MICS4 Survey Design Workshop Multiple Indicator Cluster Surveys Survey Design Workshop Survey Quality Control.

MICS4 Survey Design Workshop

Multiple Indicator Cluster SurveysSurvey Design Workshop

Survey Quality Control


Objectives

• Identify factors affecting the accuracy and reliability of survey data

• How to prevent and correct errors

• The essential role of supervision in the field and providing feedback to the team


Overview of presentation

• Preparation for fieldwork• Data collection• Field supervision• Data processing





Preparation for fieldwork• Questionnaire

– Adaptation (relevant, tailored to needs)– Translation (for uniformity)– Pretest (good flow, capture relevant information)

• Sample design and size– For reliability, accuracy, feasibility

• Interviewers– Selection (systematic process, have surplus)– Training (trainers, length, practice)





Data Collection

• Logistics & discipline• Implementation of sample• Monitoring of fieldwork


Data Collection:

• Material & equipment (questionnaires/forms)• Transport• Accommodation• Payments to field staff• Discipline & reasonable expectations• Keep the morale of the “troops”

Logistics & discipline


Data Collection: Implement. of sample

• Non-response: failure to obtain information for selected households, eligible women, or children

• A potentially serious bias that can be minimized

• Interviewers will need to make return visits to households (call-backs)


1. Interviewer is unable to do the selected household

2. Interviewer unable to meet eligible respondents

3. Respondent refuses to be interviewed

Types of non-response:



1. Interviewer is unable to do the selected household. Why?


– Structure not found– Occupied structure inaccessible– Structure non-residential, vacant, or demolished




2. Interviewer unable to meet eligible respondents. Why?


– No one at home at time of call– Respondent temporarily absent Will need call-backs


Call-backs

• Interview that is not completed requires a “call-back” or follow-up visit

• Three call-backs required (different times and days)

• Supervisors and interviewers keep track using control sheets

• Requires good tracking of work to ensure that all planned interviews are completed before leaving cluster



3. Respondent refuses to be interviewed. What to do?


– Approach respondent from her point of view– Postponed interview to another day– Have field editor carry out the interviewThe number of refusals should be closely

monitored. Need to find out reasons for frequent refusals.


Data Collection: Monitoring field work

• Training is a continuous process

• Observation and supervision throughout the fieldwork are a part of the training

• Team supervisors and field editors play very important roles in continuing this training and in ensuring the quality of MICS4 data


Data Collection: Monitoring field work

• Fieldwork control sheets• Direct observation of interviews• Review of completed questionnaires (editing)• Spot-checking households


Fieldwork Control Sheets

• Interviewers’ work is monitored and evaluated by keeping accurate record of assignments

• Both supervisors and interviewers have control forms to maintain

• These forms should be returned to the director of field operations along with the completed questionnaires

• Interviewer is responsible for ensuring that control sheet is up-to-date


Observing Interviews, part 1

• To evaluate and improve interviewer performance

• To look for errors and misconceptions that cannot be detected through editing

• Why? Precise but inaccurate answers • Who? The supervisors or field editors (females

only for women’s questionnaire)


Observing Interviews, part 2

• How often? 5-6 interviews per week, more at start of fieldwork

• How? Just take notes without disrupting the interview

• What to do after? Supervisors speaks to interviewers


Editing Questionnaires in the Field

• Fundamental to survey quality

• Need to ensure:

– Accurate and complete information in each questionnaire

– Correct count of questionnaires


Editing Questionnaires in the Field

• Done daily by field editor; supervisor can assist

• Finish all questionnaires before leaving cluster

• Provides an opportunity to identify and correct mistakes before leaving cluster

• Results and errors are discussed with interviewers


Systematic Spot Checking

• How to spot-check household composition?– Supervisors should complete columns 2-8 in the

HH questionnaire and compare with that of the interviewer

– Check about 5% of households (5-6 per week)– All team members must be spot-checked;

provide feedback if necessary


Evaluating Interviewers Performance

• Daily discuss the quality of interviewers work • Point out mistakes discovered during observation of

interviews or noticed during editing• Discuss examples of actual mistakes, but be careful not to

embarrass individual interviewers. • Re-read relevant sections from the Interviewer's Manual

with the team to resolve problems. • Encourage the interviewers to talk about any situations

they encountered in the field • Discuss whether situations are handled properly, and how

to do it in the future





Field supervision• Staff from survey headquarters

– Senior staff

– Quality control team

• Purpose

– Visit teams to observe interviews, review work

– Re-visit selected clusters; spot-check households

• Very important during first few weeks of fieldwork! (Possible to start teams in one location?)





Data Processing

• Integral aspect of survey quality (DP programmes, double-entry, secondary editing)

• Problems detected in DP should be communicated back to teams (phone calls and visits)

• Field-check tables

• Observations from reviewed questionnaires


GENERAL INSTRUCTION TO EDIT QUESTIONNAIRES

According to MICS Manual

Under no circumstances should we make up an answer

MICS4 Survey Design Workshop Multiple Indicator Cluster Surveys Survey Design Workshop Survey Quality Control.

Documents

quality of mics4 data

reliability of survey

sample slide

questionnaire slide

cluster slide

team slide

practice slide

uptodate slide