Professional Documents
Culture Documents
Research Method
• IV cannot be manipulated
• Inappropriate wording or placement of questions – guiding the
respondents to collect the desired information
• Response rates continue to decline due to the nature of the
research
Types
Steps
1. Identify a research topic
2. Conduct a review of the literature
3. Research objective(s)
4. Develop research question(s) & hypotheses
5. Develop the survey
1. Constructing questions
2. Cover letter
3. Consent form
4. Validity test
5. Pilot test
6. Reliability test
6. Administer survey
7. Data coding
8. Analyzing data
9. Interpreting data
Construction of questions (basic rules)
1. Understanding the goals/objectives of the project/research
2. Questions should be clear and unambiguous (validity)
3. Questions must accurately communicate what is required from the
respondents
4. Don’t assume respondents understand the questions you are asked
5. Questionnaire design depends on the choice of data collection
technique (e.g., mail survey– easy to read; telephone survey—
cannot use long lists of response options, etc.)
Construction of questions (cont…)
• Types of questions (i.e., closed-ended or open-ended)
• Selection-type items (Likert, ranked, checklist)
• Define or explain ambiguous terms. E.g., Do you spend a lot of time
each week studying Science?
• Avoiding leading questions = don’t you agree…
• Avoid sensitive questions = did you have sex during the past two
weeks? Do you watch porn movies? What is your annual income?
• Do not ask double-barreled questions = “and”
External Validity
Sampling
• Simple random sampling
• Stratified sampling
• Cluster sampling
• Systematic sampling
Simple Random Sampling
• All individuals in the defined population have an equal and
independent chance of selection for the sample.
• Is completely out of the researcher’s control
• So, every individual has the same probability of being selected
• The sample may not have the exact same ratio of males to females,
but the ration will be close
Stratified random sampling
• Some populations can be subdivided into subgroups. For example,…?
• Strata/stratum
• Strategically select participants from each subgroup (equal selection)
• It’s good if the research goal is to compare the behaviour of
participants from different subgroups of the population.
Difference between stratified and cluster
samplings
• Stratified = must have an equal selection from each group that is
proportionate to the population.