Nonprobability Sample
Designs
1. Convenience samples 2. purposive
or judgmental samples 3. snowball
Unrepresentative sample
Some characteristics are
overrepresented or
Typical Problems in sampling
frames
1. Incomplete frames-units are missing from list 2. cluster of elements-listed in clusters rather than individually-city blocks 3. blank
True Experiments must have
at least 3 things
1. An experimental and control group 2. variation in the independent variable before
assessment of change in the dependent variable(treatment) 3. random assignment to
The Classic Experimental
Design
Experimental group, control group
randomization pretest posttest
Target population
A set of elements larger than or
different from the population sampled
and to which the researcher would like
Systematic sampling
Select every kth element in a population,where k is determined by dividing the population sixe by the desired sample size. Select a random number between
0 and k and picking that element in the
Survey Sampling
Sampling designed to produce
information about particular
Stratified samples
Done by dividing the population into
groups(strata) that are homogeneous
on one or more traits,then sampling
from each of these groups
Stratified Proportionate
sample
The number of elements selected from each stratum is proportional to that stratum’s representation in the
population
The same number of sampling units from each stratum or a uniform sampling fraction (n/N)
Stratified Disproportionate
sample
Chosen to yield numbers in a stratum to allow intensive analysis of that particular stratum
Variable sampling fractions,total number in each stratum is different,population parameters have to be
Standard error
Allows the researcher to determine the probability that a given sample estimate is close to the actual
population value.
S.E.=standard error,the distribution of all samples about the mean of the samples is S.E.Calculate
Simple random sampling
Numbering all population elements,then selecting enough random numbers to complete a sample of the desired size.It is
simple but inconvenient with large populations
Scale
Type of composite measure composed of several items that have logical or empirical structure among
them
Take account of differing intensity of indicators e.g. Likert scale, Guttman scale
Sampling Theory
Major objective is to provide accurate
estimates of unknown parameters in
population from sample statistics
Sampling Frame
A list of all elements or other units
containing the elements in a
Sampling Error -contd
The larger the sampling error,the
less representative the sample.
Sampling Error
Any difference between the
characteristics of a sample and
the characteristics of a population
Sampling distribution
When an infinite number of independently selected sample values such as the means are placed in a distribution,the distribution is
called the sampling distribution
Sample generalizability
Refers to the ability to generalize from
a sample ,or subset of a larger
Sample
A subset of a population that is used to
study the population as a whole.
Representative sample
A sample that “looks” like the
population from which it was selected
in all respects that are potentially
Random selection procedures
Ensure that every sampling unit of the population has an equal and known
probability of being included in the sample,the probability is n/N n=sample, N=population
Random Selection
Each element has an equal chance of
selection independent of any other
Quota sample
Select respondents such that quotas of
various types of people are filled in
proportion to their prevalence in the
Quasi-experimental design
Subjects are not randomly
assigned to to the experimental
and control or comparison group
Purposive or judgmental
sample
Select a sample that, in their
subjective judgment,is
Procedures of Control
1. Randomization or random assignment-removes bias from the assignment process by relying on chance-flipping coin or
random number table assures that case has an equal probability of being assigned to either group 2. matching- or
pairwise matching,for each case in experimental group, another one with identical characteristics is selected for the
Probability vs. Nonprobability
Sampling
Probability sample allows estimates to population from sample Nonprobability
sample-list of sample population is
Probability Sample Designs
1. random sample 2. systematic
samples 3. stratified samples-
proportionate, disproportionate 4.
cluster samples 5. multistage samples
pretests
Measures the dependent variables prior to the experimental intervention,they provide a direct measure of how much the experimental
and comparison groups changed over time,tests effects of intervention
PPS-probability proportionate
to size
Type of multistage cluster sample in
which clusters are selected,not with
equal probabilities(EPSEM) but with
probabilities proportionate to their sizes
Population-finite or infinite
Finite population-contains a countable number of sampling units
Infinite population-consists of an endless number of sampling units,an unlimited
Population
The entire set of individuals or other
entities to which study findings are to
be generalized
Whole=population
Weighting
Assigning different weights to cases that were selected into a sample with different
probabilities of selection.,each case given weight equal to the inverse of its probability of