Text Ch.8 Flashcards
Sampling links in with ______ validity
external
-the extent to which the findings drawn from the cases under examination may be used to make generalizations about phenomena outside the original study. Concerned with representativeness.
Sampling Error is the difference between ______ _______ and ______ ______
Population Parameter and Sample Statistic
Population Parameter Vs Sample Statistic
p. p= population characteristics, expressed in numeric terms when the responses of each member (or case) of the population are measured
s. s=
Population Parameter Vs Sample Statistic
p. p= population characteristics, expressed in numeric terms when the responses of each member (or case) of the population are measured
s. s=a subset of the p.p
3 factors in sample representativeness
- Sample Frame
-a list of the items or people forming a population from which a sample is taken.
Consider:
– Availability (or lack thereof)
– Completeness
– Accuracy - Sample Selection
- probability sampling
- non probability sampling - Sample size(if a probability sample)
- as sample size goes up, error goes down
Describe differences between probability and non probability sampling
Probability Sampling
- Based on probability theory
- Can estimate likelihood of representativeness
- Can estimate accuracy of sample statistics
- Appropriate for statistical analysis
types: • Simple random • Systematic selection • Stratified • Cluster
Non-Probability Sampling
-NOT based on probability theory
-CanNOTestimate likelihood of representativeness
-CanNOTestimate accuracy of sample statistics
-NOT appropriate for statistical analysis (including percentages/frequencies)
Types: Accidental and Purposeful(most typical)
Critiques of Sampling Frames: for probability sample and for Non-probability
Probability sample:
- What are possible gaps in the list?
- How could the sampling frame possibly introduce bias?
Non-probability sample:
- Convenience or purposive sample?
- Potential for ‘closed loops’?
.Does the goal of the research match the fact that it is non-probability or probability
Qualitative Research and NonprobabilitySampling
Words/text, rather than numbers
Deep description
Seeks transferable, dependable findings
Does not argue representativeness
Why use Non Probaility sampling
• Exploratory research
• Extremely small populations
• Unavailable/inadequate sampling frames
• High refusal populations
**Typically will stress that results may not be representative
Critiquing Sampling Size
Probability sample: . Homogeneity of population . Number of variables . Sample selection - Cluster sampling needs large sample sizes
Data Saturation
Data saturation: point at which data collection no longer provides the researcher with new information
Non Probability sampling: how to choose sampling size
Focus on quality of information obtained, rather than quantity of cases
Data saturation: point at which data collection no longer provides the researcher with new information
population
in research the group that a researcher wishes to generalize about(ex: Canadian citizens); in content analysis, the texts most appropriate for answering the research questions
sampling
the process of drawing a sample of cases from a larger population
First step in a research project is to clearly identify the population being studied: 3 factors
- Unit of analysis(indv,political parties,etc)
- Geographic location
- Reference perios
representative sample
one that accurately represents the larger population from which it was taken
sampling frame
is a list of all of the units in the target population
sample selection method
2 main categories
- probability
2. nON PROBABILITY
Probability sampling
sampling based on probability theory allows us to estimate the likelihood that our sample provides a representative picture of the population
-random selection of a sample, each case in a population has an equal opportunity to be selected
simple random sampling
the process by which every case in the population is listed and the sample is selected randomly from this list
sampling distribution
theoretical distribution of a sample statistic (ex:the mean) for a given sample size
confidence interval
the range of values within which the population parameter is likely to fall
Sample Size
.when probability sampling techniques are used, sampling error is reduced as the sample size increases
Homogeneity vs. Heterogeneity
refers to how similar a population is with respect to salient variables
refers to how dissimilar a population is
the more homogenous a population, the smaller the sample size needed
3 factors that influence sample size needs
- Homogeneity of a population
- Desired Degree of accuracy
- The complexity(number of variables) of the research
systematic selection
a probability sampling technique in which the researcher calculates a selection interval and uses it to select cases from the sampling frame
ex: is selection interval is 20, you select a number from 1-20 and start going 6, 26,46,66,86,106
- less random and less accurate than random sampling
Stratified Sampling
.involves breaking the population into mutually exclusive subgroups, or strata, and then randomly sampling each group
- each group is its own sampling frame
- disproportionate stratified sampling is done when one subgroup of the population is way smaller but needs to be enhanced for sampling
- Disproportionate is NOT representative of the population as a whole
cluster sampling
is the process of dividing the population into a number of subgroups, known as clusters, and then randomly selecting clusters within which to randomly sample
- saves time and money
- not really representative either
Types of probability sampling(3)
- Systematic selection
- stratifies sampling
- Cluster Sampling
Types of non-probability sampling(2)
- accidental sampling
- convenience or haphazard
- the researcher gathers data from indv whome he or she accidently encounters or are convenient
- self selection is just as biased - Purposive(judgemental) sampling
- involves researcher selection of specific cases; the researcher uses his or her judgement to select cases that will provide the greatest amount of info
- snowball sampling is often employed to study social networks
- quota sampling when accidentall or purposive sampling is combined with stratification
T OR F
Non Probability sampling methods employ random selection of cases
FALSEEEEE
-this means that we cannot identify margins of error or confidence intervals; thus it is more difficult to make generalizations and draw conclusions about the general population
When are non-probability sampling appropriate?
when researchers are not seeking to make generalizations about a larger population, but are pursuing more in-depth descriptive information
-also good for when probability sampling is not feasible: when sampling frames are inadequate, members of the population are likely to refuse to participate, or the population of study is extremely small
Snowball sampling is what type of sampling?
describe Snowball sampling
non probability, purposive sampling
the researcher begins by identifying a few cases and , from these, gets referalls for other cases and continues to branch out
spurious
the relationship between the independent and dependent variables that, while initially thought to be causal, is non-causal and is a function of the presence of athird variable, which causes the variation in both variables
quota sampling is what type of sampling?
describe quota sampling
non probability and probability, purposive and stratified sampling
the researcher combines purposive or accidental sampling with stratification; the researcher identifies a number of target groups(strta) and then sets a quota number that must be met for each group
