Questionnaire development

• User consultations
• Evaluating the suggestions
• Questionnaire content and development

User consultations

Planning for the next census begins even before the current census cycle is finished. Through a series of workshops held across Canada in 2002, users provided recommendations for the types of questions they would like to see on the 2006 Census questionnaire. A cross-section of data users were also surveyed by mail and asked to rate the importance of proposed new census questions. The data requirements identified during these consultations were used to develop the content and design of the census questionnaire.

Evaluating the suggestions

Before going for testing, the workshop and submission suggestions had to meet certain criteria before being judged suitable for inclusion in the Census of Agriculture:

• Is this topic of national interest?
• Are data worthwhile at more detailed geographic levels than provincial or national?
• Will farmers easily understand the question?
• Can the question be answered — that is, do the farm operators have the information to answer the question?
• Will farm operators be willing to answer it?
• Will there be a broad demand for the data generated by the question?
• Can the question be answered by either "Yes," "No," or a quantitative response?

Questionnaire content and development

Although the questionnaire is updated every census to reflect users' changing requirements as identified through the Canada-wide workshops, certain basic or core questions appear on every census. These questions — such as those on farm operators, land area, livestock numbers and crop areas — are considered essential by Statistics Canada and other major users of Census of Agriculture data. Repeating basic questions allows the census to measure change over time, while adding new questions and dropping others allows data to be collected that reflect new technologies and structural changes in the agriculture industry. For example, the questions on manure-management practices were tested and improved to reflect operators' use and understanding of farming terms and practices on operations today. Other sections also modified and refined were those on land tenure, organic production, injuries, irrigation, and bees.

New or changed questions were developed in head office and tested a number of times with farm operators across Canada through one-on-one interviews on their farms and in focus groups. Farm operators selected for testing reflected regional diversity — in types of agriculture, production techniques, languages and terminology, and in policies or issues that could affect the sensitivity of questions. This testing proved that some questions would not perform well on the census, and that the wording of other questions would require fine-tuning. Respondent burden, content testing results, user priorities and budgets were all taken into consideration in determining the final content of the 2006 Census of Agriculture questionnaire. It was approved by Cabinet in the spring of 2005.