Description
- This assignment is a two part assignment. Part 1: According to assignment #2, Please answer the questions below
- Identify a framework that you believe may serve as the lens for your study. Use a theoretical framework if you are a PhD student or a conceptual framework if you are an EdD or EdS student.
- State your revised (based on instructor feedback) problem and purpose statements from Week 2.
- Explain how your chosen framework is relevant to your topic and problem
- Present at least one problem and framework-informed research question for both a qualitative and quantitative study
- For the qualitative research question, propose two or more sub-questions
- For the quantitative research questions, present at least one null and corresponding hypothesis. (Later, you will choose a design and make sure the design is aligned with your RQ)
- Discuss how your research questions and hypotheses align with your chosen framework and purpose statement in addressing the research problem
Length: Length: 2- 3 pages
Part 2:
The second part of the assignment, you will explore the role of paradigmatic perspective in considering your research findings and operationalizing research through quantitative methodology and corresponding research designs. Where the problem and purpose statements frame what is to be studied, the paradigmatic perspective helps frame the manner in which you will view the results. The methodology and design frame how the study will be conducted.
Of course, before beginning any research project, the question of why must be addressed. In contemplating why educational research is necessary, consider whether you would implement a new instructional strategy in the classroom, for example, based solely on a recommendation from a colleague. More likely, as a graduate-level scholar, you would thoroughly research the effectiveness of any strategy that may benefit your learning. Likewise, as an educational researcher, you have a responsibility to research strategies, policies, and initiatives before authorizing or administering them in your institution or educational settings.
The underlying purpose of all research is to provide answers based on the best available evidence. As a doctoral-level researcher, your obligation is to continue to refine understandings and practices through high-quality research studies. It is essential to understand the history, principles, and traditions of educational research for you to conduct responsible, ethical, and rigorous research. Keep in mind that, to a novice researcher, all of the structures, commonalities, and differences can at first be confusing. As you go through this course and future research courses, it is important to demonstrate your understanding of the traditions and conventions of quantitative and qualitative methods and the associated designs and data-gathering approaches so that you can make informed choices for your research study design.
Paradigmatic Perspective
As a novice researcher, your worldview will help you determine your own identity as a new researcher and to frame how you conduct your future research. Experienced researchers can adopt different paradigmatic perspectives through which to evaluate the findings of their research, adopting the one having the most relevance in determining what they want to know. Where the theoretical or conceptual framework is the lens through which you view the problem, the paradigmatic perspective is related to the why of the research as it frames how you will view the findings. Most topics can be approached for different higher purposes (not to be confused with the purpose statement) with the results evaluated through aligned frames. For instance, if you are interested in the cohort default rates of student borrowers, you can approach it in many ways. If you are most interested in how the cohort default rate differs amongst categories of students or between types of institutions, you could come at the problem from a positivist paradigmatic perspective using a quantitative correlational design. Please note that certain paradigmatic perspectives align with specific methodologies (you will see examples of perspectives that fit better with qualitative methodology in next week’s lesson).
Positivism, which Auguste Compte is known for, is heavily invested in facts; Compte is often credited for the first attempts to systemize human knowledge (Mills, 1907). Positivists asserted that scientific verification required rational, mathematical, or logical proof; therefore, prior worldviews founded in metaphysics and theism were no longer acceptable and scientific reasoning emerged as a dominant mode for new knowledge.
Figure 4
Positivism, Realism, Quantitative
To investigate anything other than what can physically be observed—much of what occurs in the social sciences—other worldviews must be considered. You will need to decide what you want to accomplish: to determine causes or outcomes, to develop increased understanding, to support marginalized populations, or to use multiple approaches in tackling a real-world problem.
Figure 5
Understanding Theories and Phenomena
Among other considerations, your worldview will likely influence whether you choose a qualitative, quantitative, or mixed-methods study. The most common research paradigmatic perspectives include (Creswell & Creswell, 2018):
Methodologies
At NCU, methodology refers to the tradition or approach used to guide the study. The three basic methodologies in research are qualitative, quantitative, and mixed-methods. Quantitative research is described as the collection of numerical data to identify relationships, make predictions, or test causal inferences.
We will discuss qualitative research next week and because mixed-methods is a complex approach often employing a multi-stage design where the quantitative data informs the analysis of the qualitative data and vice versa (not just one qualitative design and one quantitative design studied together), it is not recommended for novice researchers. Therefore, in this course, the focus will be on quantitative and qualitative methodologies as operationalized through various aligned research designs.
A common misconception by novice researchers is that you add one piece of a quantitative study and one piece of a qualitative study and that equates to a quality mixed-methods study. But, the mixed-methods methodology uses a sophisticated combination of qualitative and quantitative data collection methods using a sequential or concurrent approach. Most faculty members will strongly suggest against using mixed-methods methodology for a dissertation or capstone project as it will add a significant amount of time to your work (and there is a very high likelihood of not reaching the necessary quality standards in time to graduate).
Quantitative Methodology
Quantitative research typically involves the collection and analysis of numerical data to better understand relationships, effects, similarities, and/or differences. Quantitative researchers use these numerical results to formulate some form of an accurate description, understanding, truth, theory, and/or prediction concerning a topic of study. For a study to be truly quantitative, the study must be based on valid, reliable, numeric data, and hypotheses must be generated and tested (usually statistically). Because quantitative research involves numerical data, various statistical analyses are employed to evaluate the results. Quantitative researchers must, therefore, possess a working knowledge of the various quantitative designs and statistical analysis techniques.
Quantitative research also involves analyzing numerical data with the use of inferential statistical analyses to address null and alternative hypotheses. Quantitative research results tend to be more generalizable than qualitative data; a representative sample from a larger population is often utilized, allowing the numerical results and corresponding conclusions to be applied more broadly beyond the research site.
For each quantitative research question, there needs to be a null and alternative hypothesis presented.
Quantitative Research Designs
Selecting a research methodology and corresponding design to accurately address a specific problem worthy of graduate-level research requires considerable understanding, planning, and time. An appropriate choice of methodology and corresponding design can produce valid and convincing contributions in your specialization or area of interest, limiting other plausible interpretations of the results. It is essential for you to initially understand the foundations, principles, and traditions of your chosen methodology in order to apply them appropriately and ensure they are in alignment with your research interests. If the research is not developed and conducted accurately per the methodology and design, then your results will not be meaningful.
As you explored in Week 2, your research methodology and design are driven by your research problem (which is based on a legitimate need for research) and your worldview. In Week 3, you learned how your research methodology and design are also driven by your research questions.
While there is a large array of specific quantitative designs, they can all be positioned under three major approaches: experimental, quasi-experimental, and nonexperimental (correlational or causal-comparative). Since hypotheses are not generated and tested in descriptive studies, such studies can be viewed as either quantitative or qualitative methodologies.
Which design you use depends on several issues that we will discuss briefly in this course (see Table 1) and in more detail in your advanced quantitative methodology course (if applicable).
Table 1
Design/Hallmark/Example Table
Design |
Hallmark/Thumbnail |
Example |
|
Correlational/De Facto |
Looks for relationships with no implication of causality. Avoid words like impact, influence, cause, etc. |
The relationship between hours spent in test preparation and scores on state-mandated math exams or an introductory freshman life skills course and retention. |
|
Causal comparative/Ex Post Facto |
Attempts to determine the cause or consequences of differences that exist between groups after a treatment has already occurred. |
Looking for significant differences in student awareness of diversity and inclusion where one group of students had completed a diversity course and another group did not. |
|
Quasi-Experimental |
No acts of randomization; existing groups are used; provides some measure of causality. |
Two very similar schools or universities use two different math curricula. To what extent is there a significant difference in math scores? |
|
Experimental |
A sample is randomly selected from the target population and subjects are randomly assigned to treatments. Best opportunity to assess causality. |
From a school district or university, teachers or professors are randomly selected to participate in a study of the mode of delivery for professional development. Half of those selected are assigned to a traditional face-to-face setting, half to an online medium. To what extent does the mode of delivery significantly impact the outcome achievement of the professional development programs? |
Quantitative Data Collection Methods and Analysis
Generally, measurements used for data collection in quantitative research are used to generate numeric variables. In quantitative research, numeric data are converted into a measurable form—a variable—to be assessed or statistically analyzed.
Variables representing events, people, or objects can be converted or operationalized to numeric values. In quantitative research, concepts need to be converted into a measurable form—an operationalized variable—to be assessed. An operationalized variable can be categorical (e.g., hair color) and typically codified by categorical numeric values for analysis.
These concepts can be measured numerically, and variables can be observable (e.g., height or scores) or abstract (e.g., self-esteem as measured with an instrument). You can add, subtract, or perform some kind of numerical function with quantitative variables. For example, in quantitative research, motivation can be measured by the number of times a student attends optional tutoring. The variable of motivation would be operationally defined as attendance at optional tutoring sessions.
Variables fit into four categories. Numeric variables can be nominal (a category is “named” or “tagged” using a number), ordinal (ranked such as on a Likert scale— from very satisfied = 5 to very dissatisfied = 1), or interval/ratio variables (the distance between the variables have meaning). Quantitative variables are at least interval-level (differences) or ratio-level (amounts) variables.
These data are then collected by administering a measurement instrument. For instance, a survey is an instrument that can be used within a huge array of designs under the quantitative methodology. For example, in using a Likert scale of 1-5 those perceptions are presented numerically.
Instruments, such as a survey, require pilot testing and statistical testing to verify they are valid and reliable, which can be very time-consuming. Before planning to develop your instrument, research what is required in that effort very closely and consider modifying an existing survey instead. Most dissertation chairs will strongly recommend against creating an instrument for your study as this will typically add about six months to your dissertation.
Validity refers to the appropriateness of measurement in relation to the variable. Consider the online tutoring and motivation example previously discussed. For measuring motivation by attendance at optional tutoring sessions on the face of it (face validity) would seem reasonable and appropriate. By contrast, if the variable were again motivation, measuring motivation by how often the student sat in the same desk in the classroom does not on its face appear to be a valid measure. Validity is one of the main concerns in all types of research. Validity, generally, is an indication of the soundness of the research. There are many statistical analyses used to measure validity as well which we will not go into detail in this course.
We must also consider both internal and external validity:
Internal Validity is affected by the assumptions, limitations, and delimitations of the study. Design choices, potential issues with the research instrument, inability to control for extraneous variables, population and sample size, time constraints, etc. all contribute to issues of internal validity because they the affect interpretation of the data by the researcher.
External Validity is the extent to which your findings can be generalized or transferred to the larger or like groups. Some factors that affect external validity are researcher bias/experimenter effects, research environment, timeliness of the study, good definitions of variables, and the extent to which the sample is representative of the population.
There are four types of validity:
- Content Validity refers to the extent to which the content of a test or instrument represents what it purports to measure.
- Construct Validity refers to the extent to which a test or instrument actually measures what it intends to measure.
- Criterion Validity is the extent to which the results of a test are consistent with other tests of the same thing.
- Face Validity is the extent to which the content of the test or instrument is aligned with the aims or purpose of the study.
A measurement instrument must have high validity, but it also must have high reliability. Reliability is the ability of a variable to give consistent results. Using the same online tutoring and motivation example, both measures would give consistent results. However, the first approach (number of tutoring sessions) is both valid and reliable (at least at face value). By contrast, the second approach of position in the classroom is not valid but is reliable. Where validity is the extent to which the results of a test really measure what they are supposed to measure, reliability is the extent to which results can be replicated if the research is repeated in the same environment. Reliable results are not necessarily valid. For instance, the polygraph (i.e., lie detector test) produces reliable measures of blood pressure, pulse, respiration rate, and galvanic conductivity but it is not considered a valid measure of truthfulness by the courts because there is no scientifically supported connection (i.e., no construct validity) between those measures of physical response and truthfulness. In quantitative studies, variables must be measurable in a valid and reliable way. The various ways in which data are actually gathered within each method and respective design are illustrated in Figure 6. To a large extent, the data collection method and category of variables is guided by the educational problem to be examined.
Figure 6 provides a non-comprehensive visual as to the nesting nature of these terms and some specific examples.
Figure 6
Quantitative Charts
This week, you will delve further into articulating an actual educational problem worthy of graduate-level research for a quantitative research study. Remember to be working on your CITI training, as your completion report is due in Week 7.
Be sure to review this week’s resources carefully. You are expected to apply the information from these resources when you prepare your assignments.
References
Creswell, J. W., & Creswell, J. D. (2018). Research design: Qualitative, quantitative, and mixed methods approaches (5th ed.). SAGE.
Mill, J. S. (1907). Auguste Comte and positivism (5th ed.). Kegan Paul, Trench, Trubner & Company.
Northcentral University Library. (2018). During this week, you learned about paradigmatic perspectives, quantitative methodology, and quantitative designs. Last week, your assignment was to envision your research focus from both theoretical and applied research approaches.
This week, you are to consider research into your topic using quantitative methodology. To complete this assignment, you will use the Week 3 discussion of research questions and the Week 2 problem and purpose statements (as refined through feedback received). These statements should align with the focus and scope (i.e., theoretical or applied) of your degree program. In this assignment, you will propose alignment of paradigmatic perspective, quantitative methodology, and one quantitative design.
Organize your work as follows:
- Part 1: one-paragraph summary of your evolving topical area of interest
- Part 2: your refined problem and purpose statements
- Part 3: one potential research question and related hypotheses
- Part 4: a side-by-side presentation (see example) of paradigmatic perspective, quantitative methodology, and at least two quantitative designs
Paradigmatic Perspective |
Methodology |
Design(s) |
|
Quantitative |
||
|
Quantitative |
- Part 5: one page stating the justification of your choices, including how your choices best suit investigating your refined problem statement and fulfilling the purpose of the research
Length: 2-3 pages, excluding title and reference pages.
References: Include a minimum of 4 scholarly resources, which can include research guides.
