Home 
Contact 
Profile 
Videos 
Prepared and written by Professor
Omar Hasan Kasule Sr for medical students at the Faculty of Medicine King Fahad
Medical City, Riyadh, Saudi Arabia on November 19, 2014
Weekly objectives
§
List the factors required for a valid population
health survey.
§
Describe the differences between an individual
and a population approach to a health problem.
Learning Objectives
§
Define population health survey.
§
Describe methods for planning and conducting a
population health survey.
§
Describe appropriate design of the instruments
used for data collection (questionnaires, interviews) in population health
surveys.
§
Show how sampling and sample size are decided.
Nature & Definition
§
Surveys are observational studies that involve a
large number of respondents than the usual sample
§
They supplement information from the census and
vital statistics registration
Objectives
§
Measure of health and disease
§
Assessment of needs
§
Assessment of use of services
§
Assessment of effectiveness of care
Types Of Surveys
§
Surveys can be census surveys covering the whole
population or sample surveys covering selected samples.
§
Either of these may be carried out as a mail
survey, a telephone survey, a personal interview, or as an observational
survey.
Administrative Planning 1
§
Literature survey
§
State the objectives: identify the problem being
investigated. Determine the priority of the problem.
§
Formulate a hypothesis
§
Define the population
§
Define the sampling frame
§
Define the size of the sample and the method of
sampling
Administrative Planning 2
§
Train investigators
§
Consider logistics: clearance, manpower,
materials and equipment (e.g. maps), finance, transport, communication,
accommodation
§
Prepare and pre-test study questionnaire
Study Design
§
Type of study: cross sectional or longitudinal
§
Sampling unit: household is usually employed as
basic sampling unit
§
Sampling methods: simple random sampling,
systematic sampling, stratified sampling, cluster sampling, multistage sampling
Sample Size
§
Sample size is
n = Z2 p(1-p) / d2
§
where d is the precision usually set at 0.05 or
0.1.
§
If n<0.1N, the formula for sample size becomes n
= {Z2 p(1-p) / d2} / {n/(1 – n/N)
Data Collection
§
Existing data
§
Questionnaire
§
Physical examinations
§
Direct observation
§
Laboratory investigations
Features Of A Good Questionnaire 1
§
Brief.
§
Questions are clear with no ambiguities.
§
Questions are not leading
§
The questions are in a logical order (e) skip
patterns are clear and are logical
Features Of A Good Questionnaire 2
§
The most important questions are at the start;
sensitive questions are kept at the end
§ The closed questionnaire is preferred to the
open one. The closed one must have categories for ‘other’ and ‘I do not know’
to take care of unexpected alternatives.
Methods Of Questionnaire Administration:
§
Postal,
§
Telephone,
§
Diaries,
§
Interview
Data Management
§
Item analysis for outliers
§
validation
Biases
§
Observer bias
§
Sampling bias
§
Selection bias
§
Response bias
§
Dropout bias
§
Memory bias
§
Self selection bias
§
Lead time bias
Uses Of The
Survey Report
§
Communicate the findings
§
Application for funding
Ethical Issues
§
Sampling
§
Confidentiality
§
Research on human subjects
Legal
Considerations In Field Surveys
§
Gaining access to subjects
§
Protection of records
§
Access to personal and medical records
§
Trade and financial information
§
Report contagious or dangerous diseases
discovered during research
§
Subpoena of epidemiologists to testify in court
§
Taking compulsory measures: immunization,
isolation, quarantine