Field team supervision and monitoring
The Team leader provides overall quality control for the team by monitoring the performance of each member and ensuring that he or she is doing the job according to protocol and within expected timelines. The Team leader guides the team members respectfully and helps solve problems when they are not performing adequately. Roles for team members are clearly defined in their terms of reference, performance standards are set by the Survey coordinator, and expectations should be clearly communicated during training. The Cluster control form can be used to monitor the tasks of each individual (such as the number of interviews completed or blood draws conducted) and can help the Team leader assess and monitor individual and team progress.
As described in Module 8, Survey supervision and personnel, supervision should take place at multiple levels. Field visits by major stakeholders should be encouraged. They can help motivate teams to maintain a high quality of data collection despite potentially challenging conditions, and reinforce the importance of their work. These visits should be made mainly by people who participated in developing the survey protocol or in the training. Members of the Technical committee are likely to be able to detect and resolve any issues with survey implementation. Anyone making survey site visits should give feedback to the Survey coordinator. Where issues are identified, the Survey coordinator can then ensure that all teams are aware of the issues and that uniform corrections can be made as needed. The following paragraphs describe specific tasks for supervising and monitoring interviewers, phlebotomists and field laboratory technicians.
Supervising and monitoring interviewers
Supervising and monitoring interviewers require a number of tasks: observing each interviewer regularly, re-conducting interviews with participants when needed and reviewing the quality of all data collected. These actions are critical throughout the training, the pilot test, the beginning of the survey implementation, and during data collection. In the beginning of the data collection period, interviewers are still learning how best to collect data and conduct their interviews, and may need extra support to ensure a high level of quality. At the end of the survey, data quality risks decreasing due to fatigue.
When observing interviews, the Team leader should sit close to the interviewer without disturbing the participant. The Team leader needs to assess whether the interviewer is interpreting the responses correctly, recording the correct responses, following skip patterns and following the recommended rules of conduct. The interview should not be interrupted unless there is a serious mistake or a problem with the equipment. Feedback, both positive and negative, should be provided to the interviewer in private, away from the interview setting and after the interview is over.
Spot-checks are useful to ensure that interviewers actually interviewed all selected members in a household. Following up and re-interviewing appropriate household members by asking certain critical questions (such as the number of individuals in the household, dates of birth, and other questions that could falsely create skip patterns to make the interview time shorter) can help identify instances when interviewers are not interviewing a particular household member selected for interview. Reasons for this error could include accidentally interviewing the wrong household member, entering the wrong age for a household member (thus making that individual ineligible for interview) or making other data entry errors. Re-interviews should be done in approximately 10% of households. If possible, the re-interview should be conducted on the same day or at least while working in the same cluster to ensure that the household members and any visitors to the house are still present. If omissions are found, the Team leader should discuss the mistake with the interviewer, who should then return to conduct interviews with any respondents missed. Errors made by an interviewer should be followed by a verbal or written warning, and the case should be documented and reported up the line of command. In order to prevent additional errors, the Team leader should observe the interviewer more frequently. In cases where the interviewer repeats errors, the Team leader should discuss the situation again with the interviewer and may decide that the errors are grounds for dismissal.
If using paper-based surveys, every questionnaire must be thoroughly checked in the field prior to leaving the community so that any errors can be corrected. This is a long process and may require that a data editor be included in each field team. Small errors can cause significant problems during analysis.
Editing paper questionnaires in the field must be done with a pen of a different colour than was used by the interviewer. Missing responses, unreadable responses, and inconsistent or incorrect responses should not be corrected by the editor. Instead, using the different colour, a question mark should be put next to the item concerned. The page number or question number should be written on the front or back of the questionnaire for ease of finding the problem. The error should be discussed with the interviewer, and feedback (both positive and negative) provided. Each error should be brought to the attention of the Team leader, who will decide how to proceed. It may be that the interviewer can make a correction immediately without contacting the respondent, otherwise, the interviewer needs to return to the household to obtain the correct data. Errors that come from multiple interviewers need to be addressed by the whole team. The interviewer, field data editor or Team leader should never make up an answer.
Phlebotomists and field laboratory technicians
The Laboratory supervisor (or Team leader, if there is no Laboratory supervisor) needs to observe the performance of all phlebotomists and laboratory technicians to ensure that they are following the protocol and standard operating procedures. The most critical times to observe phlebotomy and to provide quality control on specific tasks is during the training and the pilot test, as well as at the beginning of implementation. These tasks, known as quality control points, include labelling correctly, taking appropriate time to draw blood, demonstrating the ability to collect blood, collecting the correct amount of blood, properly maintaining the cold chain, and storing and transporting blood and other specimens correctly. If the number of specimens is not consistent with the number of household members who were eligible for specimen collection, the reasons for the inconsistencies need to be verified.
Accidental errors should be corrected. Purposeful errors are more difficult to identify, but if the same phlebotomist has a higher frequency of problems with specimen collection or field readings, it could indicate a problem with their performance. Spot-checking and observing work in the field are the best ways to detect mistakes.