CHAPTER 6

Question 1

Sampling: What is a census and a sample?

Answer 1

Census: all the members of a population

Sample: a selection of members from a population

Question 2

Define observational error, random error, and bias

Answer 2

Observational error (or just error): Any difference between reported results and true scores

Random error: Refers to mistakes that are equally likely to occur (flip of a coin)
ex. If you sample 500 university students to represent the entire university population, there could be a chance that you get a bunch of smokers in your sample or a bunch of nonsmokers (*should even out once your sample gets big enough)

Bias: A form of systematic error (some mistakes are more likely)

Question 3

What are the two major types of sampling with examples?

Answer 3

Nonprobability Sampling: Any technique in which samples are selected in some fashion not suggested by probability theory

• reliance on available subjects sampling
• purposive (judgmental) sampling
• snowball sampling
• quota sampling
• selecting informants

Probability Sampling: Samples are selected according to probability theory

• EPSEM (equal probability of selection method)
• simple random sampling
• PPS (probability proportionate to size)
• systematic sampling

Question 4

What is probability theory?

Answer 4

A branch of mathematics that provides the tools researchers need to produce representative samples and to statistically analyze results derived from such samples

Question 5

What is the reliance on available subjects (Nonprobability sampling method)
What’s it most commonly referred to as?

Answer 5

*most commonly referred to as CONVENIENT

The use of those available at a particular time
- convenient and inexpensive
- useful for pretesting questionnaires or other social measurements
- used as a stepping stone

ex. pregnancy test advertising - wait outside of a maternity store

Question 6

What is Purposive (Judgmental) Sampling
(Nonprobability sampling method)

Answer 6

A type of non-probability sampling in which you select the units to be observed on the basis of your own judgment about which ones will be most useful or representative
- sample selected based on research question and knowledge of the population
- useful when studying small subset of larger population and members of subpopulation are easily identified

Ex. researcher studying the cultural identity of Canadian students → wanted to make sure she had enough cultural variation - purposefully selected kids with different backgrounds

Question 7

What is Snowball Sampling
(Nonprobability sampling method)

Answer 7

Non-probability sampling method often employed in field research whereby each person interviewed may be asked to suggest additional people for interviewing
- useful when members of a population are difficult to locate
- often used for exploratory purposes
- representativeness is problematic

ex. Words first study of puppy subculture
How would locate people - started with purposive sampling to locate people → then used snowball sampling - asked people who else to talk to - community and groups that could help locate others

Question 8

What is Quota Sampling
(Nonprobability sampling method)

Answer 8

Non-probability sampling in which units are selected into the sample on the basis of pre-specified characteristics, so that the total sample will have the same distribution of characteristics assumed to exist in the population being studied
- begins with a matrix or table describing the target population
- way to try to make it more representative

Question 9

What is Selecting Informants
(Nonprobability sampling method)

Answer 9

Informant: Someone well versed in the social phenomenon that you wish to study and who is willing to tell you what he or she knows
- applicable to field research
- informants may be somewhat marginalized group members (may bias the view you get or their marginal status may limit access to the group)

Ex. shoplifting study at self scanners - mixed methods - one method was to talk to managers with companies and find out what they knew
Especially important in field research

Question 10

What is probability sampling?

Answer 10

The general term for samples selected in accord with probability theory, typically involving some RANDOM selection mechanism
- researchers want precise, statistical descriptions of large populations –> large-scale surveys

Question 11

What is sampling bias?

Answer 11

SYSTEMATIC error derived from using non-probability samples that produce unrepresentative results
- may be conscious or unconscious

Question 12

What are advantages of probability sampling?

Answer 12

1. biases are avoided
2. allows researcher to have reasonable expectation that the sample reflects the population

Question 13

Define …
Element
Population
Study Population

Answer 13

Element = the unit of which a population is composed, and which is selected in a sample

Population = the theoretically specified aggregation of the elements in a study

Study Population = the aggregation of elements from which a sample is actually selected

*elements compose a population

Question 14

How do units of analysis differ from elements?

Answer 14

• elements are used in sample seleciton; units of analysis in data analysis
• elements are often the same as the units of analysis
• element is a unit about which inferences are made
• element is the same as cluster in cluster sampling

Question 15

What is random selection? What is the sampling unit?

Answer 15

Random selection = a sampling method in which each element has an equal chance of selection independent of any other event in the selection process
- requires a list of cases that constitutes the population
ex. flip a coin (should be 50/50 in the long run)

Sampling Unit = the element or set of elements considered for selection in some stage of sampling

Question 16

What is a sampling frame?

Answer 16

The list or quasi-list of units making up a population from which a sample is selected

Question 17

What is simple random sampling?

Answer 17

A type of probability sampling in which the units composing a population are assigned numbers

A set of random number is then generated, and the units having those numbers are included in the sample

Question 18

What is Systematic Sampling with a Random Start?
How do you find the sampling interval?

Answer 18

A type of probability sampling in which ever kth unit in a list is selected for inclusion in the sample
- it is a functional equivalent of simple random sampling and usually easier
- to avoid bias, choose the first element from the list at random

How to compute k (the sampling interval):
Divide the size of the population by the desired sample size

Question 19

Describe how to find the sampling interval AND the sampling ratio (how they relate)

Answer 19

INTERVAL: the standard distance between elements selected in the sample
population size / sample size

RATIO: the proportion of elements in the study population that are actually selected
sample size / population size

**sampling interval and sampling ratio are inverse of each other

Question 20

What is stratified sampling?

Answer 20

Not an alternative to random or systematic sampling; rather, it represents a refinement of both

• takes into account stuff that’s important about the people - we’re different in key ways - HETEROgenous populations with variation
• modification in their use
• a method for obtaining a greater degree of representativeness (decreasing the probable sampling error)

**SIZE of the sample is less important than the REPRESENTATIVES

Question 21

What is periodicity?

Answer 21

An example would be if you’re choosing every 10 apartments to be in your sample, but every 10 are in the corner
- it’s a risk if there is a pattern to your population

Question 21

What is stratification?

Answer 21

The grouping of the units composing a population into homogeneous subsets (strata) with heterogeneity between subsets, and to select the appropriate number of elements from each
- improves the representativeness of a sample, at least with regard to the variables used for stratification

Question 22

What is cluster sampling?

What is multistage cluster sampling?

Answer 22

CLUSTER
Used when it is either impossible or impractical to compile an exhaustive list of the elements that compose a large population
- saves time and money, but not as accurate as other probability samples
ex. select three streets randomly and use all residents on the streets in your sample

MULTISTAGE CLUSTER
*would be better option if you cannot sample every resident on all three streets because of budget constraints

Natural groups (clusters) are sampled initially, with the members of each selected group being subsampled afterward
- clusters are sampled 1st and members of each cluster randomly sampled 2nd using EPSEM (try to maximize the amount of clusters you can sample)
- includes more than one probability sample

ex. select a sample of uni’s, get lists of all the students at all the selected schools, and then draw samples of students from each

Question 23

What is Probability Proportionate to Size Sampling (PPS)

Answer 23

It is a multistage cluster sample in which clusters are selected with probabilities proportionate to their sizes (not equal probabilities)
- this allows that each cluster is given an equal chance of selection as well as each element within each cluster
- whenever the clusters sampled are of greatly differing sizes, it’s appropriate to use the modified sampling design called PPS

Question 24

What does disproportionate sampling and weighting infer?

What is weighting?

Answer 24

Infers that it is sometimes appropriate to give some cases more weight than others

Weighting is a procedure employed in connection with sampling whereby units selected with unequal probabilities are assigned weights in such a manner as to make the sample representative of the population from which it was selected

Question 25

What is the problem that occurs when getting samples of minority groups (sample weighting)

Answer 25

Problem: in order to get sufficient cases of minority groups, researchers need to oversample
- ex. Indigenous peoples constitute about 4% of the population; therefore, in a random sample of 2000 Canadian adults, we would expect to select 80 indigenous cases –> let’s say we end up 160 cases and include that many people in the sample - now our sample overrepresents indigneous peoples

Question 26

What are the 3 steps to sample weighting?

Answer 26

STEP 1. Calculate the proportions (sample proportion and population proportion):

• sample proportion of indg ppl (160/2000) = 0.08
• population proportion of indg ppl
  (80/2000) = 0.04
• sample proportion of non-indg ppl
  (1840/2000) = 0.92
• population proportion of non-indg ppl
  (1920/2000) = 0.96

STEP 2. Determine the weight for each group by dividing desired population proportion by actual sample proportion:

indg weight = 0.04 / 0.08 = 0.50

non-indg weight = 0.96 / 0.92 = 1.043

STEP 3: Apply the weights by multiplying each group sample size by its weight:

indg = 0.5 x 160 = 80

non-indg = 1.043 x 1.840 = 1920

*now the weighted sample reflects the correct proportions of indg and non-indg in the population