA
|
the same subjects are test on or exposed to all levels of an independent variable
|
|
what is the simplest case of a repeated measures ANOVA
|
dependent t-test
|
|
what are 3 advantages of a repeated measures ANOVA
|
provides some control over individual differences among particpiants, fewer participants needed, can study something across time
|
|
what are 4 problems with repeated measures ANOVA
|
carryover/order effects, practice effects, fatigue or boredom, sensitization
|
|
on an ANOVA graph, what do parallel lines mean, and how are main effects calculated
|
no interaction; average of the individual line itself for main effects
|
|
what is the study of changes in behavior across years
|
developmental research
|
|
what is the focus of developmental research?
|
cross-age comparison
|
|
what are the 2 basic developmental research designs
|
longitudinal and cross-sectional studies
|
|
what are the 4 limitations of longitudinal studies
|
time consuming, some move away, some die, familiarity with test items
|
|
what is the cohort problem
|
problem with cross-sectional developmental design where not all age groups are from the same population/cohort
|
|
what are 4 methodological problems of developmental research?
|
unrepresenative score (outliers), unclear semantics, lack of reliability (in young children), statistical problem (violate ANOVA assumption of equal variances)
|
|
in measurement, what does validity refer to
|
whether a test measures what it is intended to measure
|
|
what are the two important classifications of validity in experimental research
|
internal and external validity
|
|
what are two questions that internal validity addresses
|
Did the independent variable make a difference? Are there alternative hypotheses that can account for the results?
|
|
what are two questions external validity addresses
|
how generalizable is the effect? to what populations, settings, or treatment variables can this effect be generalized?
|
|
Give two examples of threats to internal validity, describing and giving examples.
|
maturation: associated with aging, as children get older, they naturally get larger, stronger, faster, and more coordinated
testing: the effect that taking a test once has on taking it again; a group of students hitting 20 forehand shots in tennis may do better 3 days later without treatment, they learned something from performing the test the first time |
|
give 2 threats to external validity, explaining and giving examples
|
reactive or interactive effects of testing: this is a problem that occurs because a pretest may make the particpiant more aware of sensitive of the upcoming treatment; in a fitness assessment given followed by a training program, an individual might try harder in the training program because they know their fitness is low
interaction of selection bias and the experimental treatment: a drug education program may be effective towards college freshman but not 3rd year medical students |
|
what is the rosenthal effect
|
the experimenter anticipates that certain participants in the group will perform better (halo effect)
|
|
what is the hawthorne effect
|
participants performance changes when attention is paid to them
|
|
what is the Avis effect
|
a threat to internal validity where participants in the control group try harder just because they are in the control group
|
|
what is a placebo and what does it do
|
a false treatment that controls for psychological effects
|
|
what is a blind setup
|
participants dont know whether they are receiving the experimental or control treatment
|
|
what is a double blind setup
|
neither participant nor the experimenter knows which treatment is given to participants
|
|
what is "nested"
|
between-subjects: subjects appear only in one condition or group
|
|
what is the opposite of "nested" or "between-subjects"
|
within-subjects/crossed/repeated measures: each subject receives all levels of the independent variable, the subjects act as their own controls
|
|
what is a combination of between-subjects and within-subjects called
|
mixed designs
|
|
what are variables that cannot be activitly manipulated called
|
organismic variables
|
|
when writing statistical notation, does within- or between-subjects come first and what does this mean
|
between subjects comes first (where subjects appear in only one condition like gender or age)
|
|
what are 3 purposes for presenting research at scientific conferences
|
visibility, feedback, and networking
|
|
what are three formats for research presentations
|
poster, oral, symposia
|
|
what are the 5 steps for publication of a research article
|
submit manuscript to the editor electronically, editor selects reviews/referees (either blind or unblind), reviewers submit critique/recommendations to editor, editor makes a decision, possible revisions and fruther review
|
|
what are the 4 common voting categories for the review process
|
accept with optional editing, accept contingent on making required changes, unacceptable in present form, reject unconditionally
|
|
what are the 4 criticisms of the peer review process
|
inconsistent critiques by reviewers (low inter-rater reliability), too subjective, reviewer biases, long delays
|