Difference between revisions of "Survey Pilot Participants"

Jump to: navigation, search
(Characteristics)
Line 12: Line 12:
 
|-
 
|-
 
|-
 
|-
 +
! style="text-align:center;"|
 
!
 
!
!
+
! style="height:50px; width:250px; text-align:center;"| Stage 1 - Pre-Pilot
! Stage 1 - Pre-Pilot
+
! style="height:50px; width:250px; text-align:center;"| Stage 2 - Content-focused Pilot
! Stage 2 - Content-focused Pilot
+
! style="height:50px; width:250px; text-align:center;"| Stage 3 - Data-focused Pilot
! Stage 3 - Data-focused Pilot
 
 
|-
 
|-
 
| '''Respondents'''
 
| '''Respondents'''
| ''Never use any respondents in the final sample. Neighboring villages / schools / firms are good options.''
+
|  
 +
| Respondents for the pilot should be as similar as possible to respondents for the actual survey in terms of age, gender, education, and socioeconomic (income) status.  
 +
 
 +
Neighboring villages, schools, or firms are good options for selecting respondents for the pilot.
 +
 
 +
Never use any of these respondents in the final sample.
 +
| style="vertical-align:top;"| Test the protocols for [[Sampling|sampling]] in this stage. Use these protocols to select respondents for the pilot.  
 +
 
  
| Respondents should be as similar as possible to actual respondents.
+
 
Include variations in age, gender, education, and income / socioeconomic status.
+
This helps to find out if the '''sampling frame''' (list from which respondents are selected) needs to be revised. It also helps to reduce [[Selection Bias|selection bias]].  
| Test sampling protocol by using it to select respondents.
+
| style="vertical-align:top;"| Re-test your '''sampling protocols'''.
Helps troubleshoot (e.g. find out sampling frame is outdated) and reduces bias in selection.  
+
 
| Test your sampling protocol, including any revisions based on previous piloting stage
+
 
 +
 
 +
 
 +
 
 +
Include revisions based on the previous round of '''piloting'''.
 
|-
 
|-
| rowspan=5 | '''Interviewers'''
+
| rowspan=4 | '''Enumerators (or interviewers)'''
| ''Language''
+
|style="height:100px;"| '''Language'''
| Fully fluent (read/write/ speak) in local language and language of research team
+
| Fully '''fluent''' (can read, write and/or speak) in the local language(s) and the language of research team.
| Fully fluent (speaking and reading) in local language and language of research team
+
| Fully fluent in the local language(s) and the language of research team.
| Fully fluent (speaking) in the local language and language of research team
+
| Fully fluent in the local language(s) and the language of research team.
 
|-
 
|-
| ''Experience''
+
| '''Required experience'''
| Experience as qualitative enumerator
+
|style="vertical-align:top;"| '''Qualitative''' experience as an enumerator. For instance, experience in identifying issues in wording and structure of questions.
Sectoral knowledge  
+
 
|Experience as quantitative enumerator
+
'''Sectoral''' knowledge. For instance, knowledge about agricultural patterns in a survey that aims to collect data on farm output.
Sectoral knowledge
+
| style="vertical-align:top;"| '''Quantitative''' experience as an enumerator. For instance, experience in entering data electronically.
|Experience as quantitative enumerator
+
 
Experience with tablets
+
 
 +
Sectoral knowledge.
 +
| style="vertical-align:top;"|Quantitative experience as an enumerator.
 +
 
 +
 
 +
 
 +
Experience working with phones or tablets.
 
|-
 
|-
| ''Background''
+
| '''Background'''
| Diversity in age / race / religion / gender / etc
+
| style="vertical-align:top;"| The pool of interviewers should be diverse in terms of age, race, religion, gender etc.
| Be mindful of (and learn about) cultural requirements (should women interview women? Can women interview men?)
+
| style="vertical-align:top;"| Interviewers should be mindful of (and learn about) cultural considerations. For instance, if women can interview men, or if men can interview women in the study area.
| Interviewer pool should reflect decisions regarding composition of final survey team, based on previous stage
+
| style="vertical-align:top;"| The background of the team of interviewers should be similar to that of the final survey team, which is decided based on previous stages of '''piloting'''.
 
|-
 
|-
| ''Size of Team''
+
| '''Size of the team'''
| Very small (~2). Always accompanied by research team member.
+
|style="vertical-align:top;"| Very small - not more than 2.  
| Small (~2-4). Ideally each enumerator accompanied by research team member or other observer.
+
 
| Larger (~4-8). The more interviews the better for debugging and data checks.
+
A member of the research team should always accompany the enumerators.
 +
|style="vertical-align:top;"| Small - between 2 and 4.  
 +
 
 +
Ideally, a member of the research team should monitor each enumerator.
 +
|style="vertical-align:top;"| Large - between 4 and 8.  
 +
 
 +
More enumerators help simplify the process of [[Monitoring Data Quality|monitoring data quality]] and '''debugging''' (identifying errors) in the [[Questionnaire Programming|coded instrument]].
 
|-
 
|-
| ''Who?''
+
| '''Survey firm'''
| Typically done before survey firm is on board, using local research assistants (STCs hired through the WB or a local PI), local government staff, or university students
+
|  
| Typically done before survey firm is on board, using local research assistants (STCs hired through the WB or a local PI), local government staff, or university students
+
| Not part of this stage.
| Best case scenario: survey firm leads. Staff who will be field supervisors for the upcoming survey act as interviewers. They gain experience with the instrument, which also helps for enumerator training.  
+
 
 +
This stage is typically conducted before survey firm is on board.
 +
| Not part of this stage.
 +
 
 +
This stage is typically conducted before survey firm is on board.
 +
| Part of this stage.  
 +
 
 +
In fact, the survey firm should lead the pilot in this round.  
 
|-
 
|-
| '''Research Team'''
+
| '''Members of the research team'''
| ''Field Coordinator participates in all stages''
+
|
| PI should participate directly. If not possible, plan for daily debriefs and discussions.  
+
|style="vertical-align:top;"| The [[Impact Evaluation Team#Field Coordinators (FCs)|field coordinators (FCs)]]. Ideally they should participate in each stage.
| Research Manager or Impact Evaluation Coordinator should participate directly.   
+
 
| Programmer participates directly (to adapt / debug in real time)
+
The [[Impact Evaluation Team#Principal Investigators (PIs)|principal investigators (PIs)]] should participate directly. They should also conduct daily discussions to '''debrief''' (exchange notes with) the enumerators.  
 +
|style="vertical-align:top;"| The field coordinators (FCs).
 +
 
 +
 
 +
The [[Impact Evaluation Team#Impact Evaluation Coordinator|impact evaluation coordinator]] (or research manager) should participate directly.   
 +
|style="vertical-align:top;"| The field coordinators (FCs).
 +
 
 +
 
 +
The '''programmer''' should participate directly (to debug the instrument in real time)
 +
 
 +
Members of the research team who will be '''field supervisors''' in the actual survey should act as interviewers in this stage. This gives them experience with the instrument, which helps them conduct [[Enumerator Training|enumerator training]].
 +
|-
 +
| '''Others'''
 +
|
 +
| Local [[Impact Evaluation Team#Research Assistants (RAs|research assistants (RAs)]]. Includes short-term consultants (STCs) hired through the World Bank or through a local principal investigator (PI).
 +
 
 +
Local government staff.
 +
 
 +
University students.
 +
| Local research assistants (RAs).
 +
 
 +
 
 +
 
 +
 
 +
 
 +
Local government staff.
 +
 
 +
University students.
 +
| Local research assistants (RAs).
 +
 
 +
 
 +
 
 +
 
 +
 
 +
Local government staff.
 +
 
 +
University students.
 
|}
 
|}
  

Revision as of 00:43, 13 April 2020

Typically, conducting a survey pilot requires several stages of planning and discussions. Apart from the impact evaluation team (or research team), survey pilot participants include interviewers, respondents, and even local government agencies. For instance, during the actual field data collection, such as in a computer-assisted personal interview (CAPI), it is the interviewer who reads out the questions to the respondent. But in the piloting stages leading up to the actual survey, people like the field coordinators (FCs) and the principal investigators (PIs) oversee the overall preparation and finalize the survey protocols.

Read First

Characteristics

Stage 1 - Pre-Pilot Stage 2 - Content-focused Pilot Stage 3 - Data-focused Pilot
Respondents Respondents for the pilot should be as similar as possible to respondents for the actual survey in terms of age, gender, education, and socioeconomic (income) status.

Neighboring villages, schools, or firms are good options for selecting respondents for the pilot.

Never use any of these respondents in the final sample.

Test the protocols for sampling in this stage. Use these protocols to select respondents for the pilot.


This helps to find out if the sampling frame (list from which respondents are selected) needs to be revised. It also helps to reduce selection bias.

Re-test your sampling protocols.



Include revisions based on the previous round of piloting.

Enumerators (or interviewers) Language Fully fluent (can read, write and/or speak) in the local language(s) and the language of research team. Fully fluent in the local language(s) and the language of research team. Fully fluent in the local language(s) and the language of research team.
Required experience Qualitative experience as an enumerator. For instance, experience in identifying issues in wording and structure of questions.

Sectoral knowledge. For instance, knowledge about agricultural patterns in a survey that aims to collect data on farm output.

Quantitative experience as an enumerator. For instance, experience in entering data electronically.


Sectoral knowledge.

Quantitative experience as an enumerator.


Experience working with phones or tablets.

Background The pool of interviewers should be diverse in terms of age, race, religion, gender etc. Interviewers should be mindful of (and learn about) cultural considerations. For instance, if women can interview men, or if men can interview women in the study area. The background of the team of interviewers should be similar to that of the final survey team, which is decided based on previous stages of piloting.
Size of the team Very small - not more than 2.

A member of the research team should always accompany the enumerators.

Small - between 2 and 4.

Ideally, a member of the research team should monitor each enumerator.

Large - between 4 and 8.

More enumerators help simplify the process of monitoring data quality and debugging (identifying errors) in the coded instrument.

Survey firm Not part of this stage.

This stage is typically conducted before survey firm is on board.

Not part of this stage.

This stage is typically conducted before survey firm is on board.

Part of this stage.

In fact, the survey firm should lead the pilot in this round.

Members of the research team The field coordinators (FCs). Ideally they should participate in each stage.

The principal investigators (PIs) should participate directly. They should also conduct daily discussions to debrief (exchange notes with) the enumerators.

The field coordinators (FCs).


The impact evaluation coordinator (or research manager) should participate directly.

The field coordinators (FCs).


The programmer should participate directly (to debug the instrument in real time)

Members of the research team who will be field supervisors in the actual survey should act as interviewers in this stage. This gives them experience with the instrument, which helps them conduct enumerator training.

Others Local research assistants (RAs). Includes short-term consultants (STCs) hired through the World Bank or through a local principal investigator (PI).

Local government staff.

University students.

Local research assistants (RAs).



Local government staff.

University students.

Local research assistants (RAs).



Local government staff.

University students.

Related Pages

Click here for pages that link to this topic.


The Field Coordinator plays a central role. Ideally other members of the research team will also participate (particularly important for pre-pilots and content-focused pilots). Staff from the relevant Government Ministry or implementing partner should be included as much as feasible (depending on availability and direct engagement). Typically, it is most productive for all pilot participants (field coordinator, PIs, local partners) to accompany skilled interviewers and observe and take notes, rather than conducting the interviews themselves.


Additional Resources