Data Collection
Data collection is a crucial step in the research process, commencing after a research problem has been clearly defined and a research design/plan has been established. Researchers must decide whether to gather primary data or utilize secondary data for their study.
Primary data refers to information collected for the first time, making it original in character. This type of data is gathered specifically for the current research purpose, either by the researcher themselves or by someone else on their behalf. The collection of primary data is an original process.
Methods of Collecting Primary Data:
- Observation Method
- Interview Method
- Structured Interviews
- Unstructured Interviews
- Questionnaires
- Structured Questionnaires
- Unstructured Questionnaires
- Schedules
- Other Methods: warranty cards, distributor audits, pantry audits, consumer panels, mechanical devices, projective techniques, and depth interviews.
Secondary data refers to information that has already been collected and processed by someone else for purposes other than the current study. This type of data collection involves compilation rather than original generation.
Sources of Secondary Data: Secondary data can be published or unpublished:
- Published Sources: Government or semi-government publications (e.g., census, vital statistics, labor force surveys, health reports, economic forecasts), technical and trade journals, books, magazines, newspapers, reports from various associations (business, industry, banks), and reports by research scholars or universities.
- Unpublished Sources: Diaries, letters, unpublished biographies and autobiographies, and data available with scholars, research workers, trade associations, and public/private organizations.
Collection of Primary Data
The choice of a primary data collection method depends on factors such as the study's purpose, available resources, researcher's skills, and the socioeconomic-demographic characteristics of the study population.
Here are the important methods for collecting primary data, particularly in surveys and descriptive research:
- Observation method
- Interview method
- Through questionnaires
- Through schedules
- Other methods such as: Warranty cards, Distributor audits, Pantry audits, Consumer panels, Using mechanical devices, Projective techniques, Depth interviews, Content analysis
1. Observation Method
This method involves the investigator's direct observation of phenomena without questioning respondents. It is a purposeful, systematic, and selective way of watching and listening to interactions or phenomena as they occur.
It is most appropriate when researchers are more interested in behavior than in individuals' perceptions, or when respondents cannot provide objective information. It's useful for studying group interactions, dietary patterns, worker functions, or individual behavior/personality traits.
Advantages:
Eliminates subjective bias if done accurately.
The information relates to current happenings, not complicated by past behavior or future intentions.
It is independent of the respondent's willingness to respond and is less demanding of their active cooperation.
Particularly suitable for studies involving subjects who cannot give verbal reports of their feelings.
Limitations:
- Respondents may alter their behavior if aware they are being observed.
- Interpretations can vary among observers.
- There is a possibility of incomplete observation or recording.
- It can be an expensive method and provides limited information.
- Subjective interpretation is a pitfall in non-controlled observation.
- Unforeseen factors can interfere with the observation.
- Some people are rarely accessible to direct observation, creating obstacles.
Types of Observation:
Structured vs. Unstructured Observation:
Structured observation involves a careful definition of units to be observed, a standardized style of recording, and selection of pertinent data. It is appropriate for descriptive studies.
Unstructured observation Occurs without pre-arranged characteristics, aiming for a spontaneous picture of life and individuals. It is more likely used in exploratory studies..
Participant vs. Non-participant Observation:
Participant observation: The researcher actively participates in the group's activities, observing and recording natural behavior, and sometimes gaining information that would otherwise be unobtainable. It's predominantly a qualitative research design.
Non-participant observation: The researcher remains a passive observer, watching and listening without involvement.
Disguised observation: The researcher's presence is unknown to the people being observed.
Controlled vs. Uncontrolled Observation:
Uncontrolled observation: Takes place in a natural setting. Its aim is to get a spontaneous picture of life and persons. The main pitfall is subjective interpretation.
Controlled observation: Takes place according to pre-arranged plans, often involving experimental procedures and the use of mechanical or precision instruments for accuracy and standardization. It tends to supply formalized data for generalizations.
2. Interview Method
An interview is a person-to-person interaction with a specific purpose in mind, where an interviewer asks questions and records the responses. This interaction can be face-to-face or conducted via telephone or other electronic media.
Personal Interviews:
An interviewer asks questions in a face-to-face contact. This can be a direct personal investigation, where the interviewer collects information personally from the sources, or an indirect oral investigation, where the interviewer cross-examines other people who have knowledge about the problem.
Direct investigation is suitable for intensive studies, while indirect investigation is often used by government-appointed commissions and committees.
Structured Interviews: Utilizes a predetermined set of questions with highly standardized recording techniques. The interviewer follows a rigid procedure, asking questions in a prescribed form and order. Questions can be either open-ended or closed, as specified in the interview schedule. This method is economical and provides a safe basis for generalization.
Unstructured Interviews: are characterized by flexibility in approach to questioning, content, and structure. The interviewer has considerable freedom to ask supplementary questions, omit some questions, or change their sequence as needed. This flexibility makes comparing interviews difficult and analysis more time-consuming. It is a central technique in exploratory or formulative research studies.
Other Types of Interviews:
Focussed interview: Focuses on a given experience of the respondent and its effects. The interviewer has the freedom to decide the manner and sequence of questions and to explore reasons and motives.
Clinical interview: These interviews are concerned with broad underlying feelings or motivations or the course of an individual's life experience.
Non-directive interview: The interviewer encourages the respondent to talk about a topic with minimal direct questioning, acting as a catalyst.
Merits of the Interview Method:
More information and that too in greater depth can be obtained.
The interviewer, through their skill, can overcome resistance from respondents, and the interview method can yield an almost perfect sample of the general population.
There is greater flexibility, especially in unstructured interviews, as there is always an opportunity to restructure questions.
The observation method can also be applied to recording verbal answers to various questions.
Personal information can be obtained easily under this method.
Non-response generally remains very low.
The language of the interview can be adapted to the respondent's ability or educational level, avoiding misinterpretations of questions.
The interviewer can collect supplementary information about the respondent’s personal characteristics and environment, which is often valuable for interpreting results.
Demerits of the Interview Method:
It is very expensive. This is especially true when a large and geographically spread sample is taken.
There is a possibility of bias from both the interviewer and the respondent.
Certain types of respondents, such as important officials, executives, or high-income individuals, may not be easily approachable, potentially leading to inadequate data.
This method is relatively more time-consuming, especially with large samples and when follow-up calls are necessary.
The presence of the interviewer on the spot may over-stimulate the respondent, sometimes leading to imaginary information to make the interview more interesting.
The organization required for selecting, training, and supervising field staff is more complex with formidable problems.
Interviewing can introduce systematic errors.
Effective interviewing presupposes proper rapport with respondents to facilitate free and frank responses, which is often a difficult requirement.
The researcher may introduce their own bias in framing questions and interpreting responses.
Pre-requisites for Interviewing:
Interviewers should be carefully selected, trained, and briefed. They must be honest, sincere, hardworking, impartial, and have technical competence and practical experience. Field checks are necessary to ensure they are not deviating from instructions.
Telephone Interviews:
Telephone interviews involve contacting respondents via telephone to collect information. While not the most widely used method, they play an important role, particularly in industrial surveys in developed regions.
Merits of Telephone Interviews:
- More flexible than mailing methods.
- Faster way of obtaining information, making it a quick method.
- Cheaper than personal interviewing methods, with a relatively low cost per response.
- Callbacks are simple and economical.
- Higher rate of response than mailing methods, with non-response generally remaining very low.
- Replies can be recorded without causing embarrassment to respondents.
- The interviewer can explain requirements more easily to the respondent.
- Access can be gained to respondents who otherwise cannot be contacted for various reasons.
- No field staff is required for direct, in-person interviews.
- Representative and wider distribution of the sample is possible.
- It can be a useful tool for follow-up to questionnaires.
Demerits of Telephone Interviews:
- Respondents have little time for considered answers, as the interview period is typically short, often not exceeding five minutes.
- Surveys are restricted to respondents who have telephone facilities, excluding those without access.
- Extensive geographical coverage may be restricted by cost considerations.
- Not suitable for intensive surveys where comprehensive answers are required for various questions.
- The possibility of interviewer bias is relatively more.
- Questions have to be short and to the point, and probes are difficult to handle effectively.