Week 2 - Survey Design Flashcards
List the three research types
Experimental, Quasi-experimental and Non-experimental
What are the purposes of Exploratory and Descriptive research?
Information gathering
What are the purposes of Explanatory and Predictive research?
Theory testing
What is a Research Process?
Examples of iterative (constant) research process models and where survey design and sampling fits in
List the survey research process
- Reality / observation / theory
- Problem definition / hypotheses
- Research method design (incl. survey)
- Collect data
- Analyse
- Results
- Discuss (generalise / apply)
- Disseminate (get reviewed / publish)
- New study?
List the “Hourglass” notion of research
- Start with broad questions
- Narrow it down
- Operationalise
- Observe
- Analyse data
- Research conclusions
- Generalize back to question
List the types of surveys
- Self administered
- Interview administered
List the stages of questionnaire develipment
- Formulate a generic questionnaire
- Expand on the questionnaire
- Pre-test/redraft
- Finalise and implement
Key parts of a survey
- Title page
- Information sheet
- Consent form
- Instructions
- Questionnaire
- End pages
Details of the information sheet
- Study name
- List of researchers
- Purpose of study
- What is required
- Risks/costs/rewards
- How the results will be used
- Ethics approval
- More info: complaints etc
Why pre-test a survey?
- for feedback
- revise redundant, misunderstood and non-competed questions
- double check layout etc
- test on anyone convenient
Why Pilot-test a survey?
- analyse data
- revise survey
- test on a small sample from the target population
Avoid questions that are:
- Double-barreled
- Double negative
- Leading
- Loaded
- Objective
- Subjective
Define a double-barreled question
Questions which have more than one concept or purpose - they should be split into different questions.
E.g. “What should the speed limit be for trucks and cars?”
Split them.
Define a double negative question
They’re confusing because responding with “no” creates a double negative.
E.g. “Do you disapprove (neg) of gay marriage?” Answering with no (neg), even though the answer is positive is confusing.
Define a leading question
A question that suggests an answer.
E.g. “Do you think Jess should earn more money than she is currently paid?!”
Define a loaded question
a question that suggests a socially desirable answer.
E.g. “Have you stopped beating your wife?” Yes = You were previously hitting her, No = You’re still hitting her.
Define an objective question
A true answer exists and the participant could give an accurate answer, but the question is made hard by how it’s delivered.
E.g. “How many times last year did you see a Dr?”
Who would remember specifics that long ago?
Define a subjective question
Asks silly personal perceptions that can’t really be answered..
E.g. “how well did you understand the medical advice you were given?”
There are many possible answers depending on the person.
Define an objective question
There’s a verifiably true answer
Define a subjective question
Based on perspective of the respondent
What is a Dichotomous question?
-Closed ended
-only two response options e.g. yes, no or male, female etc
-
What is a Multichotomus question?
- closed ended
- when more than two answers can be chosen
What is a ‘the list’/ multiple response question?
- closed ended
- when a list of answers to choose from is provided
What is a ranking question?
-closed ended
-one which helps measure the importance of several items.
E.g. “What is most important to you: 1. 2. 3. etc”
-ordinal data
What is a verbal frequency scale?
- closed ended
e. g. “Over the past month, how often have you cried?”
1. All the time 2. Often 3. Occasionally 4. never etc
What is a likert scale?
-closed ended
Measures strength of feeling perception.
E.g. How much do you agree with bla?
1. agree 2. neutral 3. agree etc
What is a graphical rating scale?
- closed ended When a cross can be marked along a scale. E.g. "how much did you like the movie?" didn't-----------------------------did -common online
What is a semantic differential?
-closed ended When a tic/across can be placed along each row to show your opinion. E.g. bad - good strong - weak unattractive - attractive
What is a non-verbal (idiographic) scale?
-closed ended
e.g. point to the face that shows how you feel about studying.
:) :/ :( >:(
What is Nominal data?
people or events are sorted into unordered categories. (“Are you male or female?”)
What is Ordinal data?
people or events are ordered or placed in ordered categories along a single dimension. (“How would you rate your health - very good, good, fair, or poor?”)
What is Interval data?
numbers are attached that provide meaningful information about the distance between ordered stimuli or classes. e.g. 1st, 2nd, 3rd
What is Ratio data?
numbers are assigned that have absolute 0 meaning such as a count or measurement by an objective, physical scale such as distance, weight, or pressure. (“How old were you on your last birthday?”)
Define ‘Target Population’
The people who I wish to study/generalise
e.g. Catholics living in Aus
Define ‘Sampling Frame’
Who has the chance at being selected?
Define a ‘Sample’
The people who get selected for the study
Define ‘Representatives’
Is your chosen sample a good indicator of the target population?
List the types of probability sampling
- Simple random - out of a hat
- Systematic random - e.g every 10th person
- Stratified random - e.g. Every 10th person from every grade
List the types of non-probability sampling
- Convenience
- Purposive - purposely targeting specific people - e.g. people who dropped out of uni
- Snowball - People not easily identified, contacted by referral e.g. drug dealers
