SAMPLING TECHNIQUES AND SURVEY DESIGN

random sample or sample

- Every member of the population must have an equal chance of being selected.

- It is important because it gives us an unbiased sample.

(NOTE: If our sample is unbiased, it represents the population we are studying)

= unbiased sample = representative of the population

sampling or selection bias

nonresponse bias

response or interviewer bias

volunteer bias

4 Types of Biased Samples:

Sampling or selection bias

- some subjects are more likely to be included in a survey than others.

(NOTE: Also known as the "convenience sampling technique." They are more accessible and free)

Nonresponse bias

- Subjects who do not respond to a survey question would answer it differently than those who do respond to it.

(NOTE: Also known as the one that got away)

Response or interviewer bias

- The subject gives a different response than the respondent truly believes.

Volunteer bias

- are used since they might be more interested in the survey and answer questions or participate differently than randomly selected subjects

(NOTE: Not considered a random sample if volunteers only answer, it would cause possible bias)

Acquiescence Bias (Yea-saying)

Social Desirability Bias

Extreme Responding

Central Tendency Bias

4 Different Response Bias:

Acquiescence Bias (Yea-saying)

- Tendency to agree with statements regardless of content.

Typical Trigger: Agree/disagree Likert items; deference to authority.

Example: Respondent answers "Agree" to both "I feel happy" and "I feel sad."

Social Desirability Bias

- Overreporting "good" behaviors, underreporting "bad" ones.

Typical Trigger: Sensitive topics (health, ethics, habits).

EXAMPLE: Claiming to exercise daily when they don't.

Extreme Responding

- Favoring endpoints of a scale (always 1 or 5).

- intentionally

Typical Trigger: Personality traits,

cultural norms.

Example: Choosing "Strongly Agree" for most items.

Central Tendency Bias

- Clustering around the midpoint of scales.

(NOTE: Neutral answers or play safe, because they fear committing to an opinion)

Typical Trigger: Fear of committing to an opinion.

EXAMPLE: Always selecting "Neutral."

sample

- is used to get information about a population for several reasons:

þ It saves the researcher time and money.

þ It enables the researcher to get information that they may not be able to obtain otherwise.

þ It enables the researcher to get more detailed information about a particular subject.

(NOTE: The best thing to do is to select a sample that will be representative with respect to some characteristics, preferably those pertaining to the study. Inclusion and Exclusion Criteria.)

Simple Random Sampling

Systematic Sampling

Stratified Sampling

Cluster Sampling

4 Sampling Methods:

Simple Random Sampling

- Use methods such as random numbers, which can be generated from calculators, computers, or tables.

- For a sample size n, all possible samples of this size have an equal chance of being selected from the population.

(NOTE: One limitation: If the population is extremely large, it is time-consuming to number and select the sample elements.)

Incorrect methods:

- Ask "the person on the street."

- Ask a question by either radio or television and have listeners or viewers call the station to give their responses or opinions.

- Ask people to respond my mail or email.

Fishbowl method

- Number each element of the population and then place the numbers on cards.

- Place the cards in a hat or fishbowl, mix them, and then select the sample by drawing the cards.

(NOTE: Correct method of Simple Random Sampling, But if not mixed well, numbers chosen are placed in the bowl last.)

Random Numbers

- Each digit, 0-9, has an equal probability of occurring.

- In every sequence of 10 digits, each digit has a probability of 1/10 occurring.

- This can be generated by a computer in a table.

(NOTE: Correct method of Simple Random Sampling. Number the elements of the population sequentially and then select each person by using random numbers.)

Systematic Sampling/Systematic Sample

- sample obtained by numbering each element in the population selecting some random starting point, and then selecting every kth element from the population to be included in the sample.

- But be careful on how the items are arranged on the list!

Stratified Sampling/Stratified Sample

- obtained by dividing the population into subgroups, called strata.

Example: (1) Males who smoke, (2) males who don't smoke, (3) females who smoke, and (4) females who don't smoke

(NOTE: However, it may not be representative on the basis of other characteristics.)

strata

- according to various homogeneous (similar) characteristics and then randomly selecting members from each stratum for the sample.

Cluster Sampling/Cluster Sample

- Obtained by selecting a pre-existing or natural group, called a cluster, and using the members in the cluster for the sample.

- We may use all units of a cluster if it is feasible, or we may select only part of the cluster to use as a sample.

- This selection is done by random methods.

Advantages:

- A ___ sample can reduce costs.

- It can simplify fieldwork.

- It is convenient.

Disadvantage:

- the elements in the ___ may not have the same variations in characteristics as elements selected individually from a population.

- The groups of people may be more homogeneous (alike) in specific ____ such as neighbourhoods or clubs.

Double Sampling

Multistage Sampling

Convenience Sampling

Snowball Sampling

4 Other types of sampling techniques:

Double Sampling

- a very large population is given a questionnaire to determine who meets the qualifications for a study (census sampling ➡ purposive sampling).

- After the questionnaires are reviewed, a second, smaller population is defined.

- Then a sample is selected from this group.

Multistage Sampling

- uses a combination of sampling methods.

Convenience Sampling

- selects subjects from the population who are available to use; not usually representative of the population = biased conclusions.

Snowball Sampling

- use of referrals

Step 1:

Steps for Conducting A Sample Survey: Decide what information is needed.

Step 2:

Steps for Conducting A Sample Survey: Determine how the data will be collected (mail survey, phone interview, etc.).

Step 3:

Steps for Conducting A Sample Survey: Select the information-gathering instrument or design the questionnaire if one is not available.

Step 4:

Steps for Conducting A Sample Survey: Set up a sampling list, if possible.

Step 5:

Steps for Conducting A Sample Survey: Select the best method for obtaining the sample (random, systematic, stratified, cluster, or other).

Step 6:

Steps for Conducting A Sample Survey: Conduct the survey and collect the data.

Step 7:

Steps for Conducting A Sample Survey: Tabulate the data.

Step 8:

Steps for Conducting A Sample Survey: Conduct the statistical analysis.

Step 9:

Steps for Conducting A Sample Survey: Report the results.

Survey

- conducted when a sample of individuals is asked to respond to questions about a particular subject.

Interviewer-administered

Self-administered

2 types of surveys:

Interviewer-administered

- requires a person to ask the questions; face-to-face in office, street, or in the mall, or via telephone.

Self-administered

- done by email, mail, or internet; or in a group setting like classroom.

Asking biased questions

Using confusing words

Asking double-barreled questions

Using double negatives in questions

Ordering questions improperly

(5) Avoid these mistakes when writing questions:

Asking biased questions

AVOID: leading on to answer a question the way researchers want to:

Using confusing words

AVOID: respondent misinterprets the meaning of the words and answers in a biased way:

Asking double-barreled questions

- questions that contain compound sentences that require respondent to answer to two questions at the same time:

Using double negatives in questions

AVOID: "not" is used twice in a sentence.”

Ordering questions improperly

AVOID: by arranging the questions in a certain order, the researcher can lead the participant to respond in a way that the researcher may otherwise not have done.

Open-ended questions

Close-ended questions

2 Types of questions asked influence the responses:

Pilot Study

- should be done to test the design and usage of the questionnaire (i.e. validity).

- It helps researchers to pre-test the questionnaire to determine if it meets the objectives of the study.

- It also helps in rewriting questions that are misleading, ambiguous, etc.