1) decide whether to carry out an in-surgery, post-consutation, or postal survey

2) dip into the GPAQ Manual (very useful) available from the 'downloads page of the Manchester Research Dept website http://www.gpaq.info

3) decide whether you want the analysis to practice or GP level (no decision needed here for a single handed practice)

4) print out the latest version of either the post-consultation or the postal questionnaire (available from http://www.gpaq.info the 'downloads' page). Take care not to mix these versions, nor mix versions 1 and 2.

5) make the appropriate number of copies. A MINIMUM of 50 completed questionnaires is needed per GP for statistical reliability. Lower numbers (per GP) can be collected for larger practices needing only reports for the practice overall. See the Manual for more details. An individual practice can make copies royalty-free.

6) select eligible patients, a representative sample (the Manual is helpful on this). For postal surveys, some practices have identified patients (over 16) who have attended the surgery within the last year, and made a random selection of patients from this list.

7) Do carry out your survey as early as practically possible. October to March is very busy. (If you would like us to analyse the results) let us know your timescale as soon as you are able, and send the completed questionnaires to us as early as you can.

8) For the post consultation survey practices found that:

• (for both surveys) if the questionnaire is to be printed (rather than photocopied), it is better if this is done A3 double sided, and folded. Pages can't get lost, and there are no paperclips or staples involved and it halves the weight of paper.
• it helps to have someone (maybe a student) looking after the survey. Receptionists tend to be very busy, and many questionnaires slip through from patients under 16 and/or had not actually seen the doctor.
• it is useful to have a large box in the surgery for patients to 'post' the questionnaires into. Questionnaires should ideally not be folded, and not be in envelopes. It takes for ever to take them out of envelopes, unfold and stack the questionnaires.
• if possible, questionnaires should be completed in the surgery. Only a proportion of those which go out of the door are returned in time.

9) Although our expertise is numerical, which entails practices keeping (or photocopying) the 'Comments' section, we are piloting analysis and summary of the Comments Section for 2007/8 for questionnaires received across the summer. See also Q&A No 16.