Research design & statistics Flashcards
an experiment-first-and-explain-later approach to research
inductive method
in this method, the researcher first observes the phenomenon, conducts a series of experiments on it, and then proposes a theory based on the results of those experiments
inductive method
an explain-first-and-verify-later approach to research
deductive method
In this method, based on initial observation, the investigator explains an event by proposing a theory and then attempts to verify the theory by conducting experiments
deductive method
hypothesis testing is what kind of method of research?
deductive method
what is the purpose behind inductive or deductive reasoning?
to build a theory
stating that X causes Y is an example of a
theory
a proposed answer to a research question, but verifiable through additional research
a hypothesis
a hypothesis is a more specific prediction stemming from a _______-
theory
“When people who stutter are put in highly stressful speaking situations, their amount of stuttering will increase” is an example of a theory or hypothesis?
hypothesis
What are the 2 kinds of hypotheses?
null and alternative
What does the null hypothesis state?
that 2 variables are not relates
Two variables are related; perhaps one is the cause of another. What kind of hypothesis is this?
alternative hypothesis
“stressful situations an stuttering are not related; when people who stutter are placed in a highly stressful speaking situation, stuttering does not increase”. This is an example of…
null hypothesis
What would be the alternative hypothesis to “stressful speaking situations and stuttering are not relates”?
there is a cause-effect relationship b/w stressful speaking situations and stuttering
What is the goal of research?
To reject the null hypothesis and accept the alternative hypothesis
systematic observation leads to measured values of those events. What are the measured values referred to as?
Data
What is the difference between quantitative and quantitative data?
qualitative data are stated in words; quantitative data are given in numbers
“The client has a severe articulation disorder characterized by multiple omissions of phonemes” is an example of what kind of data?
qualitative data
“In a 5-minute spontaneous speech sample, the client omitted word-final phonemes in 75% of the contexts” is an example of what kind of data?
quantitative data
What are the 2 kinds of criteria that data should meet?
validity and reliability
The degree to which an instrument measures what it purports to measure
Validity
a child language test is valid if it measures language skills (T/F)
true
What are the 4 types of validity?
Predictive/criterion validity, concurrent validity, construct validity, content validity
The accuracy with which a measure predicts future performance on a related task
predictive/criterion validity
the degree to which a new measure correlates w/ an established measure of known validity
concurrent validity
The performance on the GRE should predict the GPA in graduate programs. This is an example of
Predictive/criterion validity
A new test of language skills is correlated with a well-established test of known validity. What does this new test demonstrate?
concurrent validity
the degree to which measures are consistent with theoretical constructs or concepts
Construct validity
A test of language development meets the theoretical expectation that as children grow older, their language skills improve. This test has what kind of validity?
Construct validity
an instrument samples the full range of skills that it purports to measure. This is what kind of validity
content validity
the consistency with which the same event is measured repeatedly
reliability
test scores are reliable if…
they are consistent across repeated testing or measurement
If reliable, a standardized test given to the same individual on 2 occasions should result in different scores (T/F)
False; the 2 occasions should result in similar scores, if the test is reliable
A clinician who measures stuttering in a client on 2 consecutive days records similar rates of dysfluencies. What is this an example of?
Reliability
Most measures of reliability are expressed in terms of a _____ ______
correlational coefficient
The correlational coefficient is a number or index that indicates the relationship b/w …
2 or more independent measures
The correlational coefficient is usually expressed through ____ ____ ____ ___
Pearson Product Moment r
An r value of 0.00 indicate that there is no relationship b/w two measures (T/F)
True; when r is 0.00, there is no relationship b/w 2 measures
What is the highest possible value of r?
1.00
What is the lowest possible value of r?
-1.00
The closer the r is 1.00, the less the reliability of the test or measurement (T/F)
False; the closer that r gets to 1.00, the greater t/ reliability of the test/measurement
Refers to consistency of measures when the same test is administered to the same people twice
Test-retest reliability
Test-retest reliability shows that the two sets of test scores are positively correlated (T/F)
True
Another name for alternative-form reliability is
Parallel form reliability
This type of reliability is based on the consistency of measures when 2 parallel forms of the same test are administered to the same people
Alternative-form reliability/Parallel form reliability
Split-half reliability is a measure of ______ consistency of a test
internal
When response to items on the first half of the test are correlated with responses to items on the second half of the test, the test is said to have
split-half reliability
split-half reliability generally overestimates reliability b/c it does not measure stability of scores over time (T/F)
true
the extent to which 2 or more observers agree in measuring an event
interobserver or inter judge reliability
if 3 judges independently rate the fluency of a speaker, there is high inter judge reliability if there is good agreement b/w judges (T/F)
True
Optimally, good agreement results in an inter judge reliability coefficient of ____ or more
0.90
refers to the extent to which the same observer repeatedly measure the same event consistently
intra observer or intra judge reliability
if the same clinician rata a child’s intelligibility over several sessions, those ratings should be consistent to assure acceptable level of _______reliability
intra observer or intra judge reliability