Data Collection Methods
Collection of Data Through Questionnaires
This popular method involves sending a questionnaire (a list of questions printed in a definite order) to people with a request to answer and return it. Respondents are expected to read, understand, and write down their replies in the provided space on their own.
Merits of Questionnaires:
Low cost, even for a large and widely spread universe.
Free from interviewer bias; answers are in the respondent's own words.
Respondents have adequate time to give well-thought-out answers.
Respondents who are not easily approachable can be reached conveniently.
Large samples can be used, making results more dependable and reliable.
Demerits of Questionnaires:
Low rate of return; bias due to non-response is often indeterminate.
Can only be used with educated and cooperating respondents.
Control over the questionnaire is lost once it is sent.
Inflexible, as the approach cannot be amended after dispatch.
Possibility of ambiguous or omitted replies.
Difficult to know if willing respondents are truly representative.
Likely to be the slowest method.
Main Aspects of a Questionnaire:
A questionnaire should be very carefully constructed.
General Form: Can be structured (definite, concrete, pre-determined questions in the same order) or unstructured. The form of questions can be closed (yes/no) or open (free response).
Question Sequence: A proper sequence reduces the chance of misunderstanding. It should be clear and smooth-moving, with easier questions at the beginning to arouse interest and cooperation. Avoid opening with questions that strain memory, are of a personal character, or relate to personal wealth.
Question Formulation and Wording: Each question must be very clear and impartial. Questions should be constructed as part of a well-thought-out tabulation plan. They should be simple, concrete, and conform to the respondent's way of thinking.
Essentials of a Good Questionnaire:
Should be comparatively short and simple.
Questions should proceed in a logical sequence from easy to more difficult.
Personal and intimate questions should be left to the end.
Avoid technical terms and vague expressions.
Questions can be dichotomous, multiple-choice, or open-ended (the latter should be used sparingly as they are hard to analyze).
Should include some control questions to check the reliability of the respondent.
Collection of Data Through Schedules
This method of data collection is very similar to collecting data through a questionnaire. The main difference is that schedules, which are proformas containing a set of questions, are filled out by specially appointed enumerators.
These enumerators visit respondents, ask the questions from the proforma in the order they are listed, and record the replies in the designated space.
In some cases, schedules may be given to respondents, and the enumerators assist them in recording their answers. Enumerators also explain the goals of the investigation and clarify any difficulties a respondent might have with a particular question or definition.
This method requires the careful selection and thorough training of enumerators so they understand the nature and scope of the investigation. Enumerators should be intelligent, capable of cross-examination to find the truth, honest, sincere, hardworking, and possess patience and perseverance.
This method is highly useful for extensive inquiries and can produce reliable results. However, it is very expensive and is typically used in investigations by governmental agencies or large organizations. The population census around the world is conducted using this method.
Key aspects of this method:
Enumerator Selection: Enumerators must be selected very carefully. They should be intelligent, possess the ability to cross-examine to find the truth, and be honest, sincere, hardworking, patient, and persevering.
Training: They must be well-trained and have the nature and scope of the investigation thoroughly explained to them.
Usefulness: This method is highly useful for extensive inquiries and can produce reliable results.
Cost: It is a very expensive method and is typically used for investigations conducted by government agencies or large organizations. The worldwide population census is conducted using this method.
Difference Between Questionnaires and Schedules
Administration: A questionnaire is usually mailed to informants to be answered without assistance. A schedule is filled out by a research worker or enumerator who can interpret questions if necessary.
Cost: Collecting data via questionnaire is relatively cheap and economical, as costs are mainly for preparation and mailing. Data collection through schedules is more expensive due to the need to appoint and train enumerators.
Response Rate: Non-response is often high for questionnaires. In contrast, non-response is generally very low for schedules since they are filled by enumerators. However, with schedules, there is a risk of interviewer bias and cheating.
Respondent Identity: With a questionnaire, it's not always clear who is replying. With a schedule, the respondent's identity is known.
Speed: The questionnaire method is often slow because many respondents do not return them in a timely manner. Information is collected promptly with schedules as they are filled by enumerators.
Personal Contact: Personal contact is generally not possible with the questionnaire method. Direct personal contact is established when using schedules.
Respondent Requirements: The questionnaire method can only be used with literate and cooperative respondents. Schedules can be used to gather information even if the respondents are illiterate.
Sample Distribution: A wider and more representative sample distribution is possible with the questionnaire method. It can be difficult to send enumerators over a wide area for schedules.
Accuracy: The risk of collecting incomplete and wrong information is higher with the questionnaire method. Information collected through schedules is generally more complete and accurate because enumerators can clarify questions for respondents.
Success Factor: The success of the questionnaire method depends more on the quality of the questionnaire itself. For schedules, success relies heavily on the honesty and competence of the enumerators.
Appearance: The physical appearance of a questionnaire must be attractive to get the respondent's attention. This is not as important for schedules, as they are filled in by enumerators.
Observation: The observation method can be used in conjunction with schedules, but this is not possible when collecting data via questionnaires.
Other Methods of Data Collection
Several other methods are used for data collection, particularly by large business houses.
Warranty Cards: These are usually postal-sized cards that dealers of consumer durables use to collect product information. The card, which contains questions, is placed inside the product package with a request for the consumer to fill it out and mail it back.
Distributor or Store Audits: These are performed by distributors and manufacturers through their salesmen at regular intervals to estimate market size, market share, and seasonal purchasing patterns. Data is obtained through observation or by copying store records, not by questioning. This method is efficient for evaluating the effect of in-store promotions on sales.
Pantry Audits: This technique estimates the consumption of goods at the consumer level. An investigator inventories the types, quantities, and prices of commodities in a consumer's pantry. The goal is often to determine what types of consumers buy certain products and brands, assuming the pantry's contents accurately reflect preferences. These audits are often supplemented with direct questioning about purchasing habits.
Consumer Panels: This is an extension of the pantry audit, where a set of consumers agrees to keep detailed daily records of their consumption, which are made available to the investigator.
Transitory Consumer Panel: This type is set up to measure the effect of a specific phenomenon, often on a before-and-after basis. It is a popular tool in advertising and social research.
Continuing Consumer Panel: This panel is set up for an indefinite period to collect data on a particular aspect of consumer behavior over time.
Use of Mechanical Devices: Devices like eye cameras, pupilometric cameras, psychogalvanometers, motion picture cameras, and audiometers are widely used to collect information indirectly, mostly in developed countries by large businesses.
Projective Techniques: These indirect interviewing techniques were developed by psychologists to infer underlying motives, urges, or intentions that respondents may be unable or unwilling to reveal.
Word Association Tests: A respondent is asked to say the first word that comes to mind after hearing a word from a list. This is frequently used in advertising research.
Sentence Completion Tests: An extension of word association, where an informant completes a sentence to reveal their attitudes. This technique helps in developing hypotheses and questionnaires.
Story Completion Tests: A researcher provides part of a story and asks the informant to complete it.
Verbal Projection Tests: A respondent is asked to explain what other people do, which may reveal their own motivations.
Depth Interviews: These interviews are designed to uncover underlying motives and desires and are often used in motivational research. They require great skill and considerable time on the part of the interviewer.
Content Analysis: This involves analyzing the contents of documentary materials like books, magazines, and newspapers, as well as other spoken or printed verbal materials. Prior to the 1940s, this was mostly quantitative, but since the 1950s, it has become more qualitative, focusing on the general message of the documents.
Collection of Secondary Data
Secondary data is data that is already available because it has been collected and analyzed by someone else. When a researcher uses secondary data, they must look into various sources to obtain it. This data can be published or unpublished.
Published data can be found in:
Publications of central, state, and local governments.
Publications of foreign governments or international bodies.
Technical and trade journals.
Books, magazines, and newspapers.
Reports from associations related to business and industry, banks, etc.
Reports prepared by research scholars, universities, and economists.
Public records, statistics, and historical documents.
Unpublished data sources include diaries, letters, unpublished biographies, and may be available from scholars, research workers, trade associations, and other organizations.
Researchers must be very careful when using secondary data, as it might be unsuitable or inadequate for their specific study. It is never safe to take published statistics at face value without understanding their meaning and limitations. Before using secondary data, a researcher must ensure it has the following characteristics:
Reliability of Data: This can be tested by finding out who collected the data, the sources, if proper methods were used, when it was collected, if there was any compiler bias, and the level of accuracy desired and achieved.
Suitability of Data: Data that is suitable for one inquiry may not be for another. The researcher must carefully scrutinize the definitions of terms and units of collection used in the original source. The object, scope, and nature of the original inquiry must also be studied.
Adequacy of Data: If the level of accuracy is inadequate for the present inquiry, the data should not be used. Data is also considered inadequate if it relates to an area that is narrower or wider than the area of the current study.
It is risky to use already available data, but it should not be discarded if it is reliable, suitable, adequate, and from an authentic source, as using it can save time and energy.