Different data collection methods and their credibility

By Abhinash on November 29, 2013

Data collection sources in research can be of two types, primary and secondary data. Data can be collected through either of the two or both methods. After defining the research problem, the types of data collection methods should be decided by the researcher:

Primary data collection

Primary data is the data obtained from the organization or the institution that originally collected the information. It involves doing studies with your target audience to find out how they feel about a particular issue. Similarly, Hollerson (2003) defines it as ‘information collected first hand, generally by original research tailor-made to answer specific research questions. It can be of the following types:

  • Market surveys.
  • Focus groups.
  • Interviews.
  • Observation methods.

Secondary data collection

Secondary research is defined as getting the information the researcher needs from sources that have done their own research and either offer it to him for a fee or for free. Thus, the information is obtained indirectly from the market. The disadvantages of this type of research are that they may not be the best questions nor asked the best people. The benefits, on the other hand, are that the information can be obtained faster. They can be collected from the following sources:

  • Journals and articles.
  • Books.
  • Magazines.
  • Newspapers.
  • Online web portals.
  • Government agencies.
  • Annual reports.
  • Independent agencies.
  • Government official reports.

For a study, one may use the survey method of primary data if the data is quantitative in nature. It is most suitable for this type of study because of its nature. On the other hand, one may refer to numerous forms of secondary research, most of which can be used in the literature.

Data collection through a questionnaire

A questionnaire is the main type of data-collection instrument in descriptive-research designs. It is defined as “a set of questions for obtaining statistically useful personal information from individuals”. The questions asked should be aimed at gaining specific answers by minimizing the possibility that respondents will give inaccurate answers. Questionnaires can be of different types, described as under:

  1. Open format questionnaire: Questionnaires with open-ended questions where the researcher seeks a comprehensive description of the subject matter.
  2. Closed format questions: Questionnaires contain limited options for the interviewee to choose from and are best in the case of quantitative studies.
  3. Leading questions: Questionnaires containing questions that lead the respondents towards a particular answer or in other words, they are forced to select that option. They are, thus, biased.
  4. Importance questions: Questionnaires containing questions that require the respondent to rate options on a scale of 1-6.
  5. Likert questions: Questionnaires developed on a Likert scale, i.e. statements that give the respondent s five categories- options like strongly agree, somewhat agree, neutral, somewhat disagree, and strongly disagree.
  6. Dichotomous questions: Questionnaire containing questions that give respondents yes or no type questions.
  7. Bipolar questions: Questionnaires containing questions that options are extreme opposites of each other.

The credibility of research findings

  • Validity: Validity concerns the extent to which the test tests what it is supposed to test. Validity is important because it enables one to have some confidence that the measure taken is close to the true measure.
  • Reliability: Reliability is essentially concerned with the extent to which assessment procedures give consistent results. Reliability concerns the degree of confidence that can be placed in the results and the data, which is often a matter of statistical calculation and subsequent test redesigning.

I am currently working as a Research Associate. My work is centered on Macroeconomics with modern econometric approach. Broadly, the methodological research focuses on Panel data and Times series data analysis for causal inference and prediction. I also served as a reviewer to Journals of Taylor & Francis Group, Emerald, Sage.


1 thought on “Different data collection methods and their credibility”