Designing investigations


Considerations when designing investigations

  • Time and cost to set up and carry out investigation 
  • Ethical issues- you must respect people's dignity and rights
  • Confidentiality- it is important to keep data secure and confidential 
  • Convenience of getting data locally 
  • Identifying the population and method to collect sample data 
  • Planning to gain response than you think you need, in case of non- response 
  • Doing a pilot survey to help work out likely responses to sensitive questions 
  • Planning what to do with anamolous results 
1 of 3

Planning for non- response when designing an inves

  • Decide on the number of responses you need to do a valid analysis of data
  • Do a pilot survey to work out the proportion of surveys which are likely to be returned 
  • Use this proportion to work out how many surrveys to send 
2 of 3

Calculating the number of surveys

For a survey, 300 responses are needed.

In the pilot survey, 50 questionnaires are sent out. 40 responses are recieved.

The proportion of responses recieved  is 40/50 =4/5 

The total number to be sent out can be called 'x', so to recieve 300 responses to the full survey, you need: 

4/5 = 300/x, so 'x' = 5 x 300/4 

                              = 375 surveys 

You need to send out 375 surveys.

3 of 3


No comments have yet been made

Similar Statistics resources:

See all Statistics resources »See all Collecting data resources »