Home > Flashcards > Print Preview
The flashcards below were created by user
MeganM
on FreezingBlue Flashcards. What would you like to do?

Used to reduce the risk of Thype II errors and strengthen statistical conclusion validity by estimating in advance how big a sample is needed.
Power analysis

Other things being equal, the more stringent this criterion, the lower the power.
The significance criterion of PA (alpha)

As _____ increases, power increases.
sample size

An estimate of how wrong the null hypothesis is; how strong the relationship b/w the independent and dependent variable is in the population.
The effect size in a PA

The probability of refecting a false null hypothesis.
Power (or 1beta)

The magnitude of the relationship b/w research variables.
The effect size (Power analysis)

Cohen's d small effect.
0.20.3

Cohen's d medium effect.
~0.5

The easiest way to boost statistical power.
increasing sample size

What is a Type I error?
a false positive

What is a Type II error?
a false negative

The probability of detecting a true relationship or group difference.
 the power of a statistical test;
 the complement of beta (1beta)

Power analysis is used to reduce the risk of what?
Type II errors

Researchers usually establisht the risk of a Type I error (alpha) as ____.
0.05

What is the conventional standard for 1beta?
.80

With power = .80, there is a 20% risk of what?
committing a Type II error

When relationships are modest, _____ sample sizes are needed to avoid Type II errors
large

Effect size must be estimated using ________ evidence.
available (it is NOT known)

Where is the evidence taken from to determine effect size?
 pilot studies
 calculated based on based on findings from earlier studies on a similar problem

The entire aggregation of cases in which a researcher is interested.
population

The type of person that is the focus of the research.
population

The criteria that specify population characteristics.
eligibility or inclusion criteria

Quantitative researchers strive to ______ errors and to estimate their ______.

Random selection of elements; greater confidence can be placed in the representativeness of these samples.
probability sampling

Elements are selected by nonrandom methods; every element usually does NOT usually have a chance for inclusion.
Nonprobability sampling

There is no way to estimate the probability that each element has of being included in a _____ sample.
nonprobability

In ______ sampling, researchers can specify the probability that an element of the population will be included in the sample.
probability

The systematic over or underrepresentation of a population segment on a characteristic relevant the the research question.
sampling bias

A _____ sample is where the researcher identifies population strata and determines how many participants are needed from each stratum.
quota (type of nonprobability)

_______ sampling involves recruiting ALL of the people from an accessible population who meet the eligibility criteria over a specific time interval or for a specified sample size.
consecutive (type of nonprobability)

Ex: in a study of ventassoc pneumonia in ICU pts, accessible pop is ICU pts of a specific hospital, a _______ sample might consist of all eligible pts admitted to that ICU over a 6month period.
consecutive (type of nonprobability)

_________ sampling uses researchers' knowledge about the population to select sample members.
purposive or judgmental (type of nonprobability)

In _____ sampling, researchers establish a list of elements from which the sample will be chosen (a sampling frame).
simple random (a type of probability sampling)

If FNP students at GU were the accessible population, a roster of those students would be the _______.
sample frame

In _______ sampling, the poulation is first divided into two or more strata.
stratified random (a type of probability sampling)

The aim in this type of sampling is to enhance representativeness.
quota (nonprobability)AND stratified random (probability)

_________ sampling designs subdivide the population into homogenous subsets (ex: based on gender or illness severity) from which an appropriate number of elements are selected at random.
stratified random (a type of probability sampling)

A difficulty with _______ sampling.
identifying attributes must be known in advance and may not be readily discernible

Quota sampling does not have the same problem as stratified random sampling. Why not?
Because in quota sampling, the researchers can ask people quesitons that determine their eligibility for a particular stratum.

_______ sampling involves selecting broad groups rather than selecting individuals.
cluster (a type of probability sampling)

_______sampling is typically the first stage of a multistage approach.
cluster (a type of probability sampling)

Selecting every kth case from a list.
Systematic sampling (a type of probability sampling)

When researchers calculate a percentage or an average based on sample data, they are estimating a __________.
population value

_____ expresses the strength of relationships among research variables.
effect size (power analysis)

Steps in sampling.
 1. identify the population
 2. specify the eligibility criteria
 3. specify the sampling plan
 4. recruit the sample

The approximate truth of an inference
validity

Validity of inferences that there truly is an empirical relationship or correlation b/w presumed cause & effect.
statistical conclusion validity

______ validity concerns the validity of inferences that it is the independent variable (not something else) that caused the outcome.
internal

_____ validity involves the validity of inferences from the observed persons, settings, and causeandeffect ops included in the study to the constructs that these instances might represent.
Construct

Threats to _______ validity concern ways in which relationships b/w variables might interact with or be moderated by variations in people, settings, time, and conditions.
external

A threat to external validity where an effect observed with certain types of people might not be observed with other types.
Interaction b/w relationship and people

______ validity has most often been a concern to researchers in connection with the ________.
measurement of outcomes

Threats to construct validity can occur if the operationalization of the construct fails to incorporate _______ of the underlying construct, or if int includes _______.
 relevant characteristics
 extraneous

Threat to construct validity where participants may behave differently b/c they are aware of their role in a study.
Reactivity to the situation

Construct validity requires careful attantion to ________, so that appropriate construct inferences can be made.
what we CALL things (like construct labels)

________ concerns the extent to which the implementation of an intervention is faithful to its plan.
Intervention/Treatment fidelity (part of unreliable implementation of a treatiment  a threat to statistical conclusion validity)

The ability to detect true relationships among variables.
statistical power (low = threat to statistical conclusion validity)

Researchers should aim to _______ group differences on the dependent variables by _______ differences on the independent variable.

Precision is an aspect of what type of validity?
statistical conclusion

When the use of homogeneity restricts the range of values on the outcome variable, the relationships b/w the outcome and the independent variable will be ______ and may lead to an erroneous inference that the variables are UNrelated.
attenuated (part of Restriction of Range, a threat to statistical conclusion validity)

______ validity is important b/c it links the ops used in a study to a relevant conceptualization and mechanisms for translating the resulting evidence into practice.
Construct

Participants behaving in a particular manner b/c they are aware of their role.
The Hawthorne effect (reactivity to the study situation, a threat to construct validity)

