Field Period

The field period of a survey is the time frame during which the survey instrument is “n the field,” as opposed to the time when the survey instrument is under development or review in the office. It is the period during which interviews are conducted and data are collected for a particular survey. Originally, it referred to the period of time when personal face-to-face interviews are being conducted by “field interviewers.” Over the course of years, the field period has come to be regarded as the period of days or months over which data for a survey were gathered from respondents, regardless of the mode of data collection that was used.

The purpose of the survey is directly related to the field period that is established. A field period might be as short as a few hours for an overnight public opinion poll or a few days for time- or event-sensitive surveys. For surveys in which the subject is less time-or event-sensitive, the field period might extend for several weeks or months. In establishing the field period for a survey, the purpose of the survey is perhaps the most significant factor. To the extent that a survey is designed to gauge public opinion in response to a specific event or activity, a short field period is appropriate. This is often the case in political [Page 279]polling such as voter preference. It might also be appropriate in those surveys designed to gauge health risks and behavior, such as the impact of flu shots on influenza.

In addition to the purpose of the survey, another consideration is what events are taking place during the proposed field period. For example, surveys of high school students may be more difficult during certain times of the year, such as summer vacation and holiday breaks. Buying patterns may be quite different during the weeks immediately preceding the beginning of a school year than they are at the beginning of summer vacation. Holidays and traditional vacation periods may make contacting potential respondents more difficult and result in additional costs to reach the target population.

Administering surveys with a short field period can be more costly, in part because of the number of attempts that must be made for each completion. For computer-aided telephone interviews, this cost is related to the number of calls that must be made to reach the numbers of completions required. A short field period might also require the use of a large number of interviewers to ensure that sufficient call attempts are made. For mail surveys, there may be additional costs associated with either pre-notification or follow-up mailing to encourage completion of the survey. Shorter field periods may also increase the cost of data processing (such as coding and data transformation) needed to meet the turnaround time required for reports or public release of the results.

Because shorter field period surveys make it difficult to make multiple attempts to those households that are not contacted on the first attempt, there is a greater potential for sample bias. Statistical weighting can be used to overcome some of this bias. Longer field periods can allow more attempts to be made to those difficult-to-reach households and reduce the potential nonresponse bias that may result.

...