This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Find out more here

FEATURE1 May 2006

Measuring up

What does a new ISO mean for the market research industry? Debrah Harding & Barry Ryan report

On March 28, the Market Research Quality Standards Association ( MRQSA ) held a conference to launch the new ISO for market and social research. The new standard, ISO 20252, is due to be formally issued by the International Standards Organisation ( ISO ) later this month.

While ISO 20252 is broadly similar to BS 7911, there are a number of issues that are new.

Executive elements
Translations of questionnaires and related material are required to be carried out by adequately qualified individuals ( mother-tongue competence in the relevant languages and with experience of questionnaire drafting ) and checked by a second person who is also qualified. The client must be offered the opportunity to review the translation and the translation process must be documented. These requirements are more stringent than BS 7911, which simply refers to the need for a review of translations.

Data collection
There are two additional requirements added to the data collection element: the data, duration of the interview and the identity of the interviewer must be recorded, for example on the questionnaire; and for qualitative recruitment respondents' identities must be confirmed. This reflects problems found in some countries with "professional" attendees. How identity is to be confirmed is not specified and so organisations can determine how best to do this.

Data processing and analysis
As BS 7911, the standard requires various details to be covered in client reporting. The following are additional areas although some are covered in different ways in BS 7911.

• That projects are carried out in compliance with ISO 20252.

• Reasons why achieved samples fall short of intended samples.

• If automatic coding is used an estimation of error rates is required.

• Methodological details to be passed to the client ( whether they want it or not! )

• Comments on the reliability of the findings including sampling variance ( for probability samples ) and estimates of non-sampling error. In the case of qualitative research a statement that results cannot be projected to an overall population is required.

There are also a number of elements within the BS 7911 standard that have not been included in ISO 20252:

• There are no specific requirements for supervisor recruitment or training. Supervisors are now regarded as other staff and the more general requirements for training and ensuring compliance apply.

• Basic training for face to face interviews is now 6 hours not 12 although organisational training must be added to the 6 hours for all interviewers and recruiters.

• Appraisals for field based interviewers who conduct more than 5 projects/waves per year must be appraised only once a year ( reduced from twice yearly ) and it is no longer mandatory that this is completed face to face.

MRQSA announced that BS 7911 will be withdrawn from 31st May 2008 ( not May 2007 as originally announced ). Organisations wishing to switch assessment to ISO 20252 as part of their surveillance cycle will be required to complete this and have a new certificate before BS 7911 is withdrawn in May 2008.

May | 2006