Research Resources

Field team management

Summary

This section discusses the logistics of managing surveyors and survey teams. Depending on the field team size, you may need to set up processes to manage upwards of 50 field staff. Thinking ahead about how they will be paid, how they should communicate with their supervisors, and protocols for safety in the field will help ensure the survey runs smoothly and without delays.

Surveyor management

For effective management of survey teams, it is essential to have clear reporting structures and regular communication. Surveyors should be divided into smaller teams (see surveyor hiring and training for more information), and report directly to a supervisor. Supervisors, in turn, can report to a field manager, project associate, or research associate, depending on the team structure. It is recommended to have daily debrief sessions with the entire team, at least during the initial few weeks of the launch of the survey. After that, you can reduce the frequency, based on the progress and the team’s adherence to protocols. While your surveyors are generally expected to follow the team’s hierarchy, they should also be encouraged to reach out to the project associate (PA)/research associate (RA) if they face any harassment or unacceptable behavior from other staff. Some best practices for effective management of your team include:

Weekly team debrief sessions/refresher trainings

These can be led by the research associate or project associate, but it’s necessary to maintain this regularity so that any red flags/challenges to smooth implementation of data collection are addressed as and when they arise. Some useful topics to include in debrief sessions are:

  • Challenge resolution to learn about specific challenges faced by surveyors and explain steps to resolve them. This also gives surveyors a chance to discuss and learn from each other’s survey tactics.
  • Error resolution to discuss common mistakes and strategies to address them.
  • Form and batch management to assign additional lists of respondents to surveyors and go over updated survey forms if changes were made to the previous version.
  • Strategies to increase productivity and time management.
     

Performance evaluations of field team members

It is a good idea to provide individual feedback on team members once a month, or some regular frequency. Ideally, an initial round of feedback is given early on (e.g., after the first week of surveying) to catch any errors and problems and make sure the data quality is high going forward. This should be based on back checks and high-frequency checks that compare survey responses, duration, and other parameters across surveyors or across teams. Well-conducted one-on-one feedback goes a long way in boosting team morale and is generally widely appreciated by the staff. 

  • Performance-related incentives: While it can sometimes be a good idea to provide monetary incentives based on performance, these should not be based on quantitative indicators such as the number of surveys completed, as doing so introduces the risk of surveyors compromising on data quality just to earn these incentives. Qualitative indicators are an option as long as they are clearly defined. Incentives must also be in line with the policies of the HR department of the hiring institution. 

  • Another way of providing less formal incentives can be to give "awards" at the weekly debrief that recognize different strengths or acknowledge great performance in front of the surveyor group. Another option is to provide thank-you cards or small gifts (e.g., candy or phone credit). If done correctly, these less formal incentives  can incentivize good performance. Be sure the gifts are appropriate for the context and have value for all surveyors who receive them.

Office hours

It is a good idea to hold sessions where enumerators can communicate with a senior field team member (e.g., the research assistant, field monitor, supervisor, etc.) a few times during the data collection activity. These are highly effective to identify and address issues within the team, as members are often more comfortable with one-on-one discussions rather than in a group.

Field team structures are also covered extensively under surveyor hiring and training.

Communications with the field team

Good communication protocols are necessary for a successful survey and for keeping field teams happy. Time should be taken at the beginning of the project to decide how the research team will communicate with the field team. WhatsApp groups or other group chat platforms are usually the most appropriate for communication between surveyors and research staff.1 Note that no confidential survey-related information should be shared via WhatsApp. Most surveyors will have access to these platforms, which will make communicating with the team(s) much easier. Strict rules about what can be posted on these platforms should be communicated with the field team at the start of the project so that only work-related messages are posted. Research teams should also decide on the following:

  • How should surveyors communicate with the research team? For example, if surveyors are not able to come to work, how should that be communicated with the research team, and how far in advance (if it is not an emergency)?
  • How should the research team communicate with field teams?
    • Work schedules: Are these consistent? Is there a new schedule every week? Some projects will have very consistent schedules. whereas others need to be very flexible. This might be communicated at a regular weekly or daily meeting, but if work schedules are not consistent, you need to find another way of communicating this. 
    • How do you communicate changes to the schedule? For example, there may need to be a pause in fieldwork due to outside factors (such as protest action) or due to internal reasons (such as changes that need to be made to the survey, particularly during piloting). If changes to the schedule are likely, this should be communicated upfront so that the correct expectations are set. 
    • Project updates for fieldworkers: It is often advisable to have regular project update sessions with fieldworkers. For example, you can show survey progress and some basic summaries of things like average survey length and discuss any survey questions that are giving surprising results or strange distributions. This both motivates the team, by showing their progress, and lets them see that you are doing checks on the data which will disincentivize cheating. Fieldworkers also appreciate individual feedback so that they can learn about what they are doing well and what they can improve on. At these sessions, fieldworkers can also give feedback on how things are going in the field, such as which questions respondents struggle with or what strategies different surveyors use to get respondents to answer difficult questions.

Field team payments

Managing finances is one of the biggest hidden headaches of keeping a field team moving. How, when, and how much the surveyors and supervisors will be paid should be part of your field plan. A few considerations include:

  • Payment lag: Keep in mind that there is often a lag between when surveyors start work and they get paid. Even if you are providing per diems and transportation allowance, this lag can be a problem if surveyors are required to buy their own transportation fuel or food upfront. Consider providing these per diems and allowances until the surveyors are paid by giving a start-of-project bonus or loan (which can be automatically deducted from their first payment). Failing to plan for this can slow the launch of a survey.
  • Paying in cash vs. direct deposit/mobile money: 
    • Direct deposit/mobile money: Some offices, such as J-PAL South Asia, do not permit cash payments and require that all payments be made through wire transfers into enumerators’ bank accounts. In countries where you can deposit funds into bank accounts or use mobile money transfers, remember that it can take time to withdraw money from a bank or cash out location. To avoid significant delays, you may need to give enough cash in advance to teams who will not have access to cash-out locations in remote areas. There may be additional documentation required if using direct deposit (e.g., a permanent account number (PAN) card is needed for bank transfers in India). 
    • Cash: May be easier to make happen logistically and result in fewer delays but is also a major security risk. Make sure you have procedures in place to get the money safely from the bank and then hold your team accountable so you can disincentivize them from embezzling funds and can catch any money problems early. 
  • For reimbursements:
    • Have a system in place for monitoring and tracking enumerators’ work-related expenditures.
      • Monitoring systems could be a bike logbook, confirmation from the monitor or supervisor for the expenditure reported, or time sheets to determine when transport costs or other field allowances should be reimbursed.
    • Communicate expectations (such as needing receipts) to enumerators. It is also useful to set expectations by clearly communicating the time lag between submitting a reimbursement request and receiving the funds, as well as the documentation requirements for reimbursement, during surveyor training. To learn more about budget management and tracking, please refer to the related research resource.

Safety and security

Although safety of participants is covered by the institutional review board (IRB), it is important to also consider the safety of the survey team, particularly when the research concerns sensitive topics or takes place in unsafe areas. It is important that research teams enforce safety and security protocols to prevent issues as much as possible and so that surveyors are informed of the best course of action in different situations that might cause physical or psychological harm. This section discusses some common risks for surveyors and sample protocols, and more considerations are covered in the "Considerations for other populations portion" of Ethical conduct of randomized evaluations. Some considerations include:

  • Surveyors should be debriefed and supported after any harmful interaction in the field. Have in place protocols for:
    • Physical risks, such as health insurance to cover treatment, sending enumerators to the doctor if they are injured, providing information for medical assistance or medical leave of absence if applicable, etc.
    • Psychological risks, such as a procedure for reporting and escalating instances of harassment. On surveys that involve emotionally difficult, consider sources of emotional support, such as counseling services that surveyors can call on if needed. 
  • Additional safety procedures may be put in place in risky areas (or, for example, if there are concerns about women traveling alone). These could include surveyors always traveling in pairs, being accompanied by field staff or a local resident, or providing separate transportation options to avoid needing to take public transportation at night. 
  • If working in risky areas or on topics that may increase the risk to surveyors, surveyors should be informed of these potential risks prior to starting the job.

Listed below are common safety concerns and example protocols to address them. Most importantly, you should know your context and consult with field staff or a local partner organization as to the most likely safety concerns and ways to mitigate them--for example, a local partner organization may already have protocols in place.

  • Infrastructural conditions: Poor roads and bridges are a particular concern in rural areas, where roads may not be paved and are thus prone to potholes or mudslides during the rainy season. Traveling on these roads can lead to technical difficulties (e.g. breakdowns or getting stuck in the mud) or vehicle accidents. Similarly, roads or bridges may get washed out and become impassable. Suggested protocols include: 
    • Avoid traveling at night as much as possible, as rural roads are also generally not well-lit. This means ending each day’s interviews early enough to get back to base while it is still light out.
    • If possible, try to minimize doing surveys during the rainy season, when roads and bridges are more likely to get washed out. 
    • Check road conditions when determining where to survey on a given day. Build in contingency plans so that you can survey in other areas if a given village is unreachable due to road conditions.
    • Discuss routes and optimize for the safest route. Although some places will only be accessible by a single route, the survey team can optimize for the safest time to travel.
  • Weather conditions: More urgent safety concerns arise from natural disasters such as earthquakes, monsoons, floods, wildfires, heatwaves and so on. Beyond the immediate safety issues, severe weather can cause mudslides, falling debris and electric poles, fires, and more. Conditions may be localized or may be state- or nation-wide. Recommended protocols include:
    • Stop all survey activities immediately if there is a reasonable safety risk.
    • Check national warnings and local forecasts regularly, particularly if survey activities are happening during a high-risk time (e.g., monsoon).
  • Violence, crime, and political instability: Violence and crime at both the local and national level pose a threat to the physical safety and security of the survey team. In addition, political instability, while itself not an immediate threat to physical safety, can also cause physical harm through mob, police, and army violence. Both violence and crime can still occur even in politically stable environments. Recommend protocols include:
    • Read about the current state of the country and also the locality where the survey is going to be conducted.
    • Learn about safety concerns and suggested protocols from local staff and ensure any concerns are communicated immediately to research staff.
    • Reach out to local police enforcement for safety precautions. Discuss a potential partnership with the local police enforcement to ensure the safety of the surveyors during the survey days.
    • Example: A project in South Africa partnered field research assistants with the local government to ensure the safety of the survey team. Although districts that were prone to crime and violence were omitted from the sample, some included districts were geographically close to these omitted districts. The local government signed a contract with the research team agreeing to keep constant patrol in the areas of the survey team during specified days and times. In return, the research team offered payment in the form of in-kind donations or money.
  • Harassment or discrimination: Surveyors can be harassed or discriminated against by survey participants, locals, and sometimes other field staff. This may be more likely to occur when participants are asked sensitive questions that they did not anticipate, or when a surveyor is seen vulnerable. In some cases, participants and locals make crude comments towards female surveyors, especially when the surveyor is alone. Recommended protocols include:
    • Pay particular attention to vulnerable groups, or those that are more likely to face harassment. Consider additional precautions for surveyors who are more vulnerable, for example by ensuring female surveyors do not travel or do interviews alone.
    • If needed, group surveyors into pairs or trios.
    • Stop the interview if the respondent shows signs of agitation. 
    • Ensure that any surveyor who faces harassment or who interviews a respondent who becomes agitated follows up immediately with a supervisor.
    • Avoid attracting too much attention by enforcing surveyors to wear proper attire and a proper ID to signal they work for a legitimate organization.
  • Health hazards and risks: Some health interventions require collecting biological specimens such as blood or recording physical measurements. If this is the case, additional protocols should be taken to protect both participants and enumerators from health hazards. These interventions might include a community health worker drawing blood, giving shots, and prescribing medicine. Extra precautions should be taken when working in places with poor access to medical facilities. Recommended protocols include:

Remote surveying considerations

Remote survey work requires additional considerations and planning to ensure smooth execution of a study. J-PAL has compiled a number of resources to assist with planning, listed in this blog post. See also J-PAL’s Transitioning to CATI checklist and the considerations (such as device management, respondent tracking, and airtime) described in our survey logistics resource. Key changes with respect to field team management relate to surveyor management and monitoring productivity.

Surveyor management
With limited in-person interaction, surveyor supervision and management becomes ever more critical. Team sizes (or ratios of surveyors to supervisors) may need to be adjusted accordingly, and debrief sessions may need to be moved to a virtual format, such as Whatsapp or Google Hangouts. During these sessions, it is especially important to understand and resolve challenges faced by surveyors and discuss any quality-related concerns, including ensuring that surveyors are using the correct version of the questionnaire and consent form. 

In addition, daily phone check-ins with a subset of surveyors can be helpful in sending the message that data quality and productivity are being closely monitored. This also provides surveyors with the chance to privately convey questions and concerns to supervisors. It can be useful to determine the subset of surveyors for check-ins based on recent productivity, data quality, and hours worked. 

Monitoring productivity
Be sure to carefully track surveyor productivity and survey completion rates, in particular by monitoring the following parameters:

  • Call status2 and status of survey forms uploaded to the server daily. As an additional check, match the total number of recordings per surveyor with the total number of forms uploaded (assuming each call recording corresponds to a new form)
  • Frequency of various reasons for an unsuccessful call (inability to reach the respondent or refusal of the respondent to participate in the study) to identify any recurring patterns. Such patterns may suggest problems with the surveyor’s demeanor on the call or delivery of the consent script.
    • Also monitor the call status of unsuccessful calls to ensure valid and unreachable numbers are attempted again.
  • Number of surveys completed daily. Monitor successful calls for survey quality, described further in the data quality checks resource
  • Number of phone calls made and HHs reached daily, which provides the number of total attempts being made each day. This in turn i) provides surveyors’ conversion rate of calls made to surveys completed and ii) validates surveyors’ effort
  • Track call timings on SurveyCTO or the data collection platform being used to monitor each surveyor’s working hours
    • Similarly, check the time variation between multiple call attempts made to the same number to ensure surveyors are not just finishing call attempts for a particular number in a narrow time frame

Last updated March 2021.

These resources are a collaborative effort. If you notice a bug or have a suggestion for additional content, please fill out this form.

Acknowledgments

We thank Ben Morse and Caroline Tangoren for helpful comments. Any errors are our own.

1.
For communication between principal investigators (PIs) and research associates/managers, it is preferable to use only one “permanent” channel (typically email) for important communications such as task delegation, and one instant messaging channel, e.g. Slack. Avoid WhatsApp or text messaging linked to personal phones to retain a project record when staff leave the team.
2.
Note that phone surveys may also be interrupted due to running out of airtime or battery, the respondent getting interrupted, etc. As a result, it’s important to plan for possible incomplete surveys by including a question indicating the respondent’s willingness to continue the survey at the start of each section.
    Additional Resources
    Safety and security resources
    1. J-PAL Africa's Letter to community watch groups (J-PAL internal resource)

    2. J-PAL Africa's Security protocols (J-PAL internal resource)

    Health-related data collection resources
    1. Field Trials of Health Interventions: a toolbox, Chapter 16: Field organization and ensuring data of high quality

    2. J-PAL 350x: Measuring Health Outcomes Enumerator training video

    3. J-PAL 350x: Measuring Health Outcomes Infection control procedures video

    4. J-PAL 350x: Measuring Health Outcomes Safety and hygiene video

    Awatramani, Vipin, 2019. Field Team Management. Delivered in J-PAL South Asia's 2019 Research Staff Training. (J-PAL internal resource)

    Goldstein, Markus. "Some tips on doing impact evaluations in conflict-affected areas" World Bank Development Impact (blog), May 18, 2016. https://blogs.worldbank.org/impactevaluations/some-tips-doing-impact-evaluations-conflict-affected-areas. Last accessed July 15, 2020.

    In this resource