Home
Subjects
Create
Search
Log in
Sign up
Upgrade to remove ads
Only $2.99/month
Home
Social Science
Psychology
Experimental Psychology
Stats Exam 2
STUDY
Flashcards
Learn
Write
Spell
Test
PLAY
Match
Gravity
Terms in this set (52)
Correlated groups t-test
To determine if the mean difference between two groups is significant.
Correlated groups t-test, when to use it
Within participants design, matched-participant designs
the assumptions for correlated groups t
random selection of participants, observations within each treatment are independent, normal population distribution, homogeneity of variance in the two populations of scores, interval or ratio data
Cohen's d (effect size)
criteria for interpreting size
.2 = small effect
.6 = medium effect
.8 = large effect
animal adv
greater control, past (genetics and experience), present (tolerance for highly controlled experiment and absence of cognitive factors), convenience, necessity
animal dis
possible lack of relevance, can't research higher cognitive processes, possibly unpleasant and or dangerous
human adv
probably relevance, research on higher cognitive processes is possible, usually pleasant, fewer ethical dilemmas about informed consent
human dis
intolerant of highly controlled studies, cognitive factors may be main cause of results, inconvenient, dangerous research not possible
2 major considerations for selection
control and relevance
ethical considerations
The Belmont Report 1979, part A is boundaries between research and practice and the part B is basic ethical principles. Basic ethical principles are respect for persons, beneficence, and justice.
Respect for persons
autonomous agents (capable of deliberation), vulnerable populations (incapable of self-determination), capability should be periodically re-evaluated
Beneficence
do no harm, maximize benefits/minimize harm, ethical dilemmas in research with children
Justice
who should bear the burden of research, who should receive the benefits, try to make sample reflect the census, offer mild money to not coerce poor people
2-Way Independent Groups ANOVA
2 way: factorial designs: at least 2 factors, every level of each factor is represented under every level of the other. Two independent one dependent factor.
ANOVA advantages
Allows us to evaluate the performance of three of more groups within now statistical test, allows us to minimize Type 1 error despite multiple comparisons, allows us to understand complex functions.
common thread between t, z, and F
All are between group variance / within group variance
terminology
variable = factor = main effect
level = treatment group
what HSD is
Honestly significant difference, only need if significant.
advantages of factorial ANOVA
1+1=3, main effect of factor one, main effect of factor two, and factor 1 x factor 2 (in two way interaction)
definition of an interaction
The effect of one IV, on the DV is dependent on the level of another IV2. Graphical it is the lines not being parallel.
Scales of Measurement
label, category, ordinal, interval, ratio
Label
distinct, strength: identification, weakness: no math, no relationship between numbers, ex. subject numbers
Category
indicate similarity, strengths: grouping (members in group are similar, but groups are different), compute frequency, Chi squared (expected freq vs observed), lowest level of scientific data, weakness: no relationship between numbers, no order, ex. group numbers in experiments
Ordinal
fixed order, strength: concept of greater than, rank/order, nonparametric statistic (can use rank order statistic), weakness: no fixed distance between orders/points, no parametric statistic, ex. order of finish in a race
Interval
arbitrary (relative) zero, fixed equal distance between points, strengths: all math and statistics on intervals, relative comparisons can be made, most powerful statistic (parametric t, z, and f tests), weakness: no absolute zero, no absolute values, ex. races (1st place is set at zero all others compared against that)
Ratio
absolute zero, strengths: absolute quantities, all math and statistics can be used, weakness: none, ex. running time in a race (absolute clock time)
Validity
has to do with the accuracy of measurement (or the purpose of the tests)
Reliability
has to do with consistency of measurement (or consistency with which the purpose is met). 0 - 0.29 weak, 0.3 -0.69 moderate, 0.7 - 1.0 strong
Types of validity
face, construct, content, criterion
face
does it look like it related to the construct it is supposed to measure (aggression)
construct
does it measure the human characteristics it is supposed to measure (motivation)
content
does the test relate to content domain (geography)
criterion
the degree to which the test (predictor) correlated with performance on relevant criterion measure in the real world (success) concurrent vs. predictive
Types of reliability
test retest, alternate forms, split half, inter rater
test retest
take more than once
alternate forms
change exam up
split half
compare scores from each equal half for an exam
inter rater
different raters will get the same score
internal vs. external validity
internal concerned with getting the truth within the boundary conditions of your experiment
external concerned with generalizability of results; if a result is externally valid it is internally valid, if a result is internally valid it may not be externally valid
internal vs. external validity
Descriptive Methods
Naturalistic observation, experimentation, correlation
naturalistic observation
field research (behavior of a species is observed and recorded in natural settings), research does not influence subject or data. strengths: behavior is real, the natural behavior is observed, weaknesses: we can't determine cause
experimentation
laboratory research (behavior of a species is observed and recorded in highly controlled setting), experimenter is an active agent in the situation, manipulates independent variable and observes changes in dependent variable. strength: causes of behavior can be discovered, weaknesses: behavior may be artificial
correlation
relationship (to specify the direction and strength of the relationship between two variables, range from + 1 to -1, positive correlation (both variables go same direction), negative correlation (one variable goes one way the other goes the opposite). strengths: can use for prediction, weaknesses: explanations (ex. storks and babies)
probability sampling
each member of the population has an equal chance of being selected
random sampling
random phone number calling
stratified random sampling
identify strata (men or women) sample according to population ratio
cluster sampling
identify clusters (general studies class), sample the whole cluster
multi-stage sampling
identify clusters, sample selectively from the clusters
non-probability
all members of the
population do not have an equal chance of selection
convince sampling - neighbors
quota sampling - quota is to non prob as stratified is to prob
frequency claims
construct validity (how variable operationalize/measured), external validity (do the results generalize to other people, settings, times)
association claims
construct validity (how are the two variables measured), external validity (do the results generalize to other people, settings, times), statistical validity (how reliable are the results)
causal claims
construct validity (how are the two variables measured), external validity (do the results generalize to other people, settings, times), statistical validity (how reliable are the results), internal validity (must rule out alternative explanations for results)
YOU MIGHT ALSO LIKE...
Research Methods
37 terms
PSC 41 midterm 2
91 terms
AP Psych Unit 2 Terms
70 terms
Psychology. Chapter 1: The Science of Psychology
37 terms
OTHER SETS BY THIS CREATOR
EDP 352K Exam 1
22 terms
PSY 364P
68 terms
HDF 337 Exam 1
71 terms
PSY 345 Exam 4
60 terms
OTHER QUIZLET SETS
Micro Exam 3 (chapter 8)
27 terms
Spanish Vocab
26 terms
Psychology: Chapter: Learning & Conditioning
12 terms
The War of the Wall
13 terms