Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
28 Cards in this Set
- Front
- Back
Sections of a Questionnaire
|
1. Interviewer Introduction and Purpose
2. Request for Cooperation 3. Instructions 4. Screening Questions 5. Questions 6. Demographic Questions |
|
Guidelines for Questionnaire Design
|
1. No unnecessary questions
2. Consider ability and willingness to respond 3. Use simple words 4. No ambiguous words 5. No biasing, leading questions 6. Don't impose implicit alternatives 7. Don't assume respondent is knowledgable 8. Don't put math on the respondent 9. No double-barrelled questions 10. Sequence of questions should be logical 11. Start with simple questions, have difficult ones later 12. Keep demographic questions at the end 13. Use special methods for personal/embarrassing questions 14. Make questionnaire appealing to the eye 15. Do data analysis with dummy data 16. PRETEST |
|
Types of Questions
|
1. Open ended
2. Close ended |
|
Reasons to Use Open-Ended Questions
|
1. Intro to a survey or topic
2. When it's important to measure the salience of an issue 3. Too many responses listed 4. Verbatim responses desired 5. Behaviour to be measured is sensitive or disapproved |
|
Advantages of Open-Ended Questions
|
1. Wide range of responses
2. Responses obtained without influence 3. Free choices |
|
Disadvantages of Open-Ended Questions
|
1. Variability of clarity and responses
2. Time consuming 3. Subjective judgements, summarization prone to error 4. Answers expand or contract according to space/time available 5. Respondents may not use same frame of reference when options aren't available |
|
Formats for Close-Ended Questions
|
1. choice from a list of responses
2. appropriate single choice rating on a scale |
|
Advantages of Close-Ended Questions
|
1. Easier to answer
2. Require less effort by interviewer 3. Easier tabulation/analysis 4. Less potential error in asking/recording 5. Directly comparable between respondents |
|
Limitations of Close-Ended Questions
|
1. Disagreement on type of responses listed
2. Answer received no matter how irrelevant question is 3. May not produce meaningful results 4. Dichotomous questions prone to measurement error because it is polarized 5. Fewer options of self expression 6. Alternative responses provide answers not considered by respondent, leading |
|
Approaches to asking sensitive questions
|
1. Casual approach
2. numbered card 3. everybody approach 4. sealed ballot technique 5. Kinsey approach 6. Randomized response technique |
|
Reasons for Pretesting Specific Questions
|
1. Variation
2. Meaning 3. Task Difficulty 4. Respondent's interest/attention |
|
Reasons to Pretest Questionnaire
|
1. Test flow for clarity/logic
2. Ensure skip patterns are clear 3. Time each section so not too long 4. Capture and maintain interest/attention |
|
Considerations for International Research
|
1. Open-ended questions to avoid cultural bias
2. Indirect form for sensitive questions 3. Nonverbal stimuli when illiteracy is high 4.Re-word questions 5. Most problems relate to attitudes, psychographic and lifestyle data |
|
Questionnaire Design Process
|
1. Determine survey objectives, resources and constraints
2. Determine data collection method 3. Determine question response format 4. Decide on wording 5. Establish flow and layout 6. Evaluate questionnaire 7. Obtain approval from all relevant parties 8. Pretest and revise 9. Prepare final copy 10. Implement |
|
Developing a Sampling Plan
|
1. Define population
2. Choose data collection method 3. Identify sampling frame 4. Select sampling method 5. Determine sample size 6. Develop operation procedures for selecting sampling elements 7. Evaluate operational sampling plan |
|
When is census appropriate?
|
1. Small population size
2. Info needed form every individual 3. Cost of making incorrect decision is high 4. sampling errors high |
|
When is a sample appropriate?
|
1. Large population size
2. Cost/time to get info is high 3. Quick decision needed 4. INcrease response quality- more individual time 5. Population is homogenous 6. Census impossible |
|
Types of Non-Probability Samples
|
1. Convenience
2. Judgemental 3. Quota 4. Snowball |
|
Types of Probability Samples
|
1. Simple Random Sample
2. Systematic 3. Stratified 4. Cluster |
|
Types of Stratified Sampling
|
1. Proportionate
2. Disproportionate |
|
Elements of Stratified Sampling
|
1. Homogeneity within group
2. Heterogeneity between groups 3. All groups included 4. Sampling efficiency improved by increasing accuracy at faster rate than cost |
|
Elements of Cluster Sampling
|
1. Homogeneity between groups
2. Heterogeneity within groups 3. Random selection of groups 4. Sampling efficiency improved by decreasing costs faster rate than accuracy |
|
When to Use Non-Probability Sampling
|
1. Exploratory stages
2. Pre-testing 3. Homogenous population 4. Lacks statistical knowledge 5. Operational ease required |
|
When to Use Ad Hoc Methods of Sample Size
|
1. Know from experience
2. Budgetary Constraints |
|
Rules of Thumb for Determining Sample Size
|
1. Each group minimum sample of 100
2. 20-50 fro each sub group 3. Use disproportionate sampling if one group of population is relatively small |
|
Factors Determining Sample Size
|
1. number of groups and subgroups within sample
2. value of info 3. accuracy rate 4. cost of sample 5. variability of population |
|
Population/Sample Characteristics
|
1. mean
2. variance/standard deviation |
|
Determining Statistical Precision
|
1. Sample reliability
2. Standard error 3. interval estimation |