Sampling

Question 1

What is Sampling?

Answer 1

Sampling is a process of selecting a sufficient number of elements from a population of interest so that by studying the sample, we may generalize the results back to the population from which these elements were chosen. (Sekaran, 1992)

Question 2

Why do we sample?

Answer 2

• Research projects usually have budget constraints
• Information can be gathered quickly - could be days, weeks, hours (or long-term like census)
• Samples (if properly selected) are sufficiently accurate

Sample may even be more accurate
– If you did research for the whole population, so first results would be invalid by the time you finish gathering info
– Should do fieldwork as quickly as possible
– Higher sample = higher probability of mistakes
You may not need to sample at all; may not make sense

To choose a sample representing the whole population, you’d think about:

• Not practical or ethical to ask very old people
• • Highest age is usually 75, 65, sometimes 79
• • Depends on target group - if children, your range might be 3-10
• Usually restricted by budget, time, number of questions, practicalities, etc.
• Qualitative research is never “representative”

Size doesn’t matter

• • If you choose only two balls from the basket, you will likely not get representative data, so you might need 10
• • If you have two pots of soup, one big one small, you would take the same spoonful to test each one

Question 3

What does sample size depend on?

Answer 3

Research purpose
Type of data analysis
On population characteristics

If population is homogeneous, the smaller sample would be ok
If population has 50 different regions, you would need a bigger sample

Question 4

What is sample size influenced by?

Answer 4

Judgement about how typical the same may be of total population

Degree of required accuracy (95% is a commonly used figure)

Difficulty and cost of using a large sample

Number of categories in which you intend to subdivide the data

Number of anticipated non-responses
– If it’s not enough, you have to draw another sample

Question 5

Sampling Element

Answer 5

Case or unit of analysis of the population that can be selected for a sample ( person, organization, group, company etc.)

Question 6

Sampling Frame

Answer 6

“That list or quasi list of units composing a population from which a sample is selected. If the sample is to be representative of the population, it is essential that the sampling frame include all (or nearly all) members of the population. See Chapter 5.”

Specific list of sampling elements in the theoretical population
– Perfect = List of 10,000 names with name and email

Question 7

Population Parameter

Answer 7

Characteristic of entire population that you estimate from a sample

Question 8

Sampling Ratio

Answer 8

“The proportion of elements in the population that are selected to be in a sample. See Chapter 5.”

Ratio of the sample size to the size of target population

Question 9

Theoretical Population

Answer 9

The group you wish to generalize to (AKA target population)

Question 10

Study Population

Answer 10

“That aggregation of elements from which a sample is actually selected. See Chapter 5.”

Part of the theoretical population that is accessible for the research (AKA accessible p, survey p, or reachable p)

Question 11

Sample

Answer 11

Group of people who you select to be in the study

Question 12

The Problem - From All to Some

Answer 12

Variable
 ↓
Statistic
 ↓
Parameter

• Must be very careful about interpreting findings
• The result is always the result just from the sample but not always representative of the whole population; not really the findings
• Research estimates rather than research findings
• Based on this number you only estimate the percent of the whole population; an assumption/estimation

Question 13

Stages in Selection of a Sample (6 steps)

Answer 13

1. Define the theoretical and study populations
↓
2. Identify a sampling frame
↓
3. Determine sampling method
↓
4. Determine sample size
↓
5. Select actual sampling units
↓
6. Collect data

Question 14

Defining the Theoretical and Study Populations

Answer 14

1st stage of selecting a sample

• • Group of specific population elements
• • Concerns questions about critical characteristics of the population

Question 15

Identifying a Sampling Frame

Answer 15

2nd stage of selecting a sample

• • List of elements from which a sample may be drawn (ex: phone book)
• • Ideally, the source should be representative of the population
• • Source should not bias the results (Sampling Frame Error occurs when certain sample elements are not represented in the study population) (ex: people not listed in a phone book)
• • RDD - Random Digit Dialing
• – you’ll still only get people with phones
• – one number might be shared or one person might have multiple numbers

Question 16

Determine sampling method

Answer 16

3rd stage of selecting a sample

Probability sampling - each person/element has an equal chance of being selected as part of the sample

• • Simple random sampling
• • Systematic random sample
• • Stratified random sample
• • Cluster (area) random sample
• • Multistage sampling

Non-probability sampling

• • Convenience
• • Purposive (judgemental)
• • Quota
• • Snowball

Question 17

Simple Random Sampling (SRS)

Answer 17

“(1) A type of probability sampling in which the units composing a population are assigned numbers. A set of random numbers is then generated, and the units having those numbers are included in the sample. Although probability theory and the calculations it provides assume this basic sampling method, it’s seldom used, for practical reasons. An equivalent alternative is the systematic sample (with a random start). See Chapter 5. (2) A random sample with a low IQ.”

The simplest form of random sampling

Objective: to select n units out of N such that each element has an equal chance of being selected

Procedure: Create a sample frame for all cases, then select cases using a purely random process
– Pull names out of a hat
– Table of random numbers
— Don’t start at beginning of list; if choosing from 185, look only at first 3 numbers of each
– Allocate random numbers to cases (use a table of random numbers, a computer random number generator, or a mechanical device to select the sample)
Ex: diceware

Question 18

Systematic Random Sampling

Answer 18

“(1) A type of probability sampling in which every Kth unit in a list is selected for inclusion in the sample—for example, every 25th student in the college directory of students. You compute k by dividing the size of the population by the desired sample size; K is called the sampling interval. Within certain constraints, systematic sampling is a functional equivalent
of simple random sampling and usually easier to do. Typically, the first unit is selected at ran‐ dom. See Chapter 5. (2) Picking every third one whether it’s icy or not. See snowball sampling (2).”

N = 100

Want n = 20

N/n = 5

Select a random number from 1 to 5: chose 4

Start with number 4 and take every 5th unit

Question 19

Stratified Random Sampling

Answer 19

“The grouping of the units composing a population into homogeneous groups (or strata) before sampling. This procedure, which may be used in conjunction with simple random, systematic, or cluster sampling, improves the representativeness of a sample, at least in terms of the stratification variables. See Chapter 5.”

• Involves dividing the population into homogenous subgroups and then taking a simple random sample in each subgroup
• By using the stratified random sampling, you can ensure a balance of particular traits (gender, age, etc.)

Two types of stratified random sampling: if true representation is 80% female and 20% male
1. Proportionate - take representative sample, 80% female, 20% male
This wouldn’t allow any further information within the sub-group of males; too small of a group to analyze
2. Disproportionate - you can take 50% male and 50% female
Will give you a more sufficient number of cases to analyze

Post-stratification wave - divide number by 10

Question 20

Cluster (area) Random Sampling

Answer 20

“(1) A multistage sampling in which natural groups (clusters) are sampled initially, with the members of each selected group being subsampled afterward. For example, you might select a sample of U.S. colleges and universities from a directory, get lists of the students at all the selected schools, then draw samples of students from each. See Chapter 5. (2) Pawing around in a box of macadamia nut clusters to take all the big ones for yourself.”

In cluster sampling, you follow these steps:

1. Identify subgroups (clusters) of the population
2. Create a list (a sampling frame) of the clusters
3. Randomly select an agreed number of clusters
4. Involve all units within sampled clusters

Ex: list of all schools in the country - these are you clusters

Question 21

Multi-Stage Sampling

Answer 21

The most important principle is that we can combine the simple methods described earlier in a variety of useful ways that help to address the sampling needs in the most efficient and effective manner possible.

Example:

1. Begin with national sample of school districts stratified by economics and educational level
2. Within selected districts, do a simple randome sample of grade schools
3. Within schools, do a simple randome sample of classes or grades
4. Within classes do an SRS of students

Question 22

Probablity (5 methods)
vs
Nonprobability Sampling (4 methods)

Answer 22

The difference between nonprobability and probability sampling is that nonprob does not involve random selection and prob sampling does

Probability sampling:

1. Simple random sampling
2. Systematic random sample
3. Stratified random sample
4. Cluster (area) random sample
5. Multistage sampling

Non-probability sampling

1. Convenience
2. Purposive (judgemental)
3. Quota
4. Snowball

Question 23

Snowball Sampling

Answer 23

“(1) A nonprobability sam‐ pling method, often employed in field research, whereby each person interviewed may be asked to suggest additional people for interviewing. See Chapters 5 and 11. (2) Picking the icy ones to throw at your methods instructor.”

Nonrandom sample in which selection is based on connections in a pre-existing network. The crucial feature is that each person or case has a connection with the others/

This method is used whenever you are dealing with a tricky subject, where people may be doing devious or illegal activities (for example hacking, virus creation, etc.)

Question 24

Quota Sampling

Answer 24

“A type of nonprobability sampling in which units are selected into a sample on the basis of prespecified characteristics, so that the total sample will have the same distribution of characteristics assumed to exist in the population being studied. See Chapter 5.”

Question 25

Purposive (judgmental) Sampling

Answer 25

“A type of nonprobability sampling in which the units to be observed are selected on the basis of the researcher’s judgment about which ones will be the most useful or representative. See Chapter 5.”

Question 26

Convenience Sampling

Answer 26

“a non-probability sampling technique where subjects are selected because of their convenient accessibility and proximity to the researcher.”