Quantitative research methodology uses a deductive reasoning process (Erford, 2015, p. 5). It is based on philosophical assumptions that are very different from those that support qualitative research. Quantitative studies fall under what is broadly described as a positivist perspective. Epistemologically, knowledge is something that is believed to be objective and measurable, and the nature of reality (that is, ontology) is such that there is one fixed, observable, and definable reality. Quantitative approaches to research emphasize the objectivity of the researcher, and because a goal is to uncover the one true reality, values (axiological assumptions) and the subjective nature of experience are not likely to be examined.
Quantitative Research Designs
Quantitative research can be categorized in different ways. Brief descriptions of some designs appear below. The chosen research design is determined by the nature of the inquiry, that is, what the researcher wants to learn by conducting the study. Counseling Research: Quantitative, Qualitative, and Mixed Methods thoroughly describes several major reseach.
Experimental research, one of the quantitative designs, involves random selection and random assignment of subjects to two or more groups over which the researcher has control. This is what distinguishes experimental studies from the other designs. Experimental studies in counseling are not that common, because many research questions do not lend themselves to random selection and assignment for ethical reasons. Experimental studies compare the effect of one or more independent variables on one or more dependent variables. Independent variables fall into two broad categories. One type of independent variable involves measuring some characteristic inherent in the study’s participants, such as their age, gender, IQ, personality traits, income, or education level. These demographic or blocking variables are not something which the researcher can manipulate, though the researcher can statistically control for them. The treatment or experimental conditions that the researcher sets up is the other type of independent variable, which is unique to experimental designs. The element of control is what permits researchers to conclude that one variable has caused a change in another variable.
Quasi-experimental research designs come in many different forms. Like experimental research, the researcher aims to compare the effect of the independent variable under their control on the dependent variable. However, the researcher does not or cannot randomly assign individual participants to treatment and control groups, so cause-and-effect relationships cannot be as strongly inferred from the results. Pre-existing conditions of one group in comparison to the other may confound the findings. An example might be a study to examine the potential effects of a new curriculum aimed at reducing bullying in a school district. You provide the training to the fourth through sixth grades in one school but not in another, assuming a large school district in which there are two or more middle schools. You could randomly select which school receives the curriculum (treatment group) and which does not (control group), but you cannot assign individuals to either group. With quasi-experimental studies, it is particularly important for the researcher to carefully consider the threats to validity in the interpretation of the results.
Quantitative studies which have the large sample sizes required to maintain sufficient statistical power may be used to examine the interactive effects of more than one independent variable. For instance, one might examine whether or not people with different personality types, as measured on the Myers-Briggs Type Indicator, respond differently to different types of counseling treatments, while also examining whether or not men and women respond in the same ways to various treatments. When previous research suggests that there may be differential effects on people due to some demographic factor, then one would need to adopt a factorial design to control for these differential effects. Otherwise, the validity of the study could be limited.
Descriptive studies attempt to improve understanding of a phenomenon, either by describing it in succinct quantitative terms or by describing its underlying factors. The goal is not to establish a cause-and-effect relationship, but to use statistics (such as descriptive statistics, correlation, or multiple regression) or data reduction procedures (such as cluster analysis, factor analysis, and multidimensional scaling) to better understand a phenomenon or relationship. Causation cannot be inferred when descriptive designs are used.
Meta-analysis is a statistical procedure which is also considered a non-experimental design (Erford, 2015, p. 139) for determining the degree to which a number of studies examining the same phenomena are in agreement. It takes the standard literature review to another level where statistics are applied in determining an overall effect size. In essence, meta-analysis combines several studies and analyzes them as though they were one big study.
Erford, B. T. (2015). Research and evaluation in counseling (2nd ed.). Stamford, CT: Cengage.
To successfully complete this learning unit, you will be expected to:
1. Summarize the methodological structure of quantitative studies.
Quantitative Research Articles Summary
After studying the introduction to this unit and completing the study activities, briefly compare the uses of the research designs employed in the studies. What is each research design used to determine (for example, relationships between variables, differences among groups)? For one of the quantitative studies, summarize how the quantitative studies, summarize how the sampling, data collection, and data analysis procedures worked together to address the hypothesis. The post should be written in your own words, not direct quotes from the article. Incorporate material from the course text in a meaningful way.
The suggested length for this post is 400–500 words.