Data Collection Methods

Data collection forms the ‘input’ step of all computing processes: Input, Process and Output. Therefore, it is important for any database analyst to know about data collection and ensure that the right data is collected. Else she will only confirm the old saying: garbage in, garbage out.

Watch the video below to learn about the many facets of data gathering.

Data Collection Methods
Data can tell a story

Data Collection Methods

Why collect right data

Today organizations need data about their clients, customers, users, employees, vendors and competitors. Using this data they can improve most aspects of their operations.
Although data can be valuable, wrong data is useless. The right data collection method can give useful insights. No database management tool can give right results with wrong data.

How to Collect Right Data
Organizations have several tools at their disposal for primary data collection. The methods range from traditional face-to-face interview to more sophisticated ways to collect data.

Top Six Primary Data Collection Methods
Questionnaires and surveys
Documents and records
Focus groups
Oral histories


The key to efficient interviews is knowing what to ask because in-person interviewing can be the most expensive.
It can be worth the cost, as the interviewer can tailor follow-up questions.


Observation involves collecting information without asking questions.
For example, if a study involves the number of people in a pub at a given time, the data should be reasonably reliable.
Observation can determine the dynamics of a situation, which generally cannot be measured through other data collection techniques. It can also be combined with a video recording.

Documents and Records

Document- and records-based research uses existing data for a study. Attendance records, meeting minutes, and financial records are examples of this type of data.
Using documents and records can be efficient and inexpensive because you’re using secondary data. However, since the researcher has less control over the results, documents and records can be an incomplete data source.

Focus Groups

A focus group is a research method that brings together 10-12 people in a room to provide feedback regarding a product, service or concept after a presentation. A trained moderator leads a 30-60-minute discussion within the group. This adds a collective element to individual data collection.

Oral Histories

An oral history is defined as the recording, preservation, and interpretation of historical information based on the opinions and personal experiences of people who were involved in the events.
For example, we may be interested in studying the effect of a tsunami on a community.
The researcher can become an extra, unintended variable that can distort the results by introducing bias.

Questionnaires & Surveys

To be meaningful, surveys and questionnaires must be carefully planned otherwise they won’t be valuable.
Dynamic online forms are a modern and effective way to conduct surveys also on mobile devices.
For instance, when someone answers no to a question about home loans, they won’t be shown the related follow-up questions about such loans. Instead, they’ll go immediately to a question on another topic.

Qualitative vs Quantitative Data Collection Methods
Some methods are quantitative: something can be counted.
Questionnaires, surveys, and documents and records are quantitative.
Others are qualitative – they consider factors like feelings or thoughts of the participants.
Interviews, focus groups, observations and oral histories are qualitative.

Qualitative Data Collection
Qualitative methods assess factors like the thoughts and feelings of the research participants and answer the question ‘Why?’

Qualitative Research Methods
Three commonly used qualitative data collection methods are : ethnographic, theory grounded and phenomenological.

Ethnography comes from anthropology, the study of human societies and cultures. It seeks to understand how people live. The researchers observe the participants in a non-directed way. They use these observations to understand the why behind their desires, decisions, or behaviors.

Grounded theory
Grounded theory uses the following methods:
Participant observation: Researchers immerse themselves in the daily lives of subjects.

Interviews: These can vary from informal chats to structured interviews.

Document and artifact collection: Researchers can learn about a group of people from looking at materials the group used. For example, a local community’s laws may shed light on opinions and provide a clearer picture of residents’ sentiments.

Phenomenology describes how people experience certain events like a natural disaster. The understanding, focus and organization of data help researchers identify patterns, make connections and explain findings.

Qualitative Data Collection: Conclusion
Each of these qualitative data collection methods sheds light on factors that can be hidden in simple data analysis. Qualitative data is one way to add context and reality to raw numbers.

Quantitative Data Collection Methods
Quantitative methods such as surveys answer the question “How much?”
Therefore, quantitative data is measurable and expressed in numerical form.
Four Primary Quantitative Methods


An example of descriptive research is a study that collects and tabulates test scores. It uses charts and tables to illustrate results.
A descriptive approach is quantitative, but it can also be qualitative.


Correlational research seeks to collect data that shows relationships between different occurrences. A positive correlation is one in which two variables either increase or decrease at the same time. A negative correlation is when an increase in one variable means a decrease in another. Example: If all other factors remain the same, when the price of a good or service increases, the quantity of demand decreases, and vice versa.
There is also a zero-correlation result, in which the relationship between two variables is insignificant.


Experimental research uses statistics to determine the cause-and-effect relationship between variables. This method uses controls for all the crucial factors that could potentially affect the phenomena of interest.

Using this method, researchers randomly assign participants in an experiment to either the control or treatment groups as in drug testing.
Experimental methods are known for producing results that are applicable to the real world.


Quasi-experimental method doesn’t randomize assignment or sampling or both and therefore produce questionable results.

Application of Quantitative Methods

Large-scale quantitative surveys can be conducted online using a list of possible responses.
Quantitative interviews are typically conducted face to face, over the phone, or via the internet. They enable researchers to add some “why” to the “how much”.

Secondary Data Collection Methods

While primary data collection is an authoritative and authentic data collection method, there are several instances where secondary data collection methods can provide value.

Understanding Secondary Data Collection

Secondary data is another person’s original bank of knowledge which is available in a library, for example.
Second-hand data can add insight to a research project. Using secondary data is more efficient and less expensive than collecting primary data.

Sources of Secondary Data

Census Bureau
Small Business Administration
Local chambers of commerce
Publicly traded corporations
Educational institutions
Colleges and Universities
Value of Secondary Data

Using secondary data saves time and money. Combined with primary research, secondary data can help researchers better understand their subjects and prepare results more efficiently .

Sampling Methods in Data Collection

Sampling is the process of identifying a subset of a population that provides an accurate reflection on the whole. Statistical methods are available that make sure a small subset of the community accurately represents the whole group.
Sampling Methods

There are five generally accepted sampling methods.

Random Sampling

Randomness eliminates elements that can affect the validity of a study.
An example of a simple random sample would be the names of 120 students being chosen out of a box from a school of 1200 students. The sample is random because each student has an equal chance of being chosen.

Systematic Sampling

Systematic sampling follows a set of rules to create regularity in sampling. An example is interviewing every fifth customer.
Convenience Sampling

It’s the least reliable sampling method. Convenience sampling involves gathering information from whoever is easiest to reach. An example would be asking college friends in the same class a question about a service.
It can be effective to gain initial primary data on product colours, for example.

Clustered Sampling

In clustered sampling, a researcher uses subgroups of a population. For example, a researcher wants to survey academic performance of high school students in Delhi. He can divide the entire population of Delhi into different districts (clusters).
Clustered sampling can be of two types — single-stage cluster sampling, where all individuals in a cluster are included or two-stage cluster sampling, where only random individuals within the cluster are chosen.

Stratified Sampling

This data collection method involves dividing a population into subgroups that share similar characteristics.
For example, a study can break respondents down by gender or age or caste.
In stratified sampling, individuals are randomly selected from each strata. In cluster sampling, only certain clusters are used.

Further Reading:

Data Gathering Techniques