211008P - CONSTRUCT THE METHODS OF DATA COLLECTION, DATA MANAGEMENT, AND DATA ANALYSIS

Presented at Alfaisal University Scholar Program in Research Excellence (ASPIRE) on October 8, 2021 by Dr Omar Hasan Kasule Sr MB ChB (MUK), MPH (Harvard), DrPH (Harvard) Professor of Epidemiology and Bioethics King Fahad Medical City.

 

WHAT KIND OF DATA WILL BE COLLECTED, AND HOW MUCH?

· Definition of a variable: quantitative (discrete and continuous) and qualitative (nominal, ordinal, ranked)

·  Types of variables: independent, dependent, confounding

·  Determine how much data to collect (target sample size) and the level of precision

· When the data will be collected (month and year);

· How long data collection will continue (duration of research);

· Where the data will be collected (location, situation).

· Access and authorization of data collection: access to people and organizations,  access to events and settings,  access to equipment, and access to documents and records. IRB approval does not guarantee access.

 

HOW WILL THE DATA BE COLLECTED?

· Record review

· Interview

· Questionnaire

· Observation

· Physical clinical measurements

· Educational or psychological tests

· Laboratory and radiological

· Environmental

 

SOURCES OF SECONDARY DATA

· Census data is a reliable source of demographic, social, economic, and health information.

· Vital events are births, deaths, Marriage & divorce, and some disease conditions.

· Routinely collected data: medical facilities, life, and health insurance companies, institutions (like prisons, army, schools), disease registries, and administrative records.

·  Observational epidemiological studies are of 3 types: cross-sectional, case-control, and follow-up/cohort studies.

· Special surveys cover a larger population than epidemiological studies and may be health, nutritional, or socio-demographic surveys.

 

PRIMARY DATA COLLECTION BY QUESTIONNAIRE

· Questionnaire design involves content, wording of questions, format, and layout.

·  The reliability and validity of the questionnaire pilot tested

·  Informed consent and confidentiality must be respected.

·  A protocol sets out data collection procedures.

· Questionnaire administration: face-to-face interview, telephone, Computer-administered.

 

PHYSICAL PRIMARY DATA COLLECTION

· Clinical examination,

· Standardized psychological/psychiatric evaluation,

· Measurement of environmental or occupational exposure,

· Assay of biological specimens (endobiotic or xenobiotic)

 

DATA MANAGEMENT

· Self-coding or pre-coded questionnaires are preferable.

· Data is input as text, multiple choice, numeric, date and time, and yes/no responses.

· Interactive data entry enables the detection and correction of logical and entry errors immediately. Data replication is a copy management service that involves copying the data and also managing the copies.

· Data editing is the process of correcting data collection and data entry errors.

· Data is validated and its consistency is tested.

· Data transformation is the process of creating new derived variables preliminary to analysis

 

DATA ANALYSIS

· Data analysis consists of data summarization, estimation, and interpretation.

· Descriptive statistics are used to detect errors, ascertain the normality of the data, and know the size of cells.

· The tests for the association are the t, chi-square, linear correlation, and logistic regression tests or coefficients.

· The common effect measures Odds Ratio, Risk Ratio, and Rate difference.

· Analytic procedures and computer programs vary for continuous and discrete data

 

ASSESSING THE SUITABILITY OF THE METHODS

· Will the methods produce data that are relevant for addressing the research questions?

· Are the methods the best available under the circumstances? Are there better alternatives?

· Will the methods work? Will they do the job?

· Risk assessment: what can go wrong? How can we detect it? Prevention and mitigation of risk.

 

ASSESSING THE FEASIBILITY OF THE METHODS

· Feasibility is best tested by a pilot study

· Resources availability

· Type of data: The use of qualitative or quantitative data: What are their respective strengths? Which is better suited to the needs of this particular research? Is a mixed-methods approach preferable?

· Depth or breadth of data: Will a case study be better than a survey, or vice versa, in terms of the particular research questions being looked at? Is there a need for depth of focus or is there a need for data drawn from widespread sources?

· The validity of the data produced: Will the data be accurate? Will they focus on the right issues? Is the chosen method better than the alternatives in terms of getting honest responses from participants?

· The reliability of the method: Will the method(s) produce the same data if the same research is repeated?

· The possibility of generalizing from the findings: Can the findings be extrapolated to other situations/examples? Is this possible and is it important? Is this crucial for the research?

· The extent to which the data are representative: Is it better to include all (or a sample) of a population or will research along the lines of a case study be more suitable? Are data based on extreme examples or special instances more valuable?

· The extent to which the methods are objective: Is this possible bearing in mind the research questions being addressed? How much does it matter?

 

LIMITATIONS OF THE METHODS

· Limitations associated with the methods; and

· Limitations caused by circumstances beyond the control of the researcher.

· Limits to how far the findings lend themselves to being generalized to other situations/examples;

· Limits to the possibility of checking the accuracy of findings;

· Limits to the ability to confirm that data comes from a representative sample of the research population;

· Limits to objectivity resulting from the role of the researcher in data collection and analysis.

· Restricted access to significant sources of data;

· Restrictions arising from the resources available (time and money);

·  Limits to the sample size.

 

ETHICAL ISSUES

· Discuss likely issues and not just mention will get IRB approval.

· Autonomy/informed consent.

· Confidentiality.

· Data protection regulations.

· Regulations about tissues especially DNA analyses.