To answer a research question, if numbers and statistics are required to explain, describe or quantify one or more phenomena, turn to quantitative research methods. Use data collection techniques such as surveys or polls.

Data collection in quantitative research

Surveys and polls are data collection methods used in quantitative research. For example, you may want to conduct a survey to determine the level of access a community’s population has to traditional food. To find out the exact number of people who consume this diet, a questionnaire can be sent to all households in the community.

Likewise, polls can assist in finding out the opinion of a population on a specific topic. To survey large groups, statistical tools and personnel are usually needed. Some companies specialize in this kind of research. However, even with limited resources it is possible to conduct a poll of your own. For example, you could go to every dwelling in a community and ask very specific questions on a predetermined topic, then compile the responses to obtain accurate data. It is best to have very few questions (one or two) and they should be simple and direct to leave no room for different interpretations.

Thus, to find out the level of access to traditional food, you could ask: Have you eaten wild meat in the past month? If yes, how many times?

Can quantitative research and qualitative research go hand in hand?

Conducting interviews with one or more individuals who do have access to traditional food and with others who do not (qualitative research), can help delve further into the question, but it will not answer the initial question. Interviews can provide a way to understand how people who consume traditional food stock up and explain why others do not have access. Qualitative and quantitative research are not in opposition. They are merely two distinct ways to get different types of information. Quantitative research aims to know, while qualitative research is more interested in understanding phenomenon.


There is a whole science to generating samples, especially when it comes to polls. For instance, when the aim is to predict an election outcome, it is important the number of respondents be high enough for results to be generalizable to the entire population. Obviously, not every elector from Quebec or Canada can be surveyed. A sample is required, a specific number of people deemed “representative” of the entire country. When creating such a sample, ensure the people identified represent in a proportional way the characteristics of the population being surveyed. For example, this could mean asking an equal number of men and women, seniors and youth, people of varied economic status, etc. The characteristics chosen will depend on the population being surveyed. This will further refine result analysis allowing, for example, to give accurate statistics on the sample’s subsets contacted (x women think the following; while respondents older than 50 think something else).

Designing a Questionnaire

A questionnaire can be a written form that all respondents fill out, thereby ensuring all research participants answer the same questions. Statistics can be drawn from these responses because they are comparable and then tallied up. Questions should be chosen with care and be based on the research question. For example: Do you want a new school in the community? Yes or No? Are you willing to pay a tax to build and maintain it? Yes or No?

By counting all the Yes responses, you will know how many people want a new school. By counting all the No responses, you will know how many do not. The response to the second question will indicate if people are willing or not to pay for the new school.

Administering a questionnaire

There are several ways to administer (to give) a questionnaire: mailing it out with a deadline to respond (to ensure participation, a prepaid response envelope should be included with the package); surveying by telephone; distributing questionnaires door-to-door with a time specified for pick up; conducting them in person, dwelling by dwelling; or even going to an event or place to reach a large number of people with the characteristics of the survey population. Each method has its pros and cons: mailings are more uncertain and expensive, but less time consuming; conducting them in person at each dwelling and writing the answers yourself ensure the desired response rate, but does not respect anonymity; distributing them in person to each dwelling, letting people fill out the questionnaire, and then returning to pick up questionnaires in sealed envelopes, are all very time consuming, but would ensure a better response rate.

Remember to come up with a system to ensure anonymity by numbering the questionnaires or use any other method to achieve this.

Data Compilation

The objective of quantitative research is to measure, usually with numbers and statistics. There is software to accomplish this and to input all data. But these tools are probably not within the reach of most community researchers. In a small community, it is possible to count survey or poll responses without computer assistance. But as stated above, it is important to limit the number of questions so that analysis is easier afterwards.

When embarking on quantitative research, unless it concerns a small in-house survey, it is best to consult specialized technicians. Not only when creating the sample, but consult them also on software data compilation. This will ensure the rigour and validity of responses.

Assess the resources required to successfully complete the survey (technical, time, funding);

Ensure the sample is representative (by identifying the criteria to describe the population to be surveyed) or aim to survey all community members;

Think about the best ways to obtain a good response rate;

Count all responses manually or with specialized software;

Ensure the confidentiality of responses.