Monday, March 14, 2011

SG: Topic 3: Developing a Questionnaire


 

What can we measure in a questionnaire


-          Types of question

o   Behaviour

o   beliefs

o   Knowledge

o   attitudes

o   attributes

 

 

-          Behavioural Questions


o   do or have done

o   measure

§  whether or not subject exhibited certain type of behaviour

§  frequency

§  time elapsed since behaviour

o   example

§  is subject working (or did work) with pres-school age child

-          Belief

o   what is true or false – rather than on accuracy of beliefs

o   beliefs not necessarily indicator of behaviour

o   example

§  what subject believes to be effect of day care centre on emotional development of pre school age children.

-          knowledge

o   access subject's knowledge of particular facts à establish accuracy of their beliefs

o   example

§  what subject knows about govt programs to assist parents with pre-school children to work part time.

o   market research example

§  product awareness

§  product attribute awareness

§  product price

-          difference between belief and knowledge

o   belief

§  what is the average number of cups of coffee you think you should drink in a day

o   knowledge

§  what is the average number of cups of coffee that an average person would consume in a day

-          attitude

o   what subject thinks is desirable

o   example

§  whether or not mothers with pre school children ought to participate in workforce

-          attribute

o   age / education / occupation

 

 

Principles of Question Design

 

 

-          reliability

-          validity

-          discrimination

-          response rate

-          same meaning for all respondents

-          relevance

 

 

-          Reliability

o   same respondent should answer question same way on different occasions

-          validity

o   question should measure what it is intended to measure

§  example

·         how healthy are you

o   measure of health

o   or

o   optimism / pessimism

-          discrimination

o   sufficient variation allowed in response

-          response rate

o   usually caused by

§  question content

·         sensitive

·         irrelevant

·         repetitive

·         poorly worded

·         difficult to understand

·         insufficient response categories

§  method of administration

§  questionnaire length

-          same meaning for all respondents

-          relevance

o   is question really necessary

 

 

 

Question Wording

 

-          is language simple

-          can questions be shortened

-          is question double barreled

-          is question leading

o   does this seem like a good idea to you

-          is question negative

-          is respondent likely to have necessary knowledge to answer question

-          will words have same meaning for all respondents

-          is there a prestige bias

-          is question ambiguous

-          is frame of reference / context of question clear

o   example

§  how often do you see your mother

§  within the last year, how often would you have seen your mother on average

-          does question contain gratuitous qualifiers

o   ……  even if it leads to a decline in living standards

-          is question dead giveaway

 

 

 

Question Response Format

 

-          closed or forced choice

-          open ended

 

 

-          closed response format

o   advantages

§  responses can guide subjects, eliminating irrelevant answers

§  easier to analyse

§  quicker to process

o   disadvantages

§  response options must be exhaustive à problem with unanticipated answers

§  response options may be too narrow

-          open ended response format

o   uses

§  response options too large to pre code

§  attitudinal info where response options unknown / feedback is required

§  get at general feelings

§  find out reasons for opinions

o   disadvantages

§  more demanding for respondents to answer

§  more time consuming to process

§  responses may not be relevant

§  difficulty to code

o   advantages

§  many possible answers are allowed

§  do not have to anticipate all responses

§  data can be analysed quant and qual

§  clear insight into respondent logic / thinking

 

 

Principles of developing question responses

 

-          principles

o   exhaustiveness / inclusiveness

§  eg, categories for marital status

§  attitude questions

·         include dk or no opinion

§  other category

o   exclusiveness

o   balancing categories

§  same number of alternatives on either side of neutral position

 

 

Numerical Rating Scales

               

-          likert scales

-          horizontal rating scales

o   opposite attitude positions

-          semantic rating scales

o   opposite adjectives

-          vertical rating ladder

 

 

Other Response Formats

 

 

-          checklists – tick all that apply

-          binary choice formats

o   dichotomous

§  do you smoke

o   paired comparisons

-          multiple choice formats

 

 

Non Commital Responses

 

-          no opinion / don't know

-          inclusion of middle alternative

 

 

Response Sets

 

-          tendancy  to respond to question in some characteristic manner regardless of content.

o   social desirability

o   acquiescence

o   non - acquiescence

-          reducing social desirability

o   mention that everyone does it.

o   use an authority

o   build in an excuse

o   ask a less specific question

 

Order of Questions

 

 

-          start with enjoyable questions

o   easily answered

o   factual

o   obviously relevant to objectives of survey.

-          go from

o   easy to difficult

o   easy to difficult

-          open ended at end of survey

-          sections

-          use filter or contingency questions

-          mix up pos / neg questions to avoid acquiescence

-          consider randomizing

-          use variety of formats

-           

 

 

Pilot Testing

 

-          pretesting

o   ask participants how they interpret question meaning

o   rephrase?

o   is range of response alternatives sufficient

o   is question necessary / redundant

o   flow

o   estimated time right ?

o   clear instructions

o   skips simple and clear to follow

o   sufficient space

 

 

No comments: