Chapter 7 - Selecting the Sample and Setting
- rarley reprsents whole population - sampling bias - results could be misleading
3 main problems with Non probability sampling
sampling frame
A list of all elements in a population.
sampling frame
A list of all subjects, objects, events, or units in the population that is used for a simple random sample
all members of popoulation odnt have independent chance of being selected for the sample
A major difference between systematic sampling and other types of sampling
- easy to obatin - saves time and money
Advantages of Convenience sampling
- sample not subject to bais - representativeness to population is maximsized - differences are purple a fucntion of chance - proabilty of choosing a non-representive smaple decreases
Advantages of simple random sampling
- enhances representativeness - comparisons can be made among groups
Advantages of stratified random sampling
random assignment
Allocation of subjects to either an experimental or a control group.
population
An entire set of subjects, objects, events, or elements being studied. It is a well-defined group whose members possess specific attributes.
sampling bais
Disadvantage of purposive sampling
- biais for smapling occurs - rarely representative of whole population - results could be misleading
Disadvantages to nonprobabilty sampling
- potential for sampling bias - the use of a sample that may not represent the population - limited ability for results to be generalized
Disdvantages of Convenience sampling
- time consuming - may be impossible to obtaini a good sample frame
Disdvantages of simple random sampling
sampling biais
Distortions that arise when a sample is NOT representative of the population from which it is drawn.
snowball
Early sample members are asked to refer other people who meet criteria in...
population
Entire set of subjects, objects, events, or elements being studied; also called the target population.
external validity
Extent to which results of a study can be generalized from the study sample to other populations and settings
power analysis
Formulas based on number of variables and study design to determine best sample sizes
through power analysis
How are sample sizes determines for quantitate research designs?
researchers need to consider the period of history during which the study was conducted and the effect of the reported findings on nursing practice and societal events during that time. can you generalize results to different periods in the past or future? can answer this quesiton with replication studies that corroborate an est relationship b/t results from each time (ex: cant apply research founf in 1950s to today)
How can Interaction of History and Treatment affect external validity?
When samples are confined to certain types of subjects, it is not reasonable to generalize results to those who do not have these characteristics. (Ex: cant compare teenagers prognosis with cancer to adults)
How can Interaction of Selection and Treatment affect external validity?
Bias exists when members of different settings agree to participate in studies (Ex: results of rehab in hosital cant be generlized to rehab in nursing homes)
How can Interaction of setting and Treatment affect external validity?
need to know the specific composition of the population
How can you get a proportional quota sampling?
Stratified random sampling achieves a greater degree of representativeness with each subgroup, or stratum, of the population
How does stratified change the degree of representivaness of the population
Decreases ability to make generalizations
How does the ability to make generlizations change for nonprobabilty samples?
the researcher collects data from whoever is available and meets the study criteria
How does the researcher collect samples in Convenience sampling
samples are taken until the researcher has a clear understanding of phenoma
How is sampling in qualitative designs?
large enough to be an accurate representation of the population and have statistical significant results
How large should a sample be?
1. idetify populatio 2. determine the smaple size 3. list members to accessible population 4. assign all indivualds on list to a consectiev number 5. Select an arbitrary number in the table of random numbers (just close your eyes and point to one). From that number, read consecutive numbers in some direction (horizontally, vertically). 6. if you have less than 9 1 number is used, less than 99 2 numbers, less than 999 then 3 and so on
How to use a table of random numbers for simple random sampling
nonprobabilty
In _________ sampling, chance plays no role in determination of the sample
individuals; groups
In cluster sampling _____ are not randomly selected but ______ are
cluster sampling
In drawing a sample of nursing students in US, 1st draw a random sample of nursing schools and then draw a random sample of students from those schools what is this?
groups; individuals
In stratified random sampling _______ are not randomly selected, but _______ are randomly selected from the different groups
simple random sampling
Method of selecting subjects for a sample, in which every subject has an equal chance of being chosen.
probability sampling
More likelihood that you will get a representative sample with what classification of sampling?
has to be representative of the whole population
Most important chacatertics of sample
QUALITATIVE; QUANTITATIVE
Nonprobabilty sampling is more approatie to be used for a _________ methods Is less desirable for ________ methods but may be the only feasible alternative
convenience (Not randomly selected. Subjects recruited until quota is filled)
Once the quota for each stratum is determined, subjects are solicited via _______ sampling
target population
Population for which study outcomes are intended.
accessible population
Population that is readily available to the researcher and that represents the target population as closely as possible.
is not
Random selection (is/is not) a prerequisite for random assignment.
methodology section
Sample & Setting should apper in what section of the report?
external validity
Samples that are biased or too small, however, threaten the ________ of the design.
Population divided into strata. Each strata group should be homogeneous then random selection from the stratified sampling frames Can be stratified according to age, gender, race, disease state, geographical location etc
Stratified Random Sampling
true (Random samples are only considered to represent the target population; however, it is possible that they do not)
T/F No sampling technique guarantees a representative sample.
false
T/F Systemic sampling is strictly probability sampling
true
T/F there are no hard and fast rules about sample size
accessible population
The population of people that is within a target population and are readily available to the researcher and that represents the target population as closely as possible.
stratified random sampling
The process of selecting a sample to identify subgroups in the population that are represented in the sample
sampling
The process of selecting a subset from a larger population.
to gain information from a small group so that findings can be generalized to a larger group
The purpose of selecting a sample
random assignment
The random allocation of subjects to either an experimental or a control group.
- Interaction of Selection and Treatment - Interaction of History and Treatment - Interaction of Setting and Treatment
Threats to the ability to generalize findings (external validity)
1. probability sample 2. non probability sample
Two main catergiores that sampling is classified under
Convenience smapling
Type of Nonprobability Sampling that is the collection of data from subjects or objects readily available or easily accessible to the researcher.
nonprobabilty sampling
Type of category of sampling that is when sample is not selected randomly.
probability sampling
Type of category of sampling that occurs when every subject, object, or element in the population has an equal chance of being chosen.
qutoa sampling
Type of nonprobability sampling in which quotas are filled.
purposive sampling
Type of nonprobability sampling in which the researcher selects only subjects that satisfy prespecified characteristics; also called judgmental or theoretical sampling.
Convenicence Sampling
Type of nonprobability sampling in which the researcher selects subjects or elements readily available; also called accidental sampling.
snowball sampling
Type of nonprobability sampling that relies on subjects identifying other subjects with similar characteristics.
network sampling
Type of nonprobability sampling that takes advantage of social networks.
snowball sampling
Type of nonprobabilty sample that is sampling that relies on previously identified members of a group to identify other members of a population
Quota Sampling; subjects are not randomly selected for each stratum.
Type of nonprobabilty sample that is similar to striftiated random sampling and its main difference;
purposive sampling
Type of nonprobabilty sampling when the researcher "handpicks" or selects certain cases to be included in the study. Used in qualitative research.
systematic random sampling
Type of probability sample in which after the first individual is selected, then the selection of every kth case from a list
simple random sampling
Type of probability sample that involves defining the population and identifying a sampling frame. Every subject has an equal and independent chance of being chosen.
cluster sampling
Type of probability sample the researcher divides the population into separate groups. Then, a simple random sample of groups is selected from the population.
disproportional
Type of quota sampling that occurs when the number of elements in each stratum is not proportional to the number in the target population
stratified random sampling
Type of random sampling in which the population is divided into subpopulations on the basis of one or more variables, and a simple random sample is drawn from each stratum.
snowball
Type of sample that is useful for studies in which the criteria for inclusion specify a certain trait that is ordinarily difficult to find
systematic sampling
Type of sampling in which every kth (where "k" is some convenient number) member of the population is selected into the sample.
probability sampling
Type of sampling in which every subject, object, or element in the population has an equal chance or probability of being chosen.
cluster sampling
Type of sampling in which the researcher selects groups of subjects rather than individual subjects; also called multistage sampling.
non-probabilty sampling
Type of sampling in which the sample is not selected using random selection.
• Convenience (accidental) Snowball/Network • Quota sample • Purposive sample
Types of nonprobabilty sampling
-Simple random -Stratified random -Cluster (multi-stage) -Systematic sampling
Types of probability sampling
developing a rich understanding; seek in-depth understanding of phenomenon. goal is to ID participants being most knowledgeable on topic.
What does sampling for qualitiative research focus on?
the characteristics of the phenomena of interest will be, or are likely to be, present in all of the units being studied.
What does sampling insure for the study?
Type of institution and geographic areas would be the units that define this population.
What is the population for this study? Researchers may be interested in describing the characteristics of private versus public institutions in different geographic locations.
Convenience
What type of sample is this an example of?? Distribute questionnaires to 1st 100 people who walk into store
quota
What type of sample is this? o Nursing school - 10% males and 90% females o Obtain a sample of 10 males and 90 females
systematic sampling
What type of sample is this? Wanting to have 200 patients from a population of 4000, you divide it 4000 by 200 and gte 20. Every 20th person will be chosen.
purposive sampling
What type of sampling is this? Want to study a super rare disease, only 20 people in the country with it, will contact those specific 20 people
when the sample represents the total population
When can the researcher conculde that study results can be generalized to include the entire population and settings being studied?
in situations in which one cannot obtain a list of individuals who share a particular characteristic
When is snowball sampling useful?
inferences; generlization
When sampling is conducted properly, it allows the researcher to draw ______ and make ________ about the population without examining every element in the population.
representative sample
When the main characteristics closely approximate those of the population you have what type of sample?
simple random sampling
Whenever researchers use the phrase "randomly selected," the samples being described are..
representative
With a __________ sample, researchers are in a stronger position to draw conclusions from the sample findings that can be generalized to the population
sample size
a key determinant of sample quality in quantitative research
accidental
another name for convenience sampling
interaction of selection and treatment
difference in partiicpants and those who refuse is an example of what kind of threat to external validity?
no
does reseracher begin with a sampling frame in which each member has an independent chance of being included in nonprob sampling?
replicate studies in different settings and times (replicate a study that was done 20 years ago)
how do you avoid interaction of setting and history with treatment?
may be done by random or stratified sampling
how is selection of individuals from within the cluster?
Researchers identify population strata and determine how many participants are needed from each stratum Researchers guide selection of subjects so that the sample includes an appropriate number of cases from each strata; can be dis/proportionate
quota sampling
description of sample, setting demographics sampling strategy
what should you include in metholodogy section rt sample and setting?
for convenience when the group is very large or spread out geographically
when is cluster sampling used?
