Flashcards in Surveys And Questionnaire Construction Deck (22)
What are the hallmarks of surveys?
- Primarily focused on breadth not depth of phenomena
- Conducted through questionnaires
What kinds of questionnaires are there?
- Open ended questions
- Exploratory research
- Closed ended questions
- Causal, correlational, or descriptive research
- To operationalise a theoretical framework
What different manners of conducting surveys?
- Face to face
- Self administered
What kinds of survey designs are there?
* Cross sectional (correlational)
* Longitudinal (causal)
* Panel (causal)
What is a cross sectional survey design?
- A study in which various segments of a population are samples an data
are collected at a single moment in time
- Prone to common method variance
What is common method variance (CMV)?
- Variance in a variable that is attributable to the measurement method rather than to the constructs the measures represent
- A source of systematic measurement error
- Common method of measurement of IV and DV can inflate the correlation between them. Can be considered as self-report bias
- Highlight problems of single-source questionnaires
What is a longitudinal survey design?
- A survey of respondents at different times, thus allowing analysis of changes over time
- Temporal separation of IVs from DV
- Tracking study - compare trends and identify changes
- Can help CMV
What is a panel survey design?
- A longitudinal survey of the same sample of individuals or households to record their attitudes, behaviour or purchasing habits over time
- Can help reduce CMV
What types of survey questionnaires are there?
- Open ended
- Ask respondents to answer questions in any way they choose
- Closed-ended questions
- Ask the respondent to make choice among a set of alternatives given by the researcher
What are the components of a questionnaire?
* Cover letter
- General instructions & scale instructions
- Sets of questions sectioned with generic headings
* Demographics of participants at beginning or end
What are the pros of using a questionnaire?
- Empirical data
- Wide and inclusive
- Conducive to quantitative data analysis
- Cost and time
What are the cons of using a questionnaire?
- Detail and depth
- Accuracy and honesty of responses
- Typically yield low response rates which can lead to sampling error
How can measurement error be avoided in survey research?
- Adhere to principles of measurement by using existing validate and reliable scales
* Multi-Source surveys in cross sectional designs to allay CMV
* Increasing response rates
What are the components of total survey error?
* Random sampling error
* Systematic/measurement error (bias)
How does random sampling error relate to surveys?
- Statistical fluctuation that occurs because of the chance variation in elements selected for the sample.
- Can only be minimised by increasing sample size
How does random systematic/measurement error relate to surveys?
- Results form imperfect aspects of research design or from a mistake in the execution of it, and also from repsondents
- Discrepancy between observed score and true scores of a construct
What kinds of random systematic/measurement errors are there in regards to surveys?
What are examples of administrative errors in surveys?
- Unreliable scales
- Construct and content invalid scores
- Inappropriate instrucitons
- Too long means hurried unreliable responses
- Data processing error
- Sample selection error
- Interviewer effects on an examiner
What kinds of respondent errors are there in regard to surveys?
* Non response
* Response Bias
What are non response errors in regards to surveys?
- People who do not participate
- A reflection of sampling design and number of respondents
- Response rates are an accepted index for non-repose error
- Low response rates reflect a self-selection bias
- Over represents extreme positions
- Under-represents indifference
What is a response bias in regards to surveys?
- A bias that occurs when respondents tend to answer questions with a certain slant that consciously or unconsciously misrepresents the truth
- Deliberate falsification or unconscious misrepresentation
- Social desirability bias
- Acquiescence bias
- Extremity bias