trng-plan.Rmd

# Planning and preparation

This chapter presents recommendations for survey designers, survey managers and trainers on how to plan and prepare effective and efficient fieldwork training. Fieldwork training should be planned well in advance and must be well prepared. It usually consists of a main interviewer training and separate training for supervisors, data monitors or other fieldworker roles.

Survey designers, please take note of the sub-chapters on [Training size](#training-size), [Trainers](#trainers) and [Timing & duration](#trainingduration), as they are budget and timeline relevant. For survey managers and trainers all sub-chapters are relevant to plan and prepare a solid fieldworker training.

## Training size {#trainingsize}

It is strongly recommended to train more trainees than are ultimately needed as fieldworkers, i.e., as interviewers, supervisors, data monitors or other roles. This allows for a [competency-based selection](#assess-select) of fieldworkers at the end of the training. Training more people than needed also compensates for any attrition during training and builds a reserve of already-trained people that can be brought up to speed and substituted relatively quickly during fieldwork, should the need arise.

For a meaningful selection process train __at least 20% more fieldworkers__ than are required if you are working with experienced teams who you have worked with before and if you are expecting few trainees to drop out during the training. For example, train at least 48 people if you are aiming for 40 fieldworkers. Increase the excess to at least 40% if you are working with unknown trainees, a new implementing partner, in a new context or expect a high number of drop-outs during training. Under these conditions, 56 people should be trained to reach 40 fieldworkers. Training a smaller excess of trainees is not recommended as it prohibits a meaningful fieldworker selection process and increases the risk of having to rely on trainees that are not up to the task.

Often, some trainees will drop out during training, particularly after day 1 or 2 once there is a clearer understanding of the required effort. Mitigate early trainee attrition by keeping on standby some of the applicants that have not been invited to the training, so they can quickly replace any trainees that have not come back. You might have to bring them up to speed in extra hours and ask them to independently catch up on material already covered. It is not advised to bring in additional trainees after day 3, as they will have fallen too far behind.

:::: {.exmpl}
<img src="images/world.svg" class="svg-inline"/> **Real-world example:** Towards the end of the fieldwork training for one survey, trainees staged a walkout and collectively refused to continue training or fieldwork unless their (somewhat unrealistic) demands were met. This put survey management in a very difficult position and the survey itself at risk. Demands could not be met and there was no time or budget to repeat the training with other trainees. Avoid walkouts by setting fair fieldworker terms and conditions, making them clear to trainees and getting their written agreement prior to the training.
:::: 

In many circumstances it is beneficial to select supervisors, data monitors and any other fieldwork roles from the pool of trainees based on their performance and capabilities observed during the training (see [Assessment & selection](#assess-select)). If any of the fieldwork roles are predetermined, the designated persons should participate in the fieldwork training without being counted towards the trainee excess (and required to demonstrate full understanding of the trained material).  

Keep the overall number of trainees as small as possible, __ideally below 50 trainees__, which corresponds to at most 36 fieldworkers. Learning outcomes quickly decrease with increasing class size, as each trainee receives less supervision and personal feedback and as it is more difficult for trainers to keep track of trainees’ performance. 

:::: {.tip}
<img src="images/tip.svg" class="svg-inline"/> **Tip:** Whenever possible, design the fieldwork plan so that fewer interviewers work over a longer period of time, requiring fewer fieldworkers to be trained.
::::

Centralized training, i.e. having all trainees in one location at the same time, is the preferred method of training.  It guarantees that all trainees hear the same training at the same time.  They all hear any questions that may arise, and the explanations to resolve any confusion.

However, if having more than 50 trainees is unavoidable (e.g., surveys in large countries or with a large sample) or centralized training is not possible (e.g., different languages or a geographically dispersed sample), it is best or can be imperative to split fieldworker training into separate rooms or even locations. In those instances, it is paramount to __standardize the training__ as much as possible across different rooms/locations to minimize the training effect, i.e., interviewer behavior being correlated to the training rooms/locations, which is very damaging for the accuracy of a survey. 

This requires training to be guided mainly by standardized and extremely well-prepared training materials, as well as effective communication between the training rooms/locations, so that all trainees benefit from relevant feedback, comments, questions, etc. raised in each room/location. On top of that, there must be sufficient capable trainers who have received a comprehensive Training of Trainers (ToT). The ToT can be a dedicated exercise or integrated as an explicit outcome of the pilot training. Several scenarios are possible if a large number of fieldworkers need to be trained:

1. __Break-out sessions__. This is the first best option to train class sizes of 50-100 trainees if they can be trained centrally. Split the class into groups of manageable size (around 15-20 trainees each) and assign dedicated trainers to each group. Use the break-out sessions to train everything that is  sensitive to class size, including the questionnaire reading and practicing. Trainers must keep notes of any additional clarifications they provided, questions they received or feedback they gave. Hold daily plenary sessions with the entire class in which feedback is given based on the notes from individual sessions. Modules that are less size-sensitive, such as the opening module or fieldwork logistics can be trained in the full class. 

1. __Parallel training__.  If joint plenary sessions are unfeasible due to an unavoidably larger class size, the trainees must be split into groups and trained either in parallel or consecutive training. (As noted above, whenever possible try to avoid this by adjusting the fieldwork plan.) For parallel training, groups are trained separately but at the same time and in the same location (i.e., different rooms of the same building). Each group should have a “resident” trainer, who is responsible for its training. Other trainers can rotate among groups, for example those responsible for training a certain module or task, bringing some consistency to the parallel sessions. Because groups and trainers do not overlap, it is harder to harmonize training across groups. It is therefore important that trainers keep good notes and debrief during the breaks and at the end of the day. 

1. __Consecutive training__. Trainees are split into groups which are trained one after the other, possibly in different locations. A core set of trainers repeats the same training for each group, providing consistency across all groups. Since each group of trainees should start fieldwork immediately after their training, this approach only works when it is possible to stagger the start of fieldwork (e.g., if fieldwork is sufficiently long). Additional trainers should attend the training for each group to supervise the field start of one group while the core trainers move on to train the next one.    

1. __Parallel locations__. It is not recommended to simultaneously train in multiple locations, as it is difficult to harmonize fieldwork training between different locations. If unavoidable, it is essential that extra effort is put into preparation of training materials, that trainers keep good notes, and that they frequently communicate and debrief with trainers from other locations.

## Trainers 

The success of fieldwork training hinges upon having enough and competent trainers.

Conducting fieldwork training is a substantial amount of work! Doing it well - even more so. [Prepare the training well](#trainingpreparation) to avoid significantly reducing quality or overwhelming trainers. Make provisions for enough trainers so they can conduct the training well. For every room, there needs to be __a minimum of two active trainers__ at any point in time: one to lead the training and another to support the class and do any of the supporting activities that happen in parallel, such as updating manuals, compiling feedback, creating practicing scenarios, etc. Larger classes require more trainers because supervision and supporting activities increase. Ideally, there should be one active trainer for every 15-20 trainees.

Trainers must be available to actively engage in the training most of the time. Avoid the situation where trainers are physically present on site but are in meetings instead of the training room, focused on tasks unrelated to the survey, or are otherwise disengaged from training. If trainers have ongoing responsibilities outside of the training (which is often the case), plan to have more than two trainers per room.

Every trainer must have:

- A good understanding of the survey in general.
- A detailed understanding of the questionnaires, definitions, protocols, manual, CAPI, etc. Ideally, they have been involved in the survey design process, participated in the pre-test or pilot and will be involved in managing and monitoring fieldwork. 
- A general understanding of the survey subject, e.g., the regional agricultural systems for an agricultural survey. 
- Good facilitation skills and the ability to lead and manage all components of the training. 

Collectively, the trainers of each room should also have: 

- A good understanding of the indicators' construction, the intended data use or analysis.  
- Subject and context expertise, e.g., know in detail the local circumstances of child education, nutrition and health for a survey on early child development. 
- Experience implementing the survey type and implementing surveys in the context.
- Experience conducting fieldworker training.
- Fluency in the trainees’ language. Do not train in a language the trainees are not fluent in! If some trainers (usually the survey designers or analysts) do not speak the training language, use an interpreter or ask other trainers to translate for them. 

In many instances, particularly in surveys with limited pre-testing and piloting, mismatches between survey design and field reality are uncovered during the training and adjustments need to be made quickly. For example, it may become evident that a locally-practiced exchange labor system cannot be accurately captured in the presented questionnaire. Trainers need the above expertise and experience to be able to identify such issues and make decisions that are analytically correct, in-line with the survey design and make sense within the local context.

:::: { .warn}
<img src="images/warning.svg" class="svg-inline"/> **Warning: **There is a tendency to send junior analysts or PhD students without sufficient survey or context experience to conduct fieldwork training. This can work, as long as other experienced trainers are present and engaged, but quickly leads to quality issues if they are the main training lead and the others become unavailable/inattentive.
::::

Similarly, handing over a sample and questionnaire to a survey firm and letting them implement the training without close oversight or any participation is usually a bad idea. Firms commonly lack the necessary analytical background, are not familiar with the details of the survey design, are differently motivated, and/or might not have the capacity to conduct the training to the required standard. It is recommended to send your own trainer(s) to be present at the training, especially if working with a new firm. They can support as needed and ensure the training meets the quality standards described here.

Subject experts are sometimes invited to facilitate training on specific modules. Unfortunately, they often have their own priorities and varying training capacity, resulting in overly complicated or confusing sessions (e.g., long theoretical presentations with details that are largely irrelevant for interviewers). If external subject experts will lead parts of your training, make sure they fully understand the relevant parts of the questionnaire and what content and level of detail are most appropriate for trainees. Prior to the training, discuss the scope of their engagement and review their contributions, such as presentations. Another role for subject experts during fieldwork training can be to help answer any questions on their subject and contribute to relevant discussions.

## Content

To equip trainees with all the knowledge and skills necessary to do a successful job as a fieldworker, training must go beyond a simple discussion of the questionnaire and cover the components listed below. See details on how to implement each of them in the [next chapter](#conducting-the-training). 

- __[CAPI use]__. Interviewers must understand and be able to use all functionality features of the CAPI software that will be used in the field, including navigating through the CAPI questionnaire, creating and sending interview files, etc. it is best to keep theory limited and introduce functionality throughout the training at the time it is relevant, so trainees can practice it immediately.

- __[Questionnaire content]__. All questionnaires must be covered in detail, including question text, answer types and options, how to administer questions and sections, the question routing (skip instructions or enabling rules), and the underlying definitions and concepts. 

- __[Interview practice]__. Repeated practicing – in groups, with respondents, on-site and in the field – is crucial for trainees to fully internalize the trained material, to learn how to put it into practice and to have exposure to real respondents and scenarios during the training.
 
- __[Interview techniques]__. Trainees must learn and practice a range of skills that are critical to collecting good data, such as how to introduce yourself and the survey, convincing respondents to participate; how to probe and inquire for clarifications without introducing bias, and how to control the interview (i.e., keep respondents focused on the interview).

- __[Expert measurements]__. Many surveys include exercises such as anthropometrics, plot measurement or capacity testing, which require interviewers to use specialist equipment and strictly adhere to measurement procedures to ensure comparable measurements. Trainees must practice until they can correctly implement the exercise and behavior across the entire field team is standardized.

- __[Pre-interview tasks](#pre-interview-tasks)__. Interviewers need to learn the format in which they will receive their assigned units (e.g., households), any selection protocols they might have to follow, how to locate and confirm the identity of units, when to revisit, replacement protocols, how to manage their workload, etc.

- __[Post-interview tasks](#post-interview-tasks)__. It is important that interviewers know how to correctly perform a range of activities after an interview, including checking the files for completion, addressing inconsistencies, leaving comments, submitting files, receiving and responding to feedback on submitted interviews, making corrections to the files, etc.

- __[Final field test](#final-field-test)__. At the end of the training, a final field test should be conducted to identify and address any outstanding issues with questionnaires or field procedures, expose interviewers to real field conditions and determine if they are ready for field work.

- __[Supervisor training]__: Supervisors must learn how to manage field teams and the relationship to the communities and respondents. They must also understand the tasks they are responsible for in the survey, such as handling money, making logistical arrangements, checking completions, communicating with survey management or data monitors, etc. 

- __[Data monitor training]__: Data monitors need to learn all the tools and background information necessary to be effective in their roles, including how to review interviews, conduct re-interviews, review audio recordings, interpret summarized information, provide effective feedback to fieldworkers, and compile feedback for survey management and archiving. 

## Timing & duration {#trainingduration}

Fieldwork training should be held __just prior to the start of fieldwork__, with a maximum of 1-3 days between training and fieldwork to allow for administrative matters, a rest day and travel to the field. An immediate fieldwork start with strong supervision and feedback in the first few days is the best way to ensure the survey is being implemented as trained and that fieldworker behavior is streamlined. 

The longer the break, the higher the risk of interviewers forgetting details, making mistakes and developing idiosyncratic behavior. If fieldwork is delayed by more than a few days, conduct a short __refresher training__ just prior to the field start to revisit and practice key points of the trained material. For longer delays, a longer refresher training will be necessary. For breaks beyond a few weeks, often complete retraining is the only solution, in part due to the likelihood of having many new trainees.

__Schedule enough time for fieldwork training__. The optimal duration depends on a range of factors, including the nature of the questionnaire, tasks required by fieldworkers, experience level of trainees, etc. As a general rule of thumb, for a  multi-topic, socio-economic survey with a questionnaire of 2-2.5 hours, allow around 15 working days (2.5 weeks) for the main interviewer training, including practicing and final field test. Training duration can vary with survey factors, but keep in mind that many components such as opening session, CAPI use, interviewing skills, and final field test need be trained independently, so even for a survey with a relatively short (45-minutes) questionnaire, at least 6 working days (1 week) of interviewing training should be scheduled. Surveys with longer questionnaires or specialized tasks may require longer fieldworker training.   

To some readers, this might sound excessively long. Many survey firms, especially those with a background in opinion surveys, will suggest much shorter training durations. However, in socioeconomic surveys, due to the substantial amount and variety of [content](#content) that fieldworkers need to learn, it is usually impossible to train field teams to an acceptable level in a shorter amount of time. Do not cut the training short, unless you are very confident that all trainees will have fully internalized all material and will be able to perform well as field workers!   

:::: {.tip}
<img src="images/tip.svg" class="svg-inline"/> **Tip:** Be prepared for possible extensions. In many surveys, trainees are not ready to go to the field at the end of the training and the training needs to be extended for a few days. If extensions have not been budgeted for, this can cause quality issues, e.g., the survey firm having to cut corners elsewhere to compensate for unanticipated training costs. Avoid this by setting out with a realistic training schedule and putting contingencies for potential extensions into the budget and timelines.
:::: 

Avoid fieldworker training that lasts longer than 3 weeks (18 working days). Trainees’ ability to absorb and remember material and general morale decrease with training length. Also, since every day of fieldworker training is quite expensive, it leads to paying for days that are increasingly ineffective while reducing the funds available for other survey activities. Conducting an efficient and yet effective training as described in Chapter \@ref(conducting-the-training) [Conducting the training](#conducting-the-training) should allow most surveys to be trained well in less than 3 weeks.

For surveys with a large number of questionnaires (e.g., school surveys) or special tasks such as anthropometrics or plot measurements that take a substantial amount of time to train (e.g., solid anthropometrics training can take 4-6 days), often it is unrealistic for interviewers to learn all questionnaires and tasks sufficiently well in the time available. Instead, divide questionnaires and tasks between __specialized fieldworker roles__. For example, anthropometrics could be collected by dedicated *measurers* and the main interviews are conducted by *interviewers*. This way, each fieldworker role can focus on their questionnaires/tasks and learn how to implement them well. Furthermore, different fieldworker roles can be trained in separate training sessions, which cuts overall training length and costs. 

__Supervisor training__ should be held outside of the main interviewer training hours, so that supervisor candidates can attend the interviewer training in full. If supervisors are designated prior to training, supervisor training can be held just prior to the main training for a couple of days. If supervisors are selected during the training, it can be held towards the end of the training, either after interviewer training has ended for the day or during the day when modules that are not relevant for supervisors are being covered (e.g., administrative days, review sessions of material they have already mastered, etc.). The duration depends on the scope of the tasks, but generally 2-3 days are sufficient. 

__Data monitor training__ is best held after the final field test so that the data collected in the field test can be used as training and practicing material. Data monitors should attend interviewer training and field testing to have a good understanding of the data collection activities they will be responsible for monitoring during fieldwork. Availability of trainers permitting, it can be held in parallel to the supervisor training, between training and field work, and/or during the first few days of field work. Generally, 2-3 days are sufficient for initial training and practicing with an additional 2-3 days of supervised work during the first days of field work. 

## Schedule {#trainingschedule}

Prepare a realistic training schedule and track training progress on an on-going basis. Keep in mind the following general considerations to plan an effective and efficient fieldwork training: 

1. __Stay flexible__. Create a schedule that identifies all modules and allocates the required time. Use the schedule to track progress, ensure you are not falling behind and to plan your logistics, e.g., on what days you will need the training venue or transport. Expect timings to shift a little during the training, as some topics/activities tend to take longer or shorter than anticipated. Do not force compliance with the original schedule; instead adjust the training schedule to the updated requirements of the training. Stay flexible to allow sufficient time to address issues, retrain topics, or dive into more detail as needed.

1. __Allocate time generously__. It is better to finish training early one day rather than keeping trainees until late, rushing content or having trainees that are not ready for the field at the end. If training starts falling behind schedule, make a decision to extend training as early as possible – before logistical arrangements make it impossible. If training progresses faster than anticipated, use the extra time for additional practicing. Only shorten the training if you are certain that the trainees are fully ready for the field.

1. __Practice frequently__. Do not leave (field) [practice](#praticing) to the end of training. Schedule daily group practices and front-of-class interviews towards the beginning of training to solidify newly learned material. Practice with respondents on-site or in the field during the latter half of the training to prepare trainees for fieldwork.

1. __Factor in time for written tests and debriefs__. From the start of training, conduct regular [written tests](#written-tests) on days that new material is covered and then review tests and results with the class afterwards. Doing the test reviews the following day also simultaneously serves as a great recap of the material learned the previous day.

1. __Frequently change training modality__. This keeps trainees engaged and thus able to better absorb and retain content. For example, break long blocks of content training with group practices or role play activities for learning interviewing techniques. 

1. __Rotate trainers__ between different questionnaire modules to spice up the training and keep up attention levels. It is tiring for trainees to always hear the same voice and for trainers to lead a full day of training.

1. __Schedule time to debrief after field days__. Avoid back-to-back field practice or testing days, which may allow trainees to form idiosyncratic, potentially undesired behaviors. After a field practice or field test, allow for enough time to debrief, provide feedback and retrain topics that have not gone well, so trainees can try to do better on them during the next field day. Typically, one day of in-class training is needed between field days.

1. __Train what is relevant__. Only train content that interviewers need to know. Allocate time on topics proportionate to their direct importance for the interviewers. For example, spend less time on the general background of the survey and more on how to introduce it properly to respondents; or spend less time on explaining how an indicator is calculated and more time on how to properly collect the data needed for the indicator.     

1. __Train material just before it can be practiced__. Otherwise, trainees often forget the content, requiring it to be retrained. For example, it can be ineffective to train in detail the cover section of a questionnaire (identifiers, visit outcome, etc.) at the beginning of the training, as trainees do not yet know the protocols, questionnaire, respondents, etc. nor are familiar with CAPI yet. Instead, train the cover section details just before interviewers will use them for the first time, (e.g., before the first field practice) and once they have a better overview over questionnaire, respondents and protocols.

1. __Split long questionnaires__. First train and extensively practice (on-site and in field) one part before moving on to another part (e.g., household questionnaire and then agricultural questionnaire, or household level sections and then individual level sections). Breaking the content into more manageable sizes makes it easier for trainees to solidify learned material. 

1. __Schedule around low energy periods__. Schedule theory and important content in the mornings when trainees are usually freshest. Trainee attention tends to decrease in the afternoon and is at its lowest after coming back from the field or directly after having a heavy lunch. Directly after lunch, schedule hands-on and interactive modules that require greater engagement from trainees, such as practicing, interviewing techniques or written tests. Avoid training after coming back from the field.   

1. __Train supervisors and data monitors outside of main training hours.__ They need to attend the main fieldworker training in full to understand field procedures. Time slot options typically include prior to the main training, if roles are predetermined, or towards the end of the training once roles have been selected. This can be during the late afternoon or on the trainee’s administrative, travel or rest days.

1. __Do not exceed 6-7 hours of training time per day__ and give frequent breaks, without the breaks being too long and taking over the day. For example, this could be a morning session from 9:00 - 13:00h, and an afternoon session from 14:00 - 17:00h, each with a flexible 15-minute break between modules. Close the main training early enough to allow for supervisor training (when needed), daily trainer debriefs and preparing for the next day. 

1. __Rest days__. If training lasts more than one week, make sure to schedule one or two rest days per week. Trainees and trainers need regular breaks for good learning outcomes. 

The training schedule in Figure \@ref(fig:training-schedule) illustrates how fieldwork training can be organized for a survey with a 2-2.5 hours interview and a questionnaire in two parts. Each day lists the training modules to be covered and the total estimated number of hours to be spent on each one. Note that many modules should be split into individual sessions and mixed as will be illustrated further below. More details on what to consider and how to deliver each module are given in the corresponding part of Chapter \@ref(conducting-the-training) [Conducting the training](#conducting-the-training).  


```{r training-schedule, fig.show='hold', fig.cap="An illustrative training schedule", echo = FALSE, warning = FALSE, comment = FALSE, message = FALSE}
knitr::include_graphics("images/training-schedule-path.svg")
```

:::: {.tool}
<img src="images/tool.svg" class="svg-inline"/> **Tool:** Click [here](resources/training_schedule.xlsx) to download an Excel version of the illustrative schedule. 
:::::

For an actual training schedule more details usually need to be included, such as a breakdown of the questionnaire sections or topics to be covered in each module and the trainers and the location for each. Table \@ref(tab:schedule-day) below shows what a detailed schedule may look like for day 2, 3, 8 or 9 of the training outlined above. 

```{r , echo = FALSE, warning = FALSE, comment = FALSE, message = FALSE}
library(kableExtra)
options(kableExtra.html.bsTable = T)

score <- tibble::tribble(~Time, ~Module, ~Trainer, ~Location, 
  "09.00 - 09.45", "Review written test & recap", "Trainer A", "Classroom",
  "09.45 - 11.00", "Questionnaire content: Labor", "Trainer B", "Classroom",
  "11.00 - 11.15", "Tea break", "", "",
  "11.15 - 12.00", "Group practice: Labor", "", "Garden",
  "12.00 - 13.00", "Questionnaire content: Health", "Trainer A", "Classroom",
  "13.00 - 14.00", "Lunch break", "", "",
  "14.00 - 14.45", "Group practice: Labor & Health", "", "Garden",
  "14.45 - 15.30", "Questionnaire content: Education", "Trainer B", "Classroom",
  "15.30 - 15.45", "Tea break", "", "",
  "15.45 - 16.30", "Front-of-class interview: all sections", "Trainer A", "Classroom",
  "16.30 - 17.00", "Written test", "", "Classroom") 

score %>%
  kable(align = "llll", caption = 'Example of detailed schedule for one training day', label = "schedule-day") %>%
  kable_styling(bootstrap_options = c("striped","condensed")) %>%
  column_spec(1, bold = T)

```

Note that on this schedule questionnaire content sessions are no more than 1.25 hours long and are followed by practicing sessions so trainees can familiarize themselves with the newly covered material. Trainers and training modality change frequently and trainees get to change venue. More material is being introduced in the morning when trainees are freshest and more practicing is in the afternoon when energy levels are lower. Active training time is 6.5 hours and there are regular breaks.

For well-established surveys, in which time requirements are better known, a structured training schedule is feasible and recommended. Having a more general and flexible schedule would be useful and more efficient for surveys that are wholly new or even just new to the survey team. In this case, the schedule can be more easily adjusted during the training to reflect both progress and deficiencies without the burden of updating and redistributing a detailed schedule  every time there is a small change. Prior to the start of training prepare a rough schedule for the entire training to identify which resources are required on what day. During training, define detailed daily schedules for only the next 1-2 days of training.

## Location

Fieldwork training is best held in a location that:

- __Has sufficient logistical capacity__: Is easy to reach for trainers and trainees, has a suitable training venue (see more on this below), sufficient accommodation and dining/eating options for your group size, good network coverage, print shops, etc. 

- __Facilitates field testing__: Is close to communities that are similar to, but outside of, the sample. If the survey is limited to a specific region or area, it is best to conduct the fieldwork training there.

- __Facilitates field start__: Allows field teams to start field work shortly after the training and trainers to supervise the first few days of field work. 

Find a training venue with a hall large enough to comfortably fit the entire team (or multiple halls if parallel sessions are held) and space to breakout into smaller teams for group exercises, such as a garden, hallways, the canteen or additional rooms. It is crucial that there is __good internet connectivity__ at the venue, either through on-site Wi-Fi, or general network connectivity. 

The venue should be either able to accommodate trainees and cater their food, or be easily accessible and close to restaurants or eateries. It should be quiet and free from distractions. Good venues tend to be hotels or conference centers with garden or open space. Avoid universities, schools, towns halls or other places that are busy and become easily unavailable. 

:::: {.tip}
<img src="images/tip.svg" class="svg-inline"/> **Tip:** Book the venue early and check availability options in case the training needs to be extended on short notice.
:::: 


## Preparation {#trainingpreparation}

Avoid wasting precious training time by preparing in advance as much of the materials as possible. Prior to starting fieldwork training, the following should be in place:  

- __Trainees informed of survey details and terms__. To minimize attrition of trainees, make sure that trainees have been informed of and agreed in advance to the overall survey parameters and contract terms. This includes the selection process, their remuneration package, insurance coverage, areas to be visited, the length of field work, the transport and lodging arrangements during field work, expected effort, etc. Often, trainees drop out of the training if their expectations were wrong. Have trainees sign a copy of the details prior to the training to indicate their understanding of the terms. 

- __Contractual matters__. All contracts, payment arrangements, insurance policies and other administrative matters should be prepared such that they do not delay training or fieldwork. It is also important that trainees are paid without delays and are covered by insurance. Avoid interrupting the training with contractual matters, e.g., pulling out individual trainees to sign contracts. Instead, schedule dedicated admin slots prior to training, after class or between training and fieldwork. After fieldworker selection at the end of the training, make sure there is time to address all contractual matters (including insurance and payments) in the (usually short) window between training and the start of fieldwork.  

- __Logistical/administrative support__. Ensure there is a dedicated person to support all logistical and administrative work that has to be done during the training, e.g., setting up the venue, organizing transport, contacting local leaders, printing, purchasing equipment, signing fieldworker contracts, etc. If these activities have to be done by the trainers, they become a substantial distraction and can significantly affect training quality and pace.

- __Projector with a BIG screen__. The projected content must be visible and legible to all trainees. You should be able to connect a PC and as well as a CAPI tablet/phone to the projector. Ideally, it should be possible to quickly link trainees’ devices to the projector, to facilitate exercises and practicing. Test the set-up prior to the __first day__ of training.

- __Stable internet connection for all devices__. Trainees need to sync the devices used for CAPI data collection several times per day. Usually, the Wi-Fi connection of a training venue is not sufficient to connect all devices at the same time. In such cases, additional mobile network routers or wifi hot-spots are required. One option is to provide during the training the internet connection that is planned to support fieldwork, e.g., dongles, mobile phone routers or SIM cards.  

- __Food and refreshments for the breaks__. If food is provided by the venue, make sure it is ready in time for the scheduled breaks. If trainees eat out for lunch, set reasonable break times that allow trainees to finish lunch and to be back on time. Incentivize punctuality (e.g., those coming back late from the break have to sing a song).  

- __One tablet/phone per trainee__. It is crucial for all trainees to learn how to navigate the questionnaire in CAPI and how to use the device. If you do not have enough devices (since there are more trainees than interviewers) borrow additional devices for the training period. Devices must meet the minimum specifications for the CAPI software (see  [here](https://docs.mysurvey.solutions/faq/what-tablets-should-i-buy-/){target="_blank"} for Survey Solutions) and have the correct CAPI version installed. Consider using an app blocker so that only pertinent apps are available to users. Stress that the device is solely for the survey and not for other uses. Add shortcut/icons on the main screen for CAPI, the softcopy of the manual, the calculator or any other relevant app. 

- __Fully set-up CAPI software__. This can include installing the CAPI software on a server, creating interviewer accounts for each trainee, importing questionnaires to the server and syncing questionnaires/assignments to individual tablets – whatever steps are needed to ensure that tablets are ready for trainees to use. If the questionnaire includes substantial pre-loading (e.g., household rosters in a panel survey), you can also create several assignments partially pre-filled with different scenarios. 

- __Device charging__. Avoid delays due to charging issues. Provide (plenty of) extension cords to charge tablets during the training and/or in the evenings. Provide power banks for field days if the device charge does not last a full day.

- __Samples of documents__. If fieldworkers need to interpret, verify or extract information from any documents or items such as immunization cards, school timetables, or anti-malaria medications, get enough copies/samples to allow for hands-on practicing. If there is variation, provide different examples, e.g., different antimalarial medication brands. 

- __Measurement equipment__. If fieldworkers need to use any special measurement equipment –height/length boards, GPS units, water testing kits, student testing booklets, etc. – provide enough units so that trainees can practice in a meaningful way. 

- __Field procedure documents__. Ensure sufficient copies of any additional documents (beyond the questionnaire and manuals) that fieldworkers need to use in the field – introduction letters, household tracking lists, maps, cluster completion sheets, referral forms, etc.
 
- __Finalized questionnaires__. The questionnaires should already be reviewed, extensively pre-tested or piloted, translated and the translations verified. The CAPI version of the questionnaire should be extensively desk-tested and error free. Mistakes can cause significant issues or delays during the training. For surveys that will be implemented using paper questionnaires, make sure that there are copies of the paper questionnaire available for each trainee to have his/her own copy.  For surveys that will be implemented using CAPI, while it can sometimes be useful to have paper copies of the questionnaire for the trainees, try to avoid this as the trainees need to be able to follow the flow as defined by the CAPI program. 

- __Fieldworker manuals__. Realistically, fieldworker manuals tend not to be used during fieldwork. Their primary value lies as a reference during training and later, for data users, to help them interpret questions and concepts. The content of manuals should mirror the actual training and how fieldworkers interpret questions and concepts. Unfortunately, this is often not the case. For manuals to be useful, good drafts for interviewer, supervisor, and any other fieldworker roles should be available to trainees from the start of training. Modifications and additions to the manuals should be made during training. Manuals should be constantly referred to throughout the training so that trainees develop the habit of consulting them as a knowledge base. It is beneficial to hand out printed versions of the manual at the beginning (as drafts) and at the end (as revised copies), or to provide regularly updated soft copies in PDF format, which trainees can access on their devices—a method that tends to work very well. 

:::: {.tool}
<img src="images/tool.svg" class="svg-inline"/> **Tool:** The R application [Survey Solutions Questionnaire Manual Generator](https://github.com/michael-cw/susoquestionnairemanual){target="_blank"}
is a useful tool to generate questionnaire manuals in Word, PowerPoint or Html format from a Survey Solutions questionnaire. 
::::

- __Draft written tests__. Prepare questions for the written tests (link) that cover all concepts and questions in the questionnaire. Test the question with the survey designer. This usually helps to uncover outstanding gaps in the design and ensures the test questions are functioning (unambiguous, understandable, etc.) and the marking is correct. Having questions prepared will allow you to quickly produce daily tests tailored to the day’s progress. Without prior preparation, creating daily tests can be challenging. Find out more on [written tests](#written-tests).

- __Confirmed in-field practice__. Select practice areas/communities well in advance. These must be locations that are not included in the sample. Make all necessary administrative arrangements such as getting permissions, introduction to village officials, etc. If necessary, arrange respondents for field practices so trainees can spend more time practicing than looking for respondents to participate (don’t do this for the final field test). If you require certain types of respondents (e.g., children for anthropometrics) check with institutions where they are concentrated, e.g., child clinics, schools, etc. Organize reliable transport for the trainees to travel to the field practice. Make sure the transport waits for the team, not the other way round.

- __Confirmed on-site practice__. Arrange respondents who will be used for on-site practicing sessions to visit the training venue. Arrange for their transport and remuneration and don't forget to also provide food/refreshment for them.    

- __Facilitation material__. This includes microphones and speakers (with batteries and fully tested), flip charts with markers in multiple colors for drawing, name tags/stickers for all participants, printed attendance lists and feedback forms, notebooks, pens in 2 colors, and clipboards for trainees, etc. It tends to be useful to have a printer or printing facilities in or very near the venue to allow ad-hoc printing of exercises, tests, etc.   

- __Field work logistics and material__. All necessary logistics need to be in place to allow fieldwork to commence immediately after training. This includes fieldworker contracts ready to be signed upon selection, covering insurance policies, payment arrangements, transport, accommodation, etc. Similarly, have all fieldwork equipment ready, such as ID cards, letters of introduction (if needed), back packs, stationary (pens, pencils, paper, etc.), raincoats/boots, battery packs, t-shirts, respondent incentives, etc.


<img src="images/LSMS-Logo_300dpi_06.png" style="height: 6em; margin-right: 10%;" /><img src="images/WB-WBG-horizontal-RGB.png" style="height: 4em;" />