Category: Practical Research 2

Quantitative research is a type of empirical research that uses numerical data to study a phenomenon or a problem. It is a systematic approach that involves the collection, analysis, and interpretation of data using statistical methods. This method of research is often used in the natural sciences, social sciences, and health sciences, and is characterized by its focus on measurement, objectivity, and generalizability.

One of the primary characteristics of quantitative research is its reliance on numerical data. This data can take many forms, including survey responses, laboratory measurements, or secondary data sources. The use of numerical data allows researchers to make precise and accurate measurements of variables, which can be analyzed using statistical techniques.

Another characteristic of quantitative research is its emphasis on objectivity. Unlike qualitative research, which is often subjective and interpretive, quantitative research seeks to minimize the influence of the researcher’s own biases and assumptions. This is achieved through the use of standardized methods and procedures for data collection, analysis, and interpretation.

Generalizability is another important characteristic of quantitative research. The goal of quantitative research is to draw conclusions that can be applied to a larger population beyond the specific sample being studied. This is achieved through the use of random sampling techniques, which ensure that the sample is representative of the larger population.

The process of conducting quantitative research can be broken down into several key stages, including:

Identifying the research question: This involves identifying a research problem or question that can be addressed using quantitative methods. The research question should be specific, measurable, and relevant to the field of study.

Developing the research design: This involves determining the appropriate research design and methodology for the study. This may include selecting a sample, determining the variables to be measured, and identifying appropriate data collection methods.

Collecting the data: This involves collecting the numerical data necessary to address the research question. This may involve administering surveys, conducting laboratory experiments, or collecting data from existing sources.

Analyzing the data: This involves using statistical methods to analyze the data and identify patterns, relationships, and trends. This may include descriptive statistics, such as means and standard deviations, as well as inferential statistics, such as correlation and regression analyses.

Interpreting the results: This involves drawing conclusions based on the data analysis and interpreting the findings in the context of the research question. This may involve comparing the results to previous research, discussing limitations and implications, and identifying areas for future research.

One of the key advantages of quantitative research is its ability to provide precise and reliable data. The use of standardized methods and procedures for data collection and analysis ensures that the data is objective and can be replicated by other researchers. This also allows for the comparison of results across different studies, which can help to establish patterns and relationships within the field of study.

Another advantage of quantitative research is its ability to identify cause-and-effect relationships. By manipulating one or more variables and measuring the effect on another variable, researchers can establish causal relationships between variables. This can be useful in developing interventions and treatments for various problems and conditions.

However, there are also some limitations to quantitative research. One limitation is its reliance on numerical data, which may not capture the full complexity and richness of human experience. This can be particularly true in fields such as psychology and sociology, where subjective experiences and emotions may be difficult to quantify.

Another limitation is the potential for researcher bias. While the use of standardized methods and procedures can minimize bias, researchers may still have their own biases and assumptions that can influence the study design and data interpretation.

In conclusion, quantitative research is an important research method used in various fields of study, including natural sciences, social sciences, and health sciences. Its emphasis on numerical data, objectivity, and generalizability make it a powerful tool for generating precise and reliable data that can inform decision-making and contribute to the advancement of knowledge.

While there are limitations to quantitative research, such as its reliance on numerical data and the potential for researcher bias, these can be mitigated through careful study design, data collection, and analysis. Additionally, the ability of quantitative research to identify cause-and-effect relationships can be particularly valuable in developing interventions and treatments for various problems and conditions.

Overall, quantitative research plays an important role in advancing our understanding of the world around us and has contributed to significant breakthroughs in various fields of study. Its systematic and rigorous approach provides a foundation for generating reliable and trustworthy knowledge, and will continue to be an essential tool for researchers in the future.

Types of Quantitative Research Method

There are two major types of quantitative research method, namely:

1) primary quantitative research method and 

2) secondary quantitative research method

On the one hand, primary quantitative research method focuses on the collection of data directly from the respondents themselves. On the other hand, as the name suggests, secondary quantitative research method involves using already existing data from previous researches or from the internet, libraries, government records, and other public sources.

Primary Quantitative Research Method

Primary quantitative research method is further divided into three sub-types, namely:

1) in terms of techniques and types of study, 

2) data collection methodologies, and 

3) data analysis techniques.

Under techniques and types of study we have 

1) survey research, 

2) correlational research, 

3) causal-comparative research, and 

4) experimental research

Under data collection methodologies we have 

1) sampling method and 

2) surveys and polls.

Under data analysis techniques we have 

1) SWOT analysis, 

2) conjoint analysis, 

3) cross-tabulation, and 

4) TURF analysis

Let us now briefly discuss each. Please note that our intention here is just to provide readers with a basic understanding of each concept. A thorough and exhaustive discussion on each concept will be done in a separate article.

Under Techniques and Types of Study

Survey

Survey research involves the collection of information from a sample of respondents through their answers to a set of questions. Survey research employs a variety of methods in collecting data, such as the use of questionnaires, online polls, online surveys, and other types of surveys.

As we can see, in survey research, researchers choose a sample of respondents from a population and administer a standardized questionnaire. The questionnaire can be a face-to-face interview, telephone interview, or a written document completed by the respondent being surveyed.

There are two types of survey, namely:

1) cross-sectional surveys and 

2) longitudinal surveys

Cross-sectional survey refers to the collection of data from a sample target of population at one specific point in time. The participants in the study are selected based on particular variable of interest. Thus, participants in cross-sectional studies are chosen based on the inclusion and exclusion criteria set by the study. For instance, researchers who are studying consumer traits may select a group of people who differ in income but study them at one point in time. By doing this, any differences in income can presumably be attributed to differences in consumption habit rather than something that happened over time.

Like cross-sectional survey, longitudinal survey is observational. However, unlike cross-sectional survey which collects data from a sample target of population at one specific point in time, longitudinal survey is conducted across various time duration. Here, researchers conduct several observations of the same subject over a period of time, sometimes lasting days, weeks, months, years, or even decades. For example, researchers might want to look at the change in blood sugar levels among men over 40 who meditate daily for 10 minutes for a period of 2 years.

Hence, because longitudinal studies go beyond a single moment in time, then this type of survey is more likely to suggest a cause-and-effect relationship between variables.

Correlational Research

According to Adi Bhat, “Correlational research is a type of non-experimental research method, in which a researcher measures two variables, understands and assess the statistical relationship between them with no influence from any extraneous variable.” See Adi Bhat, “Correlational Research: Definition with Example”, QuestionPro, available from https://www.questionpro.com/blog/correlational-research/.

It is important to note that unlike in experimental research where the researcher manipulates some variables and then measure the effects of this manipulation on other variables, in correlational research design, the researcher does not influence any variables but only measure them and look for relations (that is, correlations) between some set of variables.

As we can see, in correlational research, the researcher may want to know or discover more about the relationship between and among variables. This is important for many reasons. For one, the knowledge about the relationship between and among variables may help address a particular problem or clarify some issues or even make some predictions. For instance, we may want to correlate money with happiness. But why the need to know the correlation between money and happiness? Maybe we want to test whether the cliché “Money can’t buy happiness” is true or not.

Causal-Comparative Research

Causal-comparative research is also called quasi-experimental research.

Causal-comparative research design depends mainly on the factor of comparison. Researchers used this quantitative research design in order to establish cause and effect relationship between two or more variables, where one variable is dependent on the other. However, it must be noted that in observing the impact of the independent variable on the dependent variable, the independent variable should be established but not manipulated.

It is also important to note that the attempt to find or determine the relationships between independent and dependent variables in causal-comparative research design should be done after the action or event has occurred. As already hinted above, the main goal of a causal-comparative research design is to determine whether the independent variable affects (directly or indirectly) the outcome or the dependent variable. And, as we can see, this is done by comparing two or more groups of individuals.

There are three most commonly used methods in causal-comparative research, namely, the chi-square test, the paired samples and independent tests, and analysis of variance (ANOVA) or ANCOVA. Please note that PHILO-notes will discuss each of this method in a separate article.

Experimental Research

Experimental research is a type of quantitative research method where the researchers examine the effect of the independent variable on the dependent variable.

Experimental research is conducted to test theories or construct theoretical explanations about a particular phenomenon under investigation.  In some cases, researchers conduct experiments to discover facts that will support in making decisions, especially when there is not enough data or information to support such decisions.

In experimental research, researchers manipulate one variable and control the rest of the variables. As we may already know, the manipulated variable is the independent variable and it is called “the manipulated variable” because this is the variable that researchers change or alter in the experiment. It is important to note that in experimental research, the researchers should only have one manipulated variable at a time.

It is also important to note that there are three main variables involved in experimental research, namely, the independent variable, the dependent variable, and the controlled variable. Again, the independent variable is the variable that is altered or changed during the experiment; the dependent variable is the variable that is being tested or measured in the experiment; and the controlled variable is the variable that is kept constant during the experiment. The controlled variable must be kept constant during the experiment because any change in the controlled variable will invalidate the results of the experiment.

Take, for instance, the experiment that attempts to investigate the effect of sunlight on the growth of plants. As we can see, the independent variable would be the amount of light the plant receives, while the dependent variable would be plant growth. Hence, given the amount of light we give to the plant (independent variable), the researcher measures the growth of the plan every, say, 5 days (dependent variable). As we can see, the results of the measurement will depend upon the amount of light given to the plant. The controlled variable in this experiment could be temperature, water, size of container, humidity, and type of soil. Please note again that changes in these variables will invalidate the results of the study. Hence, they must remain constant (thus, controlled) throughout the experiment. Thus, in experimental research, there are only two variables that change, namely, the independent variable and the dependent variable.

Under Data Collection Methodologies

Data collection method is the second major step in primary quantitative research. Scholars and researchers divide data collection methodologies into two, namely:

1) sampling methods and 

2) surveys and polls

Sampling Method

In statistical research, sampling refers to the process whereby researchers select a representative group from the population to be studied. A sample, therefore, is the group of people who will take part in the study while the target population is the total group of people or individuals from which the sample can be drawn. As is well known, the individuals who will take part in the study are referred to as “participants”.

Quantitative researchers divide sampling method into two major types, namely, probability sampling method and non-probability sampling method.

On the one hand, probability sampling is a sampling technique where a sample from a larger target population are chosen with the use of a method based on the theory of probability, a theory in mathematics which is concerned with the analysis of random phenomena. It involves random selection, which allows researchers to make statistical inferences about the whole population.

It must be noted that when doing probability sampling, everyone in the target population must have equal chance of being selected. Hence, in probability sampling, participants are chosen randomly. And if the researcher wants to produce results that are truly representative of the whole population, then she needs to employ a probability sampling technique.

There are four main types of probability sampling technique, namely:

1) simple random sampling, 

2) stratified random sampling, 

3) cluster sampling, and 

4) systematic sampling

Simple random sampling technique is the basic sampling technique where researchers select a group of individuals, which is called “a sample”, from a large population. Here, each member of the group is chosen entirely by chance and each has an equal chance of being selected. Please note that this type of sampling technique is implemented when the target population is considerably large.

For example, simple random sampling technique can be simply done by assigning numbers to the population and then randomly select the numbers which will become members of the sample. To illustrate, if a researcher wants to select a simple random sample of 50 members of a particular organization, then she may assign a number to every member in that organization from 1 to 500 using a random number generator to select 50 members.

In stratified random sampling technique, a large population is divided into smaller groups or strata and members of a sample are selected randomly from these groups or strata. However, it must be noted that those groups or strata should not overlap with each other. Of course, these smaller groups or strata should represent the entire population together.

One of the best ways to do stratified random sampling technique is to group the members according to sex, age, ethnicity, economic status, and what not. And then using a simple random sampling technique, the researcher chooses members from these groups. However, it must be noted that members from each group should be distinct to ensure that each member of all group has equal opportunity of being selected.

Please note that his sample technique is also called “random quota sample”.

Cluster sampling technique usually involves selecting participants from geographically spread-out population, such as cities, provinces, or states. Here, the geographically spread-out population is divided into clusters or subgroups, and instead of sampling from each subgroup, researchers randomly select the entire subgroups.

Let us suppose the researchers want to choose 100 participants from the entire population of Australia. Since they can hardly get a complete list of all people in Australia, the researchers may randomly choose areas in Australia, such as cities or states, and randomly select from within these geographical boundaries. This may include families or universities or non-government organizations from the cluster or subgroups.

Lastly, systematic sampling technique is a type of probability sampling technique that is similar to simple random sampling. But instead of randomly generating numbers, sample members are selected at regular intervals. The process starts with the selection of sample members from a larger population through random starting point but with fixed and periodic intervals. This interval, which is known as “sample interval”, is calculated by dividing the entire population size by the desired sample size.

Let us suppose a school is planning to form a dance troupe to represent itself in a national competition. If the school is seeking to form a systematic sample of 300 students from a population of 3000, then it may select every 10^th person from the entire 3000 student population to build a sample systematically.

On the other hand, non-probability sampling technique is one in which, in the process of gathering the samples or participants, all members of the target population do not have equal chances of being chosen. At the core of non-probability sampling is the idea that samples or participants are chosen by the researcher based on her own subjective judgment. Hence, contrary to probability sampling, non-probability sampling technique does not select samples or participants randomly from a target population.

In non-probability sampling technique, subjective judgment is employed in deciding which aspects or elements are included in the sample. For instance, if the researcher would want to study the lived experiences of Asian immigrants who experience racial discrimination in New York City, then the researcher may just select between 10 and 20 samples or participants out of, say, 1000 population of Asian immigrants in New York City. And this 10-20 participants are those who directly experienced racial discrimination.

It is important to note that non-probability sampling technique is commonly employed in qualitative research method.

There are five types of non-probability sampling technique, namely: 

1) convenience sampling, 

2) consecutive sampling, 

3) quota sampling, 

4) snowball sampling, and 

5) judgmental sampling

As the name suggests, convenience sampling technique is a sampling technique where samples or participants are chosen simply because of their convenient accessibility and proximity to the researcher. Here, samples or participants are chosen since they are very easy to recruit for the study.

Of course, it would be better if the researcher will include in the study all members of the target population. But in most cases, the target population is huge and is spread geographically. That is why it is best to use a “sampling technique”, and one of the famous sampling techniques is the convenience sampling technique because aside from the fact that it is fast and practical, it is also inexpensive and easy to implement without sacrificing the quality of the output of the research.

Consecutive sampling technique is quite similar to convenience sampling technique, except with a slight variation. Here, the researcher chooses a sample or group of people and then conduct research on it over a period of time, and then moves on to another sample, and so on until the goal of the research is attained. For example, if the researcher wants to determine the lived experiences of Asian immigrants on racial discrimination in the United States, then she may select a sample from New York and conduct research on it and then she may move on to California, select a sample there and conduct a research on it, and so on.

Quota sampling technique is the process of selecting specific characteristics of the target population, such as traits and personalities, to form a subgroup or strata. The researcher may then choose members of various strata or subgroups as part of the sample in proportion to a population on interest.

Snowball sampling technique is used by researchers when potential participants from the target population is hard to locate or the sample of the study is limited to extremely small strata or subgroups of the target population. As the name suggests, it is called “snowball sampling” because, just as the snow becomes bigger and bigger when rolled down in heaps, the researcher gets more participants by way of referral after interviewing or gathering information from the initial subject.

As we can see, the snowball sampling technique involves the process of asking the initial subject to nominate another potential subject with similar trait, characteristics, or interest. The researcher will then observe, study or gather information from the nominated subject and then continue the process in the same manner until a sufficient number of subjects is obtained.

Judgmental sampling technique is a type of non-probability sampling technique where researchers create samples or select unit to be sampled based on their own knowledge, experience, skills or judgment. Judgmental sampling technique is also called “purposive sampling technique”.

Because the researcher’s knowledge, experience, skills or judgment is instrumental in the creation of the sample, scholars are convinced that this sampling technique will produce highly accurate results with minimal margin of error.

Now, once the sample is determined, the researcher may proceed with the collection of data. This is where surveys and polls come in.

Surveys and Polls

On the one hand, a survey is defined as the collection of data from a predetermined sample with the purpose of gaining information or insights on various topics of interest. It is important to note that surveys involve the gathering of data directly from the source. And given the ease of survey distribution and the wide number of respondents it can reach (of course, depending on research objective and research time), a survey is one of the most important aspects when conducting quantitative research.

On the other hand, a poll is defined as the collection of feedback with the use of close-ended questions from a sample. Unlike surveys which involve asking a wide range of multiple questions, polls involve asking only one question or one multiple choice question.

The most commonly used types of polls are election polls and exit polls. Both of these types of polls collect data or information from a large sample size but use just one multiple question. Here, the participants can choose from among the answers predefined by the researchers themselves. For example, the researchers may ask the respondents who would they vote for president in the upcoming US elections. Researchers then can restrict voters to choose just one answer to the question like, for instance, “A. Obama, B. Biden, C. Trump, or D. Undecied”.

Under Data Analysis Techniques

Data analysis is the third aspect of the primary quantitative research. Needless to say, after raw data were collected, analysis has to follow in order to derive statistical inferences from the research. It is here where the researcher will relate the results of the study to the objective of the study and then establish the statistical relevance of the results.

Data analysis, therefore, aims to extract useful information from the data gathered in order to come up with a sound decision or, perhaps, a reasoned opinion regarding the issue under investigation. This necessarily involves the process of cleaning, transforming, and modeling data.

Some of the common statistical analysis methods that researchers used in analyzing quantitative data are 1) SWOT analysis method, 2) conjoint analysis method, 3) cross-tabulation method, and 4) TURF analysis method.

SWOT stands for Strengths, Weaknesses, Opportunities, and Threats. SWOT, therefore, is a statistical analysis method that is used to assess and evaluate the strengths, weaknesses, opportunities, and threats of, say, organizations or any relevant entity for that matter. An organization, for example, may use SWOT analysis to make the most of what it got to its advantage. In this way, organizations can reduce the chances of failure by understanding what they lack which then enables it to eliminate threats that may catch them unaware.

SWOT analysis also enables an organization to develop strategies that can be the sources of its competitive advantage over its competitors, which in turn enables the organization to compete successfully in the market.

Conjoint analysis is a survey-based statistical technique that is used in market research to learn how individuals make different and complicated purchasing decisions. Researchers consider conjoint analysis as one of the most effective models in extracting consumer preferences during the purchasing process. This is because after the data are turned into “quantitative measurements” through the use of a statistical analysis, organizations or companies can then evaluate products or services in a way no other method can.

Cross-tabulation is also known as contingency tables or cross tabs. As a statistical analysis technique, researchers use cross-tabulation to find patterns, trends, and probabilities within raw data. Researchers do this by grouping different variables to understand their correlation and show how such correlation changes from one variable grouping to another.

Lastly, TURF stands for Totally Unduplicated Reach and Frequency. TURF analysis is a type of statistical analysis that researchers used in understanding the potential of a target market. TURF analysis usually identifies the number of users reached by a communication as well as the manner in which they were reached.

Secondary Quantitative Research Method

As the name suggests, secondary quantitative research is a research method that appropriates already existing data. Secondary quantitative research method is also known as desk research. This is so because researchers can just sit on their desks and conduct research using existing data or secondary data. Researchers may collect quantitative data from the internet, libraries, previous researches, or government records and then summarize and collate such existing data to increase the overall effectiveness of research.

While it is possible for researchers to come up with some new findings or conclusions from already existing data, secondary quantitative research method has the function of validating existing data with the purpose of strengthening or proving and disproving previously collected data. Some of the famously used secondary quantitative research methods are:

1) Data available on the internet, 

2) Government and non-government sources, 

3) Public libraries, 

4) Educational institutions, and 

5) Commercial information sources.