1. Introduction
Data contains raw facts or figures that a researcher captures, stores, manipulates or analyzes to discern some meaning or make a decision. Data is not important for its own sake but because it helps us find an answer to a research question. Researchers use two categories of data: primary and secondary data.
Hence, it’s important to know the definition, purpose, advantages, and drawbacks of primary and secondary data and understand the context in which we can use them.
In this tutorial, we’ll explain the difference between these two data types.
2. Understanding Primary Data
Primary data represents raw findings from first-hand fieldwork, questionnaires, interview transcripts, focus groups, observational studies, or experimental data. It’s unfiltered and unprocessed, so it’s in the form in which it was recorded.
Primary data is the foundational material from which researchers build theories, answer questions, and formulate hypotheses. So, gathering primary data is the first step of many research methodologies.
2.1. Example
Let’s say a company is market-testing a mobile app. It invited several users from a test group to use the app so they could provide feedback on improving features or usability. For example, the company may want to find out things such as:Â
- how long does it take for users to complete the tasks they want
- whether there are missing features or UI elements to add
- which functionality are users most drawn to, and why
During the test sessions, participants complete the questionnaire, and the app records all the interactions with the user. So, we know the exact timesteps at which the users performed any action, such as entering data or clicking on a menu item. Additionally, we have their textual responses to the questions from our survey. In this example, the users’ textual responses we get from usability testing constitute our primary data.
2.2. Collecting Primary Data
Surveys are one of the most common and popular methods of collecting primary data. They are structured queries about people’s attitudes, experiences, or behavior.
Further, researchers can interview study participants to get data. Interviews may be in person, over the phone, or even online. Because of their interactive nature, interviews can provide more details than surveys, often revealing things that a survey can miss.
Another method is observational study. Observational studies collect data on events, interactions, or behaviors as they occur spontaneously in nature or society. For example, the researchers can conduct an observational usability study. They can track users’ activity through the app to note where they get stuck or confused. Observations bring researchers closer to understanding the inner life of social, cultural, or ecological systems. These observations enable the rhythms, deviations, and connections that quantitative methods may not reveal.
Finally, experimental designs allow researchers to systematically manipulate (or ‘test’) independent variables by random assignment. They observe the effects on dependent variables and model these effects in controlled conditions.
For example, when designing a mobile app, researchers might want to measure and test the usability of the app interface against an alternative one. They could randomly assign users to either use Design A or Design B and compare the results in terms of specific measures and indicators of usability. Examples include time to complete a task, error rate, and level of satisfaction. In so doing, they can isolate the effect of the interface design on the usability outcomes.
3. Understanding Secondary Data
Secondary data refers to the data previously gathered, organized, and stored by another individual or organization. Secondary data can also be defined as data derived from primary data. For example, raw recordings of interviews represent the primary data, and the transcripts derived from them are the secondary data:
Secondary data can contain many items without a clear structure, as the data can come from various internal and external databases, published works, non-published documents, maps, photographs, videos, and so forth. So, a researcher first has to organize all the data into a coherent structure suitable for answering the specified research question.
3.1. Secondary Data Sources
Published books contain a substantial body of secondary data. They usually contain references to other books and articles with data that can be relevant to our research question.
In addition to published sources, researchers can look into unpublished sources, which allow them to obtain information that is not readily published. These sources can be found in government agencies, non-profits, or private research institutes and cover a wide range of highly focused topics.
Organizations also generate terabytes of internal data—financial data, customer data, corporate performance metrics, employee surveys, etc. Internal sources often contain private data that can reveal insights about organizational processes, market shifts, or consumer tastes.
External data sources are all data sources compiled in another institution or organization than ours. These include data produced by national and local government agencies, statistical bureaus, international organizations, research consortia, and others.
4. Comparative Analysis
Let’s compare the two data types:
| Aspect | Primary data | Secondary data |
| Control over the data collection process | The researcher controls data collection, from method selection and interviewing to measurement scheme details | The researchers using the secondary data don’t have control over the questionnaire, interview protocol, etc. |
| Time and resource constraints | This process is both time-consuming and resource-intensive | Although it is still more cost-effective than collecting primary data, it also requires an investment of time and effort |
| Ethical implications | The protection of confidentiality and privacy arises from direct contact between the researcher and the human participants during the collection of sensitive or personal data | The researchers might have ethical dilemmas concerning issues of privacy and confidentiality, issues of ownership, intellectual property rights, etc. |
| Time-sensitive |
Primary data collection can provide researchers with real-time data about the phenomenon under investigation | Researchers must consider whether any changes over time might influence the interpretation of the data |
While primary data offers researchers complete control over the type, volume, and style of data collected, collecting it from scratch may be costly.
On the other hand, secondary data, which is relatively inexpensive and easier to get, poses ethical and methodological issues that we need to consider. The researcher’s objectives and purposes will determine the choice of research method.
5. Conclusion
In this article, we compare primary and secondary data. The former helps in analysis with more precision and detail but demands more time and resources for collecting. In contrast, secondary data provides some advantages over primary data because researchers don’t have to gather them. This eliminates the costs of time and money needed to gather primary data. However, researchers have to be careful. They need to ensure that the secondary data they use are relevant to their work and that they have been collected properly and ethically.