Sources of Data Collection

Quantitative and qualitative data

The source you use for data collection depends on the type of research you want to carry out. There are two types of research; quantitative and qualitative.

Quantitative data

Qualitative data

Primary Sources of Data Collection

A primary source of data collection refers to data that is collected first-hand for particular research purposes. This means the person who collects it is the originator of that data. This may either refer to yourself or others. For example, you could carry out your own research and data collection or use a primary source from another person. Primary data sources give a more personal insight into the information as they rely on a first-hand account.

Examples of Primary Sources of Data Collection

Some examples of primary sources of data collection are as follows:

Questionnaires

Questionnaires are sets of written questions that people can answer and provide information. They can be either qualitative or quantitative:

Quantitative questions may contain boxes to tick, yes/no ratings or scales (i.e. rate from 1 -5). With these types of questions, it is easier to present the data objectively as statistics, e.g. '70 percent of people who took this questionnaire answered yes to this question'.

Qualitative questions may include boxes for people to write down their thoughts. This will provide a more detailed answer that will be more subjective and personal.

Fig. 1 - Questionnaires can be quantitative or qualitative.

Interviews

An interview is a spoken or written conversation in which the researcher asks questions to another person. These are usually qualitative, as there is a focus on the language. Unlike questionnaires with predetermined questions, interviews are generally carried out with both people, so there is more freedom to ask a wider variety of questions. An interview is a good way to find out more about a topic or opinion from another point of view. The person answering the questions will be able to give a more in-depth answer and will also have more freedom to express their thoughts.

Interviews can be structured, semi-structured, or unstructured. Structured interviews provide more quantitative data that is easily comparable. In contrast, unstructured interviews allow for more qualitative and relevant data that is more difficult to compare as it is more subjective.

Autobiographies/memoirs/letters

Both autobiographies, memoirs, and personal letters recount personal experiences that are often unique to the writer. Because of this, they are seen as qualitative methods of research. You could use either of these as a source to gain a deeper understanding of someone's views or opinions on a particular topic.

Critical analysis

Critical analysis deals with the skilful examination and interpretation of data to form a deeper understanding and opinion of it. Examples of texts that can be critically analysed are magazines, newspapers, and speeches. This is a useful way to gain more knowledge of a topic in a detailed manner. Critical analyses are considered qualitative sources, as there is a strong focus on an in-depth analysis that relies on critical language, as opposed to numbers and statistics.

Observations

This is a way to collect data by watching participants or events in a natural setting. Observations can be either qualitative or quantitative:

Qualitative observations are subjective - they are used to gather more personal, in-depth information about a participant. There is more of a focus on the quality of the information given by the participants, as opposed to the number of participants. As a result, fewer participants are needed for qualitative observations.

Quantitative observations are objective, focusing on a broader statistical analysis of a population as a whole instead of an in-depth analysis of a participant. As a result, more participants are needed for quantitative observations to make the data unbiased and credible.

Fig. 2 - Observations can be qualitative or quantitative.

Primary vs secondary sources

Secondary sources refer to existing data that has already been collected by another researcher and used for another purpose. This can then be taken and used for research in another project or study.

Some more examples of secondary sources are:

Books

Dictionaries/encyclopaedias

Biographies

Characteristics of good Data Collection

How do you know if the data you have collected is good? A few characteristics of a decent data collection are as follows:

Credible

The data you collect must come from trustworthy sources. Credibility is important as it gives value to your claims and makes your work more believable. If your data is not credible, this could lead to inaccuracies in your analysis, or you could end up writing about things that are not correct. Consider the following things when collecting data from someone else:

Is the source outdated or recent?

Is the information biased? If so, will this bias negatively affect my analysis?

What is the purpose of the source?

Is the originator of the source knowledgeable on the topic?

Relevant

This may seem obvious, but the data you collect must relate to your topic. For example, if you are studying the phonetic features of native vs non-native speakers of English, your research should also focus on that subject.

Ethical

If you collect data that involves other people, ensure they consent to their data being used for your research. For example, if you conduct an interview, make sure the other people involved know what they are doing and where the information is going. Any data collected from a consenting person must be kept secure, and after use, it must then be deleted. Further, the names of people and places must be removed to maintain anonymity. If participants do not wish to continue, they have the right to stop the research at any time.

Varied

It would help if you had a wide enough variety of data in order to gain a good understanding of different perspectives. If you do not have enough data, this could lead to your study being biased, which could negatively affect your analysis. Instead of providing an overview of different perspectives, it could be more in favour of a particular view, which is misleading and may lead to false conclusions!