Wednesday, January 26, 2011

Topic 4 : Developing a Questionnaire


What is a questionnaire

-          Aim : collect accurate data in a consistent format

Influences on Questionnaire Design

-          Purpose of research
o   Until we have a clear understanding of purpose of research, no point in thinking about questions

-          Who Will be answering the questionnaire
o   Style and complexity of language

-          How Will Questionnaire Be Administered
o   Self completion vs computer based

General Format of Questionnaire

-          logical sequence
o   simple then complex
o   general then specific
o   neutral before personal
o   "flows" from one topic to another
o   Vs omnibus

Basic Structure of Questionnaire

-          Title

-          Name of sponsoring organisation

-          Introduction / purpose & use of survey / confidentiality

-          Screening questions

-          Main questions

-          Verification (respondent contact details)

-          Thank you

-          Authentication : interviewer name etc / detail of interview

What can we measure in a questionnaire

-          Behavior

-          Beliefs – what people think is true

-          Knowledge – accuracy of beliefs

-          Attitudes – what people think is desirable

-          Attributes – respondent's characteristics

Behavioral Questions

-          Did respondent exhibit behavior

-          Frequency

-          Time elapsed since

-          Intended behavior

Measuring Knowledge & Beliefs

Measuring Attitudes

-          What is desirable

-          Conviction

-          Context / seasonality

-          Measured by scale

Recording Attributes

-          Descriptive information

-          Analyse sub-groups

-          Define quotas

-          Determine if sample is representative of population

Response Formats

-          open ended

-          pre-coded

-          uncoded open ended – respondent free to provide any answer

-          precoded open ended – respondent provides answer of choice, but interviewer then codes it

-          pre-coded – respondent has to select choice

o   one answer : discrete / single response question
o   multiple response


-          Open ended

o   Analysed qualitatively as anecdote

o   Quantitatively – percentage who gave certain response  (particularly where number of response options too large to precode)

o   Response options unknown

o   Get general feelings

o   Get reasons for opinions


-          Precode


-          Which format ?

o   Kind of info sought

o   Whether researcher wants to pre-condition respondent

o   Prior knowledge of possible responses

o   Ease of processing info

o   Available resources.

o   Info potentially available from respondents

o   Level of accuracy required

o   Sensitivity of questions

o   Experience of respondents.


Question Wording Rules


-          Set of response options for precoded questions must be exhaustive

-          Response categories to be mutually exclusive

-          Scale should be unidimensional

o   Scale should only measure single concept.



Structuring a Questionnaire


-          Logical order

-          Apart from primary questions

o   Secondary questions – consistency or reliability checks

o   Tertiary questions – breathers after sensitive questions

o   Probes – gather details : what else

o   Direct & indirect questions
§  Direct – about respondent
§  Indirect – info about respondent by asking about other

o   Suggestive – assume respondent has particular view on issue and these questions attempt to confirm that view

o   Filter & contingency questions – select certain groups of respondents


Minimizing Respondent Burden

-          Relevance

-          Language used: conversational / clear

-          Question length : single issue


-          Avoid leading questions


-          Dealing with sensitive issues

o   Casual approach

o   Numbered cards : which one of these statements applies to you

o   Everybody approach

o   Other people approach







               

No comments: