Top Banner
I. What is a survey?
41

I. What is a survey?

Feb 24, 2016

Download

Documents

xylia

I. What is a survey?. What is a survey?. A tool to collect information from or about people to describe, compare, or explain their knowlege, feelings, values, and behaviour . Source: F i nk , A. (2009 ) , How to conduct surveys, SAGE. What is a survey?. - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: I.  What is a survey?

I. What is a survey?

Page 2: I.  What is a survey?

What is a survey?

A tool to collect information from or about people to describe, compare, or explain their knowlege, feelings, values, and behaviour .

Source: Fink, A. (2009), How to conduct surveys, SAGE.

Page 3: I.  What is a survey?

What is a survey?

Mostly a quantitative research tool ….

• Questionnaires sent by post

• Online questionnaires

• Face-to-face structured interviews

• Telephone interviews

Page 4: I.  What is a survey?

Doing a survey – main stepsDesign survey

process

Develop questions

Test questions/tool and train

interviewers

Collect data

Analyze data

Page 5: I.  What is a survey?

II. Survey design

Page 6: I.  What is a survey?

Target group• Who do I want to interview? (who should be in and who out)

• Do I want to interview the whole target population or its sample?

Tracer studyTarget population: graduates

Sample: all graduates (of selected fields) or their sample

Page 7: I.  What is a survey?

Timing • When is the best time to contact your target population?

• How much time is needed to collect responses?

Tracer study

Timing: at least 6-9 months after graduation

Page 8: I.  What is a survey?

Urban Rural Urban Rural Urban Rural1 month 6 months 12 months

12 717 13

1913

18

13

26

17

28

22

26 49

24

39

2339

3525 23 21 20 14

Other (military/parental leave/studying/disabled)UnemployedTraining/probationWorking for familyEmigrationUnregistered employeeEntrepreneurRegistered employee

Source: The ETF Transition Study in the Kyrgyz Republic 2011/2012

Page 9: I.  What is a survey?

Timing (Repeated) Cross-sectional design

Panel design

- population of graduates asked only once

- population of graduates asked repeatedly at different time points

Page 10: I.  What is a survey?

Method of data collection

Self-administered interviews (e.g. web-based, paper-pencil )

Face-to-face interviews

Telephone interviews

Page 11: I.  What is a survey?

Face-to-face interviews

• High response rate

• Possibility to clarify ambiguous answers/provide follow-up information

- More time needed

- More resources needed (most expensive)

Page 12: I.  What is a survey?

Telephone interviews

• High response rate, but lower than in case of face-to-face

- Possibility of sample bias (not everybody has a phone)

• Less time consuming (in comparison with face-to-face interviews

• Less expensive (in comparison with face-to-face interviews)

Page 13: I.  What is a survey?

Self-administered interviews

• Saves money and time

- Lower response rate

- Possible sample bias (web-based interviews)

• People tend to be more truthful while responding

Page 14: I.  What is a survey?

Doing a survey – testing

Once a questionnaire is ready, it must be tested in the field!

- Do people understand to the questions?

- Do people complete the questionnaire?

- Do they often drop-out? - How long does it take to fill the questionnaire? - Can interviewers administer the questionnaire properly?

- Are all procedures standardised?

Page 15: I.  What is a survey?

Doing a survey – testing- Try to anticipate the actual circumstances in which the survey will be conducted

- Choose respondents similar to the ones who will eventually complete the survey

- Enlist as many people in the trial as reasonable, considering also your resources - Focus on the clarity of the questions and the general format of the survey

Page 16: I.  What is a survey?

Doing a survey – things to consider

• Reliability

• Costs

• Usefulness and credibility of results

• Comparability

• Validity

Page 17: I.  What is a survey?

Doing a survey – things to consider

• Reliability

A reliable survey provides consistent information (i.e. repeated measurement yields the same results).

…… make sure that each question means the same thing to everyone.

Page 18: I.  What is a survey?

Doing a survey – things to consider

• Validity

A valid survey provides accurate information.

Do our questions measure what they are supposed to measure?

How were the respondents selected?

Can we generalise our results? …………

Page 19: I.  What is a survey?

III. How to increase a response rate?

Page 20: I.  What is a survey?

What is a response rate?

The proportion of people who return the questionnaire out of the totalnumber of people to whom a questionnaire was distributed.

Example: If you distribute 100 questionnaires and you get back 80 of them,your response rate is 80%.

Page 21: I.  What is a survey?

Response rate matters!

• If we get responses only from a small share of respondents →

- results can be biased and not representative

- results are not significant and useful

Page 22: I.  What is a survey?

What to do?

1. Study preparation

Think about study feasibility!

Think how to reach your respondents/graduates!

- Do you know their addresses? Or their parents’ addresses?

- Do you know their telephone numbers?

- Can you contact them via internet?

- Do you need to use a snowball technique?

Page 23: I.  What is a survey?

What to do?

2. Designing a questionnaire

- Add a pre-notification letter before contacting the respondents

- Add a cover letter while distributing the questionnaires

Personalised requests and communication are important!

- Make sure that respondents see the value of the survey

Page 24: I.  What is a survey?

What to do?

2. Designing a questionnaire

- Limit the number of questions to those which are relevant

- Add clear instructions

- Use clear and simple language

- Add ‘thank you’ at the end of the questionnaire

Page 25: I.  What is a survey?

What to do?

3. Fieldwork

Reminders!

e.g. postal survey – send reminders after 2 and 5 weeks

e.g. face-to-face/telephone interviews – contact respondents at least 3 times at different hour/day

- Contact people several times- Time all contacts- Use different strategies

Page 26: I.  What is a survey?

Follow up of respondents is an important part of the survey fieldwork and can influence the quality of your results!

Page 27: I.  What is a survey?

IV. How to ask a good question?

Page 28: I.  What is a survey?

Doing a survey – preparing a questionnaire• What information is needed?

• Can such information be obtained?

• Format of questions: open vs. closed, scales etc.

• Length of the questionnaire

• The order of questions

• Piloting

Page 29: I.  What is a survey?

Questionnaire – basic rules

• Introduction – provide a context of the survey

• Assure anonymity

• Simple format

• Clean and consistent lay-out

Page 30: I.  What is a survey?

Questionnaire – questions order

• Begin with questions reflecting the subject of the study

• Ask people to recall things in a chronological order

• Group items that are of a similar topic

• Personal and demographic questions should be placed at the end

Page 31: I.  What is a survey?

Questionnaire – length

• Too long questionnaire – negative effect on a response rate

• Too short questionnaire – limited possibilities to respond our objectives

Be efficient and ask questions that are relevant and needed!

Page 32: I.  What is a survey?

Doing a survey – format of questions

Closed questionsExample: How would you rate your overall career perspectives over the next years?1. Very good2. Good3. Poor 4. Very poor

Open questionsExample: What are your overall career perspectives over the next years?------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Page 33: I.  What is a survey?

Doing a survey – format of questions

Scales

1 2 3 4

Students always like to go to the school

Students are interested in the field of study

Students are satisfied with teaching and learning services

Students are present during the entire school day

Attendance to the school is regular during the school year

Thinking about students' motivation, please rate the following statements:(Likert) 1=strongly agree, 2=agree, 3=disagree, 4=strongly disagree

Page 34: I.  What is a survey?

What is a good question?

• Avoid double negative questions

• Avoid asking two things in one question

• A question should have a meaning

• A question should be concrete

• A standard language should be used

• Avoid biased words and phrases

Page 35: I.  What is a survey?

What is a good question?

• How bad is your teacher?

• How would you rate the quality of your teachers and teaching materials?

• Would you say that the students are not motivated to study?

Not concrete and biased

Two questions in one

Double negative question

Page 36: I.  What is a survey?

What is a good question?

• Please, specify the expenditure on ICT:

• How teachers and students evaluate the quality of food provided at school?

• Do you think that math teachers are better than those teaching languages?

Abbrevition used – not clear

Two questions in one

Biased question

Page 37: I.  What is a survey?

Doing a survey – scalesNumber of response categories depends on the survey purposes (usually between 3 to 5).

Example: Do you agree or disagree with the statement that students always like to go to school?

1=agree 2=neither agree nor disagree

3=disagree

1=strongly agree 2=agree 3=disagree 4=strongly disagree

1=strongly agree 2=agree 3=neither agree nor disagree

4=disagree 5=strongly disagree

Page 38: I.  What is a survey?

Doing a survey – scales

Number of response categories must be balanced.

Example: Do you agree or disagree with the statement that students always like to go to school?

1=strongly agree, 2=agree, 3=disagree, 4=strongly disagree

1=strongly agree, 2=agree, 3=disagree

Page 39: I.  What is a survey?

Doing a survey – closed questions/scales

Number of response categories for each question must be exhaustive.

Example: What is your current marital status?

1. Never married 2. Married / Living together

3. Divorced / Separated 3. Other

4. Widowed

Page 40: I.  What is a survey?

Doing a survey – closed questions/scales

Always specify whether one answer or multiple answers are possible.

Example: What is your current marital status?

1. Never married 2. Married / Living together3. Other

Example: Why did you leave school before graduation?

1. Studies too difficult 2. I wanted to work 3. I needed to work4. I got married5. Other

Page 41: I.  What is a survey?

References

A large part of the presentation was based on - Fink, A. (2009), How to conduct surveys, SAGE.