Objectives, uses and users
Objectives are the purposes for which information is required, stated within the context of the program, research problem or hypotheses that gave rise to the need for information. Uses narrow down and specify more precisely the information needs, for example, by describing what decisions may be made based on the information collected and how such information will support these decisions. Users are the organizations, agencies, groups or individuals expected to use the project deliverables.
The first task in planning a statistical activity is to specify the objectives. A clear statement of objectives guides all subsequent steps and could be revised many times during the survey development. Meeting the objectives might require mounting a new survey, redesigning an existing survey, using existing data products, using administrative records, or a combination of these activities. The information needs as stated in the objectives must justify any response burden that will be generated. The relevance of project deliverables to the targeted user community must also be clearly stated.
The hypotheses to be tested, specific data requirements and use of the data, data quality expectations, budget constraints and expected delivery dates should all be stated in the objectives. The concept, definition, unit of analysis and the target population, which are discussed in the subsequent sections, should also be defined at this stage. This will allow the intended as well as other potential data users to determine if and to what extent the project deliverables meet their needs.
Develop survey objectives and constraints in partnership with important users and stakeholders. Establish and maintain relationships with users of information in the private and public sectors and with the general public to enhance the relevance of the information produced and to improve the marketing of products and services. Among important users are representatives of potential markets, policy makers and agents who require the information for legislated use. Before major designs or redesigns, consider conducting a feasibility study and/or pilot test. Conduct extensive and user-focused consultation routinely so as to identify content options, survey relevancy, to determine the need for a cross-sectional or a longitudinal survey as well as to develop public support for the program when it reaches the data collection stage.
Focus analysis of user needs and data requirements on finding the most cost-effective solutions for both the short and long term in the context of statistical framework. Having the users specify their analytical plan or proposed release tables in advance will help to more clearly define their detailed requirements. Before embarking on the design of a new statistical activity (or redesigning an existing one, longitudinal or cross-sectional survey), analyze currently available statistics in the area in terms of sources, frequency, quality, timeliness, etc. Determine the best trade-off between adequacy of the available statistics to meet the requirements of clients and the cost and time required to undertake a new activity to produce statistics that do not already exist.
Where explicit data quality targets exist, include them in the statement of survey objectives in terms of measurable aspects of quality for the entire population or specific domains. Targets can be set in terms of measures such as sampling error, coverage rates, response rates, and timeliness. With administrative data and derived statistical activities, the data quality will be directly related to the quality of input data sources.
During the planning process, clarify the operational constraints such as time frame, costs, resources and data collection methods. Other issues such as the use of proxy responses, respondent recall and the need to measure seasonal data variation should also be taken into account. This is conducted in phases of increasing detail and exactitude. To start with, estimates are based on broad assumptions of the methodology to be used. These estimates have to be more exact and detailed at each stage of planning. They should be based on historic information when available and updated as the objectives are reviewed.
In determining the extent to which a survey will meet user needs, seek a reasonable trade-off between these needs, the survey objectives and the budget, response burden and privacy considerations. Although the Agency may have little discretion where a legal requirement is in place, in other cases it is worthwhile to formulate alternative methodological approaches, means and modes of data collection, frequencies, geographical detail, etc. with a view to arriving at an optimum solution. It might also be necessary to conduct a pilot survey or feasibility study to find an optimum solution.
Review ongoing statistical activities at regular intervals. Statistical programs need to evolve, adapt and innovate so as to keep pace with the changing demands of the users they serve or demands of new users. The purpose of the activity or its statement of objectives needs to be reviewed periodically to enhance the relevance of the statistical product to user needs and constraints (budget, time frame, resources, etc.), as they evolve or change. Sometimes the redesign overhaul of existing surveys may be desirable to maintain the reliability of key statistical series, especially if sources of information have changed or the way in which they are made available is reengineered or rethought.
Main quality element: relevance
Describe and classify key users of project deliverables.
Describe the needs of key users and their anticipated uses of data products, in terms of analytical plans and release tables. Address any gaps between needs and deliverables.
If changes are made to a survey program where estimates are fitted into a time series, evaluate the impact on the time series of those changes.
Blanc, M., Radermacher, W. and Körner, T. 2001. "Quality and users." International Conference on Quality in Official Statistics 2001.Session 15.1. Stockholm, Sweden.
Brackstone, G.J. 1993. "Data relevance: keeping pace with user needs." Journal of Official Statistics. Vol. 9, no. 1, p. 49-56.
Statistics Canada. 2003. Survey methods and practices. Statistics Canada Catalogue no. 12-587 XPE. Ottawa, Ontario. 396 p.
- Date modified: