Quantitative Research Methods and Designs
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 research designs.
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-experi mental 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.
Unit 3 Study 1
Quantitative Research Methods and Designs
Use your Counseling Research: Quantitative, Qualitative, and Mixed Methods text to read the following:
• Chapter 5, “Experimental Designs,” pages 46–60.
• Chapter 6, “Predictive Designs,” pages 61–79.
Use the Library to read the following:
• Whiston and Campbell’s 2010 article, “Randomized Clinical Trials in Counseling: An Introduction,” from Counseling Outcome Research and Evaluation, volume 1, issue 1, pages 8–18.
You will find two appropriate journal articles to complete this unit’s discussion question. These should be different articles from the ones you for the first two weeks. Use the Library to locate two recent (no more than five years old) peer-reviewed journal articles that you may also incorporate into your pre-proposal assignment. Locate one experimental or quasi-experimental study and one predictive study that uses one of the designs discussed in Chapter 6 of Counseling Research: Quantitative, Qualitative, and Mixed Methods. Use the Finding Articles by TYPE guides from the Library to learn how to find experimental and non-experimental research articles.
Click Quantitative Research Decision Tree to view the interactive multimedia.
Below is the Quantitative Research Decision Tree to view the interactive multimedia.
Quantitative Research Decision Tree
An experiment is a study in which at least one variable is manipulated and subjects are randomly assigned to the different levels or categories of the manipulated variable(s).
Results may be more easily generalized.
Takes more time. The process of randomization limits the number of subjects.
A quasi-experiment is a study that has all the elements of an experiment except that the subjects are not randomly assigned to the groups.
Results may not be more easily generalized
Since there is no randomization the number of available subjects is expanded.
In a non-experiment, there is no manipulation of variables, nor is there any randomization. Examples include descriptive statistics*, correlational research, survey research, and others.
Easiest and quickest research design.
There is a threat to internal validity because of the possibility of uncontrolled confounding variables.
*Note: All quantitative designs contain some descriptive statistics, e.g., demographics. Using only descriptive statistics is not sufficient for a dissertation.
• Pedhazer, E., & Schmelkin, L. P. (1991). Measurement, design, and analysis: An integrated approach. Hillsdale, NJ: Erlbaum.
• Optional Resources
The following resources will help you as you prepare your Quantitative Study Evaluation Assignment:
• Finding Articles by TYPE: this guide helps you search by journal article by specific research design types.
• How Do I Find Peer-Reviewed Articles?: this offers additional explanation on distinguishing characteristics of a peer reviewed article and how to find them.
• Primary and Secondary Sources: this guide displays the differences between primary and secondary research sources.
• What is Scholarly?: this guide offers informaton for understanding what makes a resource scholarly.
Note: The Capella library defines an empirical article as one associated with a quantitative approach to research. In this course, we are using the definition that the APA Manual uses; it states that empirical studies are reports of original research (2010). Both quantitative and qualitative studies are empirical. See more information in the introduction of Unit 1.
American Psychological Association. (2010). Publication manual of the American Psychological Association (6th ed.). Washington, DC: Author.
Discussion 1: 1 page needed with 400 words and 2 references.
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 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.