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;
66 Cards in this Set
- Front
- Back
nominal
|
lowest level of measurement .
may be thought of as "naming" level respond with words, not numbers |
|
ordinal
|
measurement place participants in order from high to low
|
|
interval
|
equal interval without an absolute zero
|
|
ratio
|
equal interval with an absolute zero
|
|
reliable
|
a test is said to be _________ if it yields consistent results
|
|
validity
|
accurately measures what you're testing
|
|
1. A test with high realiability may have low validity
2. Validity is more important than reliability 3. To be useful, an instrument must be both reasonably valid and reliable |
3 principles of realibility and validity
|
|
interobserver reliability
|
when two or more researchers agree on the realiability of the measurments taken
|
|
correlation coefficient
|
check the degree of relationship between 2 quanatative results.
1.00=perfect reliability |
|
test-retest reliability
|
obtain measurments at two different points in time
|
|
parallel forms of reliabilty
|
some published tests come in two parallel forms that are designed to be interchangable with each other; they have different items that cover the same content
|
|
split-half reliability
|
checks on the consistency of scores within the test itself
|
|
test-retest reliability
parallel forms of reliability |
which measure the consistency of scores over time?
|
|
split-half
alpha |
which measures the consistency of items within a test at a single point in time?
|
|
instrument
|
any type of measurment device
|
|
content validity
|
researchers make judgements on the appropriateness of its contents
*essential for achievement tests |
|
face validity
|
judgement is based on whether the instrument APPEARS to measure what it is supposed to measure
|
|
content and face validity
|
what two types of validity rely on judgements?
|
|
predictive validity
|
To what extent does the test predict the outcome it is supposed to predict
|
|
construct validity
|
the type of validity that relies on subjective judgements and empirical data
|
|
1. origante question
2. develope rationale 3. fesability of answering question |
3 steps to formulate a question
|
|
efficacy research
|
the process of demonstrating under ideal conditions
|
|
outcomes measurement
|
the process of demonstrating under average or less than average conditions
|
|
comparative or standard group
|
*type of descriptive research
strategy to measure behvior of two or more types of participants at one point in time in order to draw conclusions about similarities and differences between them |
|
cohort studies
|
pateints who presently have a condition and/or receive treatment are followed over a period of time and compared with another group who are not affected by the condition under investigation
|
|
developmental or normative
|
designed to measure changes over time in behavior or characteristic, usually with reference to aging and maturation
|
|
cross-sectional
|
select participants from various age groups and observe differences
*norms should be established on random selected participants who are a representative of population *sample size must be appropriate to to population |
|
semilongitudinal
|
compromise of longitudinal, divide total age span into overlapping age spans
|
|
correlational
|
-asks two basic questions
1. how closely related are the variables 2. how well can performance on one variable predict performance on another |
|
survey
|
use questionaires, interviews, or combinatons
|
|
retrospective or post facto
|
done after the fact, IV occured in the past and investigator starts with effect
|
|
case-control study
|
a type of design often used in epidemiology
|
|
case history
|
examine one person in depth. looking at an unusual phenomen, hard to generalize
|
|
frequency
|
measures how many instances there were and what was the time frame
ex) he said ball 8 out of 10times *good for nominal and ordinal data *useful and objective |
|
duration
|
count the duration of a type of behavior
ex) client was dysfluent for 15 minutes during 30 minutes of speaking time *good for contineous behaviors |
|
interresponse time
|
time lapse between any two discreet events or responses
ex)want kid to sit in chair -count time elapsed between last time left seat to when he sits down to when he gets back up agaiin |
|
latency or reaction time
|
how long it takes for a client to respond to a stimulus
|
|
response amplitude
|
intensity of response
ex)voice intensity, sweaty palms, |
|
utility
|
is the date useful for your purpose?
will you be able to answer your research question with the data you're collecting |
|
sensitivity of test
|
rarely fails to identify disorder or disease
mathmatically defined as a proportion of people who likewise test positive for disorder or disease *gather all people with that trait (false-positives) |
|
speciicity of test
|
seldom identifys a person having a disease or a disorder or when they don't
mathmatically defined as ratio of people who test negative on a screening test to people without disease or disorder *might miss more (false positives) |
|
floor effect
|
test so hard, no one can pass it
|
|
ceiling effect
|
test so easy, everyone can pass it
|
|
precision
|
the reliability aspect of a test, will it give you the same scores with each administration of test
|
|
accuracy
|
both reliablity and validity; tests what you want it to test and scores are reliable
|
|
scaling power
|
refers to scale of measurement
|
|
test-retest
|
methods to estimate reliability: stability
|
|
parallel forms
|
methods to estimate reliability: equivalence
|
|
split-half
alpha kuder-richardson #20 formula |
methods to estimate reliablity: internal consistency
|
|
standard error of measurement
small SEM related to high levels of reliability |
SEM
|
|
kappa
|
statistic that takes into account fact that some agreements may occur by chance, especially in very frequently or less frequent occuring behaviors
|
|
direct replication
|
same investigator repeats same research with same or similar participants, in same physical setting, to confirm realiability of original results
|
|
systematic replication
|
try to extend generalization across participants, settings, measurments, or treatments
manipulates 1 or 2 variables |
|
threats to external validity
|
the ability for research to be generalized
|
|
selection bias
|
generalizing the results to a population when an experiment is conducted on a nonrandom sample
|
|
reactive effects of experimental arrangements
|
if the experimental setting is different from the natural setting in which the population usually operates, the effects there were observed in the experimental setting may not be generalized to the natural setting
|
|
reactive effect of testing
pretest sensitization |
a pretest might affect how a participant will respond to experimental treatment
|
|
multi-treatment interference
|
when a group of participants receive more than one treatment which could ultimately affect their responses to later treatment
|
|
threats to internal validity
|
explains how aspects other than the treatment actually working could interfer with the validity of an experiment
|
|
history
|
other enviormental influences on the participants between the pretest and the posttest
|
|
maturation
|
participant got older, wiser, or smarter between pre and post test
|
|
instrumentation
|
possible changes in the instrument
|
|
testing
|
effects of the pretest on the performance exhibited on the post test
|
|
statistical regression
|
occurs when participants are selected on the basis of their extreme scores
|
|
intact groups
|
researcher uses previously existing groups, so they are not random
|
|
selection
|
two different groups are not initially the same in all important aspects
|