DATA COLLECTION Dr Hayati Bte Kadir Shahar Unit

DATA COLLECTION Dr Hayati Bte Kadir @ Shahar Unit of Epidemiology & Biostatistics Dept. of Community Health

Learning objectives ® Understand the different types of data ® Enumerate various sources for collection of data ® Enumerate the advantages and limitations of using secondary or existing data ® Differentiate between data collecting techniques and data collecting tool ® Identify and describe common methods of data collection ® Identify the bias in data collection and how it can be reduced

Introduction ® Data collection refers to collections of information necessary to answer the research questions. ® This has to be done systematically, accurately and according to standards

Types of data ® Primary data ® Secondary data

Primary data ® First hand data collected by the investigator and is related to study being conducted & follows the study protocol ® Example of sources of primary data ® Personal interviews, written questionnaire, medical examination or tests

Secondary data ® Data collected for other purpose other than those of the study: individual who initially collected the data did not know that the data will be used for another study purpose ® Example of sources of secondary data ® Hospital medical record, health center records, past surveys conducted by various ministries, universities and NGO’s, registries of birth and deaths

Data collection techniques ® Basically ® Using it is divided into 6 techniques: secondary data ® Observing ® Interviewing ® Administering written questionnaire ® Focus Group Discussion ® Laboratory techniques

Using secondary data Locate the source May need to prepare a checklist or compilation sheet Retrieve the information needed

Using secondary data-available information or past medical records ® Advantages: ® Disadvantages: Inexpensive Ø Permit examination of past trends Ø Ø Problems with accessibility Ø Ethical issues concerning confidentiality Ø Imprecise Ø Incomplete information

Observation ® Purposeful, systematic and selective way of watching and listening to an interaction or phenomenon as it takes place ® Usually done in situation where accurate information cannot be elicited by respondents by questioning, not cooperative or unaware of the answers

Observation ® Participant observation ® Researcher participates in the activities being observed ® Respondents can be/not aware that they are being observed ® Eg. Examine the reaction of people regarding wheel chair/crutches researcher sits in a wheelchair /use crutches

Observation ® Non-participant ® Researcher observation does not get involve in the activities of the group, just observes ® Eg. Study the functions of nurses in a hospital

Observation ® Advantages ® Disadvantages More detailed and context related information Ø Permits collection of information on facts not mentioned in questionnaires Ø Ø Observer bias may occur (inter and intra) Ø The presence of data collector may influence the situation observed

Interview ® Face to face interview involves oral questioning of individuals ® Answers are usually recorded through tape recorder, video camera or written down ® Interviewing skills are important to gather correct data ® How to ask questions in written and spoken form

Interviewing ® Advantages ® Disadvantages Suitable for use with Ø Presence of illiterates interviewer can influence response Ø Permits clarification of questions Ø Reports of events may be less Ø Higher response rate complete compared than written to observation questionnaire Ø

Interview (small scale flexible interview) ® Advantages Ø Permits collection of in depth information and exploration of spontaneous remarks by respondents ® Disadvantages Influence of the interviewer on the respondents Ø Open ended data are difficult to analyze Ø

Interview (large scale fixed interview) ® Advantages Ø Easy to analyze ® Disadvantages Ø Important information may be missed because spontaneous remarks by respondents are not recorded or explored

Administering Written Questionnaires ® Advantages ® Disadvantages Less expensive Ø Permits anonymity and may result in a more honest response Ø +/- require research assistants Ø Ø Cannot be used with illiterate respondents Ø There is often a low rate response Ø Questions may be misunderstood

Focus Group Discussion ® Conducted when a researcher wants to obtain a consensus from a group regarding the area of study ® Vital for decision making based on the opinions.

Laboratory Techniques ® Advantages ® Disadvantages Ø Little application Allows for many types of treatment to a sample Ø Ø Results do not vary much in repeated experiments Ø Ø Researcher has total control over the Ø experiments Ø Ø of randomization and this makes statistical analysis difficult Very prone to observer and sampling bias Relatively expensive to carry out although can be repeated Need lab facilities to carry out Need skilled technician in lab techniques

Data collection plan ® Planning in data collection is essential for the following reasons: ® To know who will collect the data ® When to collect the data ® How long will it take to collect the data ® Efficient use of human and material resources for data collection ® To minimize errors and delays which is often the result from lack of planning

Data collection plan ®A proper plan for data collection can be divided into 3 stages: ® Acquisition of permission from relevant authorities ® Data collecting process ® Data management

Data collection plan ® Acquisition of permission include: ® Getting permission from authorities to carry out the project ® Getting informed consent from participants who are part of the project ® Relevant authorities depend upon where the research will be carried out: In a university Dean or Vice Chancellor ® In a village/’kampung’ setting village headman ® In a hospital the hospital director *often related to ethical considerations in a research ®

Data collection plan ® Data collecting process ® This is mainly dealing with logistics of data collection ® List of tasks and assignment Decide time fame for data collection ® Decide when to collect the data ® Sequence of data to be collected if several data is to be collected ® How to ensure quality data is collected-reliable and valid ® Prevention of bias at every stage ®

Source of bias in data collection ® Defective bias) instruments (main source of ® Defective questionnaires ® Not standardised equipments ® Observer bias (intra and inter) ® Data contamination during examination, withdrawal of participants, low compliance of participants