Questionnaires

Closed-ended questions

• Also called forced choice

• Respondents choose from a limited range of possible answers pre-decided by the researcher

  • E.g. yes/no/don’t know

Open-ended questions

• Respondents able to give own answer in own words

• No pre-decided choices

3 practical positives

1. Quick

• Can gather data from a large sample of people, and with a good geographical spread if postal or online

2. Cheap

• Don’t need staff to conduct them

3. Quantifiable 

• Especially when using close-ended questions as can be easily processed by a computer and turned into a graph etc.

Quick - example

Connor and Dewson (2001)

• Sent out 4,000 questionnaires to 14 HE instutitons across the country

5 practical negatives

1. Data can be limited and superficial

• Longer questionnaires are less likely to to be completed, so short ones may be better for ease and speed but not depth of data

2. Can be costly

• Depending on the group/topic/length of questionnaire, incentives may be needed

3. May not be completed by the intended recipient

• It may not be recieved, or completed by the intended recepient and there is no way to prove this either way

4. Low response rate

• Despite their ability to collect large amounts of data, this may not happen, especially w.postal questionnaires

• Follow-ups/hand collection increase the cost and time

• Could also be due to faulty questionnaire design or divisions in class

5. Inflexibility

• Once sent out, researchers are stuck with their questions and cannot change them to explore new areas of interest

Practical problems - lying/forgetting/right answerism

• People may try to please/second-guess the researcher

• They may put a purposefully ‘respectable’ answer even if they know this is not true

• Researchers cannot see for themselves what the subjects say they do vs what the actually do so cannot prove this right or wrong

Low response rate - example

• Hite (1991)

  • Sent out 100,000 questionnaires

  • Only 4.5% returned

2 aspects of questionnaires that make them reliable

1. Postal/online - no researcher present

• Means they cannot influence responses and impose their meaning on the answers

2. Questionnaires can be identical in questions, order of questions (and answer choice)

• This means they can be repeated by any researcher

3 aspects of questionnaires that positivists like

1. Detatchment

• Questionnaires are completed at a distance from the researcher, espcially postal questionnaires

• This makes them objective as the researcher’s meaning is not imposed on the person doing it

2. Representativeness

• Large sample size = more representative of wider population

• Can make accurate generalisations if they pay attention to getting a representative sample

3. Hypothesis testing

• Can identify cause and effect relationships

1 aspect of questionnaires that positivists don’t like

1. Open-ended questions

• Open-ended questions can mean non-identical answers get pushed together when creating quantitative data

Invalidity - aspect of questionnaires that interpretivists don’t like

1. Snapshots

• Only give a picture of social reality at the moment the respondent answers the questions

• Don’t capture changes in attitudes and behaviour

2. Lack of researcher/detatchment

• Questionnaires fail as they’re too detached

• There is no way to clarify what the questions mean to the respondent or deal with misunderstandings

• Also no way of knowing if the researcher and respondent interpret the question in the same way

3. Imposition of researcher’s meaning

• By choosing what to ask, the researcher has already decided what’s important to them

4. Types of questions

• Closed-ended questions prevent respondents from discussing what is actually important to them

Validity - Cicourel (1968)

• We have to share the subject’s meaning and see the world as they do

Validity - Shipman (1997)

• Researchers themselves ‘prune and bend the data’

  • This means they distort respondent’s meanings and undermine the validity of the data

Ethical positives

• No obligation to answer sensitive/invasive questions

  • Still need an ICF and to ensure anonymity

Examples - Willmott and Young - explanation

• Studied who did what in households

Examples - Willmott and Young - strengths and limitations

• Strengths:

  • Clear statistical data

• Limitations:

  • Analysis not so good

    • Any ‘help’ with trad. fem. tasks seen as shared role, even if once a month

Examples - Townsend - explanation

• Developed index to determine % of population that were poor

Examples - Townsend - strengths and limitations

• Strengths:

  • Revealed levels of poverty hidden in offical stats

• Limitations

  • Not valid

    • Said that not eating meat = poverty

      • Does vegetarianism not exist…?

Examples - Kelsall - explanation

• Postal survey to large numbers of grads in years after they finished uni

Examples - Kelsall - strengths and limitations

• Strengths:

  • Accessed a large sample

• Limitations:

  • Those who chose not to participate may have been the most interesting

Types of questionnaires - explanation, strength and limitation - postal/mailed

• Posted or emailed to a participant

• Strengths:

  • Distribute quickly and effectively to a large sample

• Limitations:

  • Low response rate

Types of questionnaires - explanation, strength and limitation - self-completion

• Subject fills in the questionnaire themself

• Strengths:

  • Saves researcher time

• Limitations:

  • Cannot clarify question meaning

Types of questionnaires - explanation, strength and limitation - researcher present

• Researcher present during completion of the questionnaire

• Strengths:

  • Can clarify question meanings and ensure all filled out okay

• Limitations:

  • Stupidly time-consuming for both parties

Limitations of website questionnaires

• Unrepresentative!

  • Website readership is often a specific demographic - political bias, socioeconomic

  • Not everyone can access the internet (class)

  • Not everyone has the time and energy to complete it (full/part time workers, disabled people)