Quantitative and Qualitative Approaches to Comparative Education

Cite this chapter.

Book cover

  • Gregory P. Fairbrother 5  

Part of the book series: CERC Studies in Comparative Education ((CERC,volume 19))

5649 Accesses

8 Citations

Among the many approaches to research, a broad classification distinguishes between the quantitative and the qualitative. Boundaries may be difficult to determine, and the approaches may not be mutually exclusive. Nevertheless, the two approaches deserve focus because they permit different types of insights.

  • Qualitative Approach
  • Literacy Skill
  • Literacy Rate
  • Direct Assessment
  • Ethnographic Study

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Unable to display preview.  Download preview PDF.

Author information

Authors and affiliations.

Department of Mathematics, Science, Social Sciences & Technology, The Hong Kong Institute of Education, 10 Lo Ping Road, Tai Po, Hong Kong, China

Gregory P. Fairbrother

You can also search for this author in PubMed   Google Scholar

Editor information

Editors and affiliations.

UNESCO, 7-9 rue Eugène Delacroix, 75116, Paris, France

Hong Kong Institute of Education, 10 Lo Ping Road, Tai Po, Hong Kong SAR, China

Bob Adamson

The University of Hong Kong, Pokfulam Road, Hong Kong, China

Rights and permissions

Reprints and permissions

Copyright information

© 2007 Springer

About this chapter

Fairbrother, G.P. (2007). Quantitative and Qualitative Approaches to Comparative Education. In: Bray, M., Adamson, B., Mason, M. (eds) Comparative Education Research. CERC Studies in Comparative Education, vol 19. Springer, Dordrecht. https://doi.org/10.1007/978-1-4020-6189-9_2

Download citation

DOI : https://doi.org/10.1007/978-1-4020-6189-9_2

Publisher Name : Springer, Dordrecht

Print ISBN : 978-1-4020-6188-2

Online ISBN : 978-1-4020-6189-9

eBook Packages : Humanities, Social Sciences and Law Education (R0)

Share this chapter

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

  • Publish with us

Policies and ethics

  • Find a journal
  • Track your research

Qualitative vs. Quantitative Research: Comparing the Methods and Strategies for Education Research

A woman sits at a library table with stacks of books and a laptop.

No matter the field of study, all research can be divided into two distinct methodologies: qualitative and quantitative research. Both methodologies offer education researchers important insights.

Education research assesses problems in policy, practices, and curriculum design, and it helps administrators identify solutions. Researchers can conduct small-scale studies to learn more about topics related to instruction or larger-scale ones to gain insight into school systems and investigate how to improve student outcomes.

Education research often relies on the quantitative methodology. Quantitative research in education provides numerical data that can prove or disprove a theory, and administrators can easily share the number-based results with other schools and districts. And while the research may speak to a relatively small sample size, educators and researchers can scale the results from quantifiable data to predict outcomes in larger student populations and groups.

Qualitative vs. Quantitative Research in Education: Definitions

Although there are many overlaps in the objectives of qualitative and quantitative research in education, researchers must understand the fundamental functions of each methodology in order to design and carry out an impactful research study. In addition, they must understand the differences that set qualitative and quantitative research apart in order to determine which methodology is better suited to specific education research topics.

Generate Hypotheses with Qualitative Research

Qualitative research focuses on thoughts, concepts, or experiences. The data collected often comes in narrative form and concentrates on unearthing insights that can lead to testable hypotheses. Educators use qualitative research in a study’s exploratory stages to uncover patterns or new angles.

Form Strong Conclusions with Quantitative Research

Quantitative research in education and other fields of inquiry is expressed in numbers and measurements. This type of research aims to find data to confirm or test a hypothesis.

Differences in Data Collection Methods

Keeping in mind the main distinction in qualitative vs. quantitative research—gathering descriptive information as opposed to numerical data—it stands to reason that there are different ways to acquire data for each research methodology. While certain approaches do overlap, the way researchers apply these collection techniques depends on their goal.

Interviews, for example, are common in both modes of research. An interview with students that features open-ended questions intended to reveal ideas and beliefs around attendance will provide qualitative data. This data may reveal a problem among students, such as a lack of access to transportation, that schools can help address.

An interview can also include questions posed to receive numerical answers. A case in point: how many days a week do students have trouble getting to school, and of those days, how often is a transportation-related issue the cause? In this example, qualitative and quantitative methodologies can lead to similar conclusions, but the research will differ in intent, design, and form.

Taking a look at behavioral observation, another common method used for both qualitative and quantitative research, qualitative data may consider a variety of factors, such as facial expressions, verbal responses, and body language.

On the other hand, a quantitative approach will create a coding scheme for certain predetermined behaviors and observe these in a quantifiable manner.

Qualitative Research Methods

  • Case Studies : Researchers conduct in-depth investigations into an individual, group, event, or community, typically gathering data through observation and interviews.
  • Focus Groups : A moderator (or researcher) guides conversation around a specific topic among a group of participants.
  • Ethnography : Researchers interact with and observe a specific societal or ethnic group in their real-life environment.
  • Interviews : Researchers ask participants questions to learn about their perspectives on a particular subject.

Quantitative Research Methods

  • Questionnaires and Surveys : Participants receive a list of questions, either closed-ended or multiple choice, which are directed around a particular topic.
  • Experiments : Researchers control and test variables to demonstrate cause-and-effect relationships.
  • Observations : Researchers look at quantifiable patterns and behavior.
  • Structured Interviews : Using a predetermined structure, researchers ask participants a fixed set of questions to acquire numerical data.

Choosing a Research Strategy

When choosing which research strategy to employ for a project or study, a number of considerations apply. One key piece of information to help determine whether to use a qualitative vs. quantitative research method is which phase of development the study is in.

For example, if a project is in its early stages and requires more research to find a testable hypothesis, qualitative research methods might prove most helpful. On the other hand, if the research team has already established a hypothesis or theory, quantitative research methods will provide data that can validate the theory or refine it for further testing.

It’s also important to understand a project’s research goals. For instance, do researchers aim to produce findings that reveal how to best encourage student engagement in math? Or is the goal to determine how many students are passing geometry? These two scenarios require distinct sets of data, which will determine the best methodology to employ.

In some situations, studies will benefit from a mixed-methods approach. Using the goals in the above example, one set of data could find the percentage of students passing geometry, which would be quantitative. The research team could also lead a focus group with the students achieving success to discuss which techniques and teaching practices they find most helpful, which would produce qualitative data.

Learn How to Put Education Research into Action

Those with an interest in learning how to harness research to develop innovative ideas to improve education systems may want to consider pursuing a doctoral degree. American University’s School of Education Online offers a Doctor of Education (EdD) in Education Policy and Leadership that prepares future educators, school administrators, and other education professionals to become leaders who effect positive changes in schools. Courses such as Applied Research Methods I: Enacting Critical Research provides students with the techniques and research skills needed to begin conducting research exploring new ways to enhance education. Learn more about American’ University’s EdD in Education Policy and Leadership .

What’s the Difference Between Educational Equity and Equality?

EdD vs. PhD in Education: Requirements, Career Outlook, and Salary

Top Education Technology Jobs for Doctorate in Education Graduates

American University, EdD in Education Policy and Leadership

Edutopia, “2019 Education Research Highlights”

Formplus, “Qualitative vs. Quantitative Data: 15 Key Differences and Similarities”

iMotion, “Qualitative vs. Quantitative Research: What Is What?”

Scribbr, “Qualitative vs. Quantitative Research”

Simply Psychology, “What’s the Difference Between Quantitative and Qualitative Research?”

Typeform, “A Simple Guide to Qualitative and Quantitative Research”

Request Information

U.S. flag

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

  • Publications
  • Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

  • Advanced Search
  • Journal List
  • Front Psychol

Quantitative and Qualitative Approaches to Generalization and Replication–A Representationalist View

In this paper, we provide a re-interpretation of qualitative and quantitative modeling from a representationalist perspective. In this view, both approaches attempt to construct abstract representations of empirical relational structures. Whereas quantitative research uses variable-based models that abstract from individual cases, qualitative research favors case-based models that abstract from individual characteristics. Variable-based models are usually stated in the form of quantified sentences (scientific laws). This syntactic structure implies that sentences about individual cases are derived using deductive reasoning. In contrast, case-based models are usually stated using context-dependent existential sentences (qualitative statements). This syntactic structure implies that sentences about other cases are justifiable by inductive reasoning. We apply this representationalist perspective to the problems of generalization and replication. Using the analytical framework of modal logic, we argue that the modes of reasoning are often not only applied to the context that has been studied empirically, but also on a between-contexts level. Consequently, quantitative researchers mostly adhere to a top-down strategy of generalization, whereas qualitative researchers usually follow a bottom-up strategy of generalization. Depending on which strategy is employed, the role of replication attempts is very different. In deductive reasoning, replication attempts serve as empirical tests of the underlying theory. Therefore, failed replications imply a faulty theory. From an inductive perspective, however, replication attempts serve to explore the scope of the theory. Consequently, failed replications do not question the theory per se , but help to shape its boundary conditions. We conclude that quantitative research may benefit from a bottom-up generalization strategy as it is employed in most qualitative research programs. Inductive reasoning forces us to think about the boundary conditions of our theories and provides a framework for generalization beyond statistical testing. In this perspective, failed replications are just as informative as successful replications, because they help to explore the scope of our theories.


Qualitative and quantitative research strategies have long been treated as opposing paradigms. In recent years, there have been attempts to integrate both strategies. These “mixed methods” approaches treat qualitative and quantitative methodologies as complementary, rather than opposing, strategies (Creswell, 2015 ). However, whilst acknowledging that both strategies have their benefits, this “integration” remains purely pragmatic. Hence, mixed methods methodology does not provide a conceptual unification of the two approaches.

Lacking a common methodological background, qualitative and quantitative research methodologies have developed rather distinct standards with regard to the aims and scope of empirical science (Freeman et al., 2007 ). These different standards affect the way researchers handle contradictory empirical findings. For example, many empirical findings in psychology have failed to replicate in recent years (Klein et al., 2014 ; Open Science, Collaboration, 2015 ). This “replication crisis” has been discussed on statistical, theoretical and social grounds and continues to have a wide impact on quantitative research practices like, for example, open science initiatives, pre-registered studies and a re-evaluation of statistical significance testing (Everett and Earp, 2015 ; Maxwell et al., 2015 ; Shrout and Rodgers, 2018 ; Trafimow, 2018 ; Wiggins and Chrisopherson, 2019 ).

However, qualitative research seems to be hardly affected by this discussion. In this paper, we argue that the latter is a direct consequence of how the concept of generalizability is conceived in the two approaches. Whereas most of quantitative psychology is committed to a top-down strategy of generalization based on the idea of random sampling from an abstract population, qualitative studies usually rely on a bottom-up strategy of generalization that is grounded in the successive exploration of the field by means of theoretically sampled cases.

Here, we show that a common methodological framework for qualitative and quantitative research methodologies is possible. We accomplish this by introducing a formal description of quantitative and qualitative models from a representationalist perspective: both approaches can be reconstructed as special kinds of representations for empirical relational structures. We then use this framework to analyze the generalization strategies used in the two approaches. These turn out to be logically independent of the type of model. This has wide implications for psychological research. First, a top-down generalization strategy is compatible with a qualitative modeling approach. This implies that mainstream psychology may benefit from qualitative methods when a numerical representation turns out to be difficult or impossible, without the need to commit to a “qualitative” philosophy of science. Second, quantitative research may exploit the bottom-up generalization strategy that is inherent to many qualitative approaches. This offers a new perspective on unsuccessful replications by treating them not as scientific failures, but as a valuable source of information about the scope of a theory.

The Quantitative Strategy–Numbers and Functions

Quantitative science is about finding valid mathematical representations for empirical phenomena. In most cases, these mathematical representations have the form of functional relations between a set of variables. One major challenge of quantitative modeling consists in constructing valid measures for these variables. Formally, to measure a variable means to construct a numerical representation of the underlying empirical relational structure (Krantz et al., 1971 ). For example, take the behaviors of a group of students in a classroom: “to listen,” “to take notes,” and “to ask critical questions.” One may now ask whether is possible to assign numbers to the students, such that the relations between the assigned numbers are of the same kind as the relations between the values of an underlying variable, like e.g., “engagement.” The observed behaviors in the classroom constitute an empirical relational structure, in the sense that for every student-behavior tuple, one can observe whether it is true or not. These observations can be represented in a person × behavior matrix 1 (compare Figure 1 ). Given this relational structure satisfies certain conditions (i.e., the axioms of a measurement model), one can assign numbers to the students and the behaviors, such that the relations between the numbers resemble the corresponding numerical relations. For example, if there is a unique ordering in the empirical observations with regard to which person shows which behavior, the assigned numbers have to constitute a corresponding unique ordering, as well. Such an ordering coincides with the person × behavior matrix forming a triangle shaped relation and is formally represented by a Guttman scale (Guttman, 1944 ). There are various measurement models available for different empirical structures (Suppes et al., 1971 ). In the case of probabilistic relations, Item-Response models may be considered as a special kind of measurement model (Borsboom, 2005 ).

An external file that holds a picture, illustration, etc.
Object name is fpsyg-12-605191-g0001.jpg

Constructing a numerical representation from an empirical relational structure; Due to the unique ordering of persons with regard to behaviors (indicated by the triangular shape of the relation), it is possible to construct a Guttman scale by assigning a number to each of the individuals, representing the number of relevant behaviors shown by the individual. The resulting variable (“engagement”) can then be described by means of statistical analyses, like, e.g., plotting the frequency distribution.

Although essential, measurement is only the first step of quantitative modeling. Consider a slightly richer empirical structure, where we observe three additional behaviors: “to doodle,” “to chat,” and “to play.” Like above, one may ask, whether there is a unique ordering of the students with regard to these behaviors that can be represented by an underlying variable (i.e., whether the matrix forms a Guttman scale). If this is the case, we may assign corresponding numbers to the students and call this variable “distraction.” In our example, such a representation is possible. We can thus assign two numbers to each student, one representing his or her “engagement” and one representing his or her “distraction” (compare Figure 2 ). These measurements can now be used to construct a quantitative model by relating the two variables by a mathematical function. In the simplest case, this may be a linear function. This functional relation constitutes a quantitative model of the empirical relational structure under study (like, e.g., linear regression). Given the model equation and the rules for assigning the numbers (i.e., the instrumentations of the two variables), the set of admissible empirical structures is limited from all possible structures to a rather small subset. This constitutes the empirical content of the model 2 (Popper, 1935 ).

An external file that holds a picture, illustration, etc.
Object name is fpsyg-12-605191-g0002.jpg

Constructing a numerical model from an empirical relational structure; Since there are two distinct classes of behaviors that each form a Guttman scale, it is possible to assign two numbers to each individual, correspondingly. The resulting variables (“engagement” and “distraction”) can then be related by a mathematical function, which is indicated by the scatterplot and red line on the right hand side.

The Qualitative Strategy–Categories and Typologies

The predominant type of analysis in qualitative research consists in category formation. By constructing descriptive systems for empirical phenomena, it is possible to analyze the underlying empirical structure at a higher level of abstraction. The resulting categories (or types) constitute a conceptual frame for the interpretation of the observations. Qualitative researchers differ considerably in the way they collect and analyze data (Miles et al., 2014 ). However, despite the diverse research strategies followed by different qualitative methodologies, from a formal perspective, most approaches build on some kind of categorization of cases that share some common features. The process of category formation is essential in many qualitative methodologies, like, for example, qualitative content analysis, thematic analysis, grounded theory (see Flick, 2014 for an overview). Sometimes these features are directly observable (like in our classroom example), sometimes they are themselves the result of an interpretative process (e.g., Scheunpflug et al., 2016 ).

In contrast to quantitative methodologies, there have been little attempts to formalize qualitative research strategies (compare, however, Rihoux and Ragin, 2009 ). However, there are several statistical approaches to non-numerical data that deal with constructing abstract categories and establishing relations between these categories (Agresti, 2013 ). Some of these methods are very similar to qualitative category formation on a conceptual level. For example, cluster analysis groups cases into homogenous categories (clusters) based on their similarity on a distance metric.

Although category formation can be formalized in a mathematically rigorous way (Ganter and Wille, 1999 ), qualitative research hardly acknowledges these approaches. 3 However, in order to find a common ground with quantitative science, it is certainly helpful to provide a formal interpretation of category systems.

Let us reconsider the above example of students in a classroom. The quantitative strategy was to assign numbers to the students with regard to variables and to relate these variables via a mathematical function. We can analyze the same empirical structure by grouping the behaviors to form abstract categories. If the aim is to construct an empirically valid category system, this grouping is subject to constraints, analogous to those used to specify a measurement model. The first and most important constraint is that the behaviors must form equivalence classes, i.e., within categories, behaviors need to be equivalent, and across categories, they need to be distinct (formally, the relational structure must obey the axioms of an equivalence relation). When objects are grouped into equivalence classes, it is essential to specify the criterion for empirical equivalence. In qualitative methodology, this is sometimes referred to as the tertium comparationis (Flick, 2014 ). One possible criterion is to group behaviors such that they constitute a set of specific common attributes of a group of people. In our example, we might group the behaviors “to listen,” “to take notes,” and “to doodle,” because these behaviors are common to the cases B, C, and D, and they are also specific for these cases, because no other person shows this particular combination of behaviors. The set of common behaviors then forms an abstract concept (e.g., “moderate distraction”), while the set of persons that show this configuration form a type (e.g., “the silent dreamer”). Formally, this means to identify the maximal rectangles in the underlying empirical relational structure (see Figure 3 ). This procedure is very similar to the way we constructed a Guttman scale, the only difference being that we now use different aspects of the empirical relational structure. 4 In fact, the set of maximal rectangles can be determined by an automated algorithm (Ganter, 2010 ), just like the dimensionality of an empirical structure can be explored by psychometric scaling methods. Consequently, we can identify the empirical content of a category system or a typology as the set of empirical structures that conforms to it. 5 Whereas the quantitative strategy was to search for scalable sub-matrices and then relate the constructed variables by a mathematical function, the qualitative strategy is to construct an empirical typology by grouping cases based on their specific similarities. These types can then be related to one another by a conceptual model that describes their semantic and empirical overlap (see Figure 3 , right hand side).

An external file that holds a picture, illustration, etc.
Object name is fpsyg-12-605191-g0003.jpg

Constructing a conceptual model from an empirical relational structure; Individual behaviors are grouped to form abstract types based on them being shared among a specific subset of the cases. Each type constitutes a set of specific commonalities of a class of individuals (this is indicated by the rectangles on the left hand side). The resulting types (“active learner,” “silent dreamer,” “distracted listener,” and “troublemaker”) can then be related to one another to explicate their semantic and empirical overlap, as indicated by the Venn-diagram on the right hand side.

Variable-Based Models and Case-Based Models

In the previous section, we have argued that qualitative category formation and quantitative measurement can both be characterized as methods to construct abstract representations of empirical relational structures. Instead of focusing on different philosophical approaches to empirical science, we tried to stress the formal similarities between both approaches. However, it is worth also exploring the dissimilarities from a formal perspective.

Following the above analysis, the quantitative approach can be characterized by the use of variable-based models, whereas the qualitative approach is characterized by case-based models (Ragin, 1987 ). Formally, we can identify the rows of an empirical person × behavior matrix with a person-space, and the columns with a corresponding behavior-space. A variable-based model abstracts from the single individuals in a person-space to describe the structure of behaviors on a population level. A case-based model, on the contrary, abstracts from the single behaviors in a behavior-space to describe individual case configurations on the level of abstract categories (see Table 1 ).

Variable-based models and case-based models.

From a representational perspective, there is no a priori reason to favor one type of model over the other. Both approaches provide different analytical tools to construct an abstract representation of an empirical relational structure. However, since the two modeling approaches make use of different information (person-space vs. behavior-space), this comes with some important implications for the researcher employing one of the two strategies. These are concerned with the role of deductive and inductive reasoning.

In variable-based models, empirical structures are represented by functional relations between variables. These are usually stated as scientific laws (Carnap, 1928 ). Formally, these laws correspond to logical expressions of the form

In plain text, this means that y is a function of x for all objects i in the relational structure under consideration. For example, in the above example, one may formulate the following law: for all students in the classroom it holds that “distraction” is a monotone decreasing function of “engagement.” Such a law can be used to derive predictions for single individuals by means of logical deduction: if the above law applies to all students in the classroom, it is possible to calculate the expected distraction from a student's engagement. An empirical observation can now be evaluated against this prediction. If the prediction turns out to be false, the law can be refuted based on the principle of falsification (Popper, 1935 ). If a scientific law repeatedly withstands such empirical tests, it may be considered to be valid with regard to the relational structure under consideration.

In case-based models, there are no laws about a population, because the model does not abstract from the cases but from the observed behaviors. A case-based model describes the underlying structure in terms of existential sentences. Formally, this corresponds to a logical expression of the form

In plain text, this means that there is at least one case i for which the condition XYZ holds. For example, the above category system implies that there is at least one active learner. This is a statement about a singular observation. It is impossible to deduce a statement about another person from an existential sentence like this. Therefore, the strategy of falsification cannot be applied to test the model's validity in a specific context. If one wishes to generalize to other cases, this is accomplished by inductive reasoning, instead. If we observed one person that fulfills the criteria of calling him or her an active learner, we can hypothesize that there may be other persons that are identical to the observed case in this respect. However, we do not arrive at this conclusion by logical deduction, but by induction.

Despite this important distinction, it would be wrong to conclude that variable-based models are intrinsically deductive and case-based models are intrinsically inductive. 6 Both types of reasoning apply to both types of models, but on different levels. Based on a person-space, in a variable-based model one can use deduction to derive statements about individual persons from abstract population laws. There is an analogous way of reasoning for case-based models: because they are based on a behavior space, it is possible to deduce statements about singular behaviors. For example, if we know that Peter is an active learner, we can deduce that he takes notes in the classroom. This kind of deductive reasoning can also be applied on a higher level of abstraction to deduce thematic categories from theoretical assumptions (Braun and Clarke, 2006 ). Similarly, there is an analog for inductive generalization from the perspective of variable-based modeling: since the laws are only quantified over the person-space, generalizations to other behaviors rely on inductive reasoning. For example, it is plausible to assume that highly engaged students tend to do their homework properly–however, in our example this behavior has never been observed. Hence, in variable-based models we usually generalize to other behaviors by means of induction. This kind of inductive reasoning is very common when empirical results are generalized from the laboratory to other behavioral domains.

Although inductive and deductive reasoning are used in qualitative and quantitative research, it is important to stress the different roles of induction and deduction when models are applied to cases. A variable-based approach implies to draw conclusions about cases by means of logical deduction; a case-based approach implies to draw conclusions about cases by means of inductive reasoning. In the following, we build on this distinction to differentiate between qualitative (bottom-up) and quantitative (top-down) strategies of generalization.

Generalization and the Problem of Replication

We will now extend the formal analysis of quantitative and qualitative approaches to the question of generalization and replicability of empirical findings. For this sake, we have to introduce some concepts of formal logic. Formal logic is concerned with the validity of arguments. It provides conditions to evaluate whether certain sentences (conclusions) can be derived from other sentences (premises). In this context, a theory is nothing but a set of sentences (also called axioms). Formal logic provides tools to derive new sentences that must be true, given the axioms are true (Smith, 2020 ). These derived sentences are called theorems or, in the context of empirical science, predictions or hypotheses . On the syntactic level, the rules of logic only state how to evaluate the truth of a sentence relative to its premises. Whether or not sentences are actually true, is formally specified by logical semantics.

On the semantic level, formal logic is intrinsically linked to set-theory. For example, a logical statement like “all dogs are mammals,” is true if and only if the set of dogs is a subset of the set of mammals. Similarly, the sentence “all chatting students doodle” is true if and only if the set of chatting students is a subset of the set of doodling students (compare Figure 3 ). Whereas, the first sentence is analytically true due to the way we define the words “dog” and “mammal,” the latter can be either true or false, depending on the relational structure we actually observe. We can thus interpret an empirical relational structure as the truth criterion of a scientific theory. From a logical point of view, this corresponds to the semantics of a theory. As shown above, variable-based and case-based models both give a formal representation of the same kinds of empirical structures. Accordingly, both types of models can be stated as formal theories. In the variable-based approach, this corresponds to a set of scientific laws that are quantified over the members of an abstract population (these are the axioms of the theory). In the case-based approach, this corresponds to a set of abstract existential statements about a specific class of individuals.

In contrast to mathematical axiom systems, empirical theories are usually not considered to be necessarily true. This means that even if we find no evidence against a theory, it is still possible that it is actually wrong. We may know that a theory is valid in some contexts, yet it may fail when applied to a new set of behaviors (e.g., if we use a different instrumentation to measure a variable) or a new population (e.g., if we draw a new sample).

From a logical perspective, the possibility that a theory may turn out to be false stems from the problem of contingency . A statement is contingent, if it is both, possibly true and possibly false. Formally, we introduce two modal operators: □ to designate logical necessity, and ◇ to designate logical possibility. Semantically, these operators are very similar to the existential quantifier, ∃, and the universal quantifier, ∀. Whereas ∃ and ∀ refer to the individual objects within one relational structure, the modal operators □ and ◇ range over so-called possible worlds : a statement is possibly true, if and only if it is true in at least one accessible possible world, and a statement is necessarily true if and only if it is true in every accessible possible world (Hughes and Cresswell, 1996 ). Logically, possible worlds are mathematical abstractions, each consisting of a relational structure. Taken together, the relational structures of all accessible possible worlds constitute the formal semantics of necessity, possibility and contingency. 7

In the context of an empirical theory, each possible world may be identified with an empirical relational structure like the above classroom example. Given the set of intended applications of a theory (the scope of the theory, one may say), we can now construct possible world semantics for an empirical theory: each intended application of the theory corresponds to a possible world. For example, a quantified sentence like “all chatting students doodle” may be true in one classroom and false in another one. In terms of possible worlds, this would correspond to a statement of contingency: “it is possible that all chatting students doodle in one classroom, and it is possible that they don't in another classroom.” Note that in the above expression, “all students” refers to the students in only one possible world, whereas “it is possible” refers to the fact that there is at least one possible world for each of the specified cases.

To apply these possible world semantics to quantitative research, let us reconsider how generalization to other cases works in variable-based models. Due to the syntactic structure of quantitative laws, we can deduce predictions for singular observations from an expression of the form ∀ i : y i = f ( x i ). Formally, the logical quantifier ∀ ranges only over the objects of the corresponding empirical relational structure (in our example this would refer to the students in the observed classroom). But what if we want to generalize beyond the empirical structure we actually observed? The standard procedure is to assume an infinitely large, abstract population from which a random sample is drawn. Given the truth of the theory, we can deduce predictions about what we may observe in the sample. Since usually we deal with probabilistic models, we can evaluate our theory by means of the conditional probability of the observations, given the theory holds. This concept of conditional probability is the foundation of statistical significance tests (Hogg et al., 2013 ), as well as Bayesian estimation (Watanabe, 2018 ). In terms of possible world semantics, the random sampling model implies that all possible worlds (i.e., all intended applications) can be conceived as empirical sub-structures from a greater population structure. For example, the empirical relational structure constituted by the observed behaviors in a classroom would be conceived as a sub-matrix of the population person × behavior matrix. It follows that, if a scientific law is true in the population, it will be true in all possible worlds, i.e., it will be necessarily true. Formally, this corresponds to an expression of the form

The statistical generalization model thus constitutes a top-down strategy for dealing with individual contexts that is analogous to the way variable-based models are applied to individual cases (compare Table 1 ). Consequently, if we apply a variable-based model to a new context and find out that it does not fit the data (i.e., there is a statistically significant deviation from the model predictions), we have reason to doubt the validity of the theory. This is what makes the problem of low replicability so important: we observe that the predictions are wrong in a new study; and because we apply a top-down strategy of generalization to contexts beyond the ones we observed, we see our whole theory at stake.

Qualitative research, on the contrary, follows a different strategy of generalization. Since case-based models are formulated by a set of context-specific existential sentences, there is no need for universal truth or necessity. In contrast to statistical generalization to other cases by means of random sampling from an abstract population, the usual strategy in case-based modeling is to employ a bottom-up strategy of generalization that is analogous to the way case-based models are applied to individual cases. Formally, this may be expressed by stating that the observed qualia exist in at least one possible world, i.e., the theory is possibly true:

This statement is analogous to the way we apply case-based models to individual cases (compare Table 1 ). Consequently, the set of intended applications of the theory does not follow from a sampling model, but from theoretical assumptions about which cases may be similar to the observed cases with respect to certain relevant characteristics. For example, if we observe that certain behaviors occur together in one classroom, following a bottom-up strategy of generalization, we will hypothesize why this might be the case. If we do not replicate this finding in another context, this does not question the model itself, since it was a context-specific theory all along. Instead, we will revise our hypothetical assumptions about why the new context is apparently less similar to the first one than we originally thought. Therefore, if an empirical finding does not replicate, we are more concerned about our understanding of the cases than about the validity of our theory.

Whereas statistical generalization provides us with a formal (and thus somehow more objective) apparatus to evaluate the universal validity of our theories, the bottom-up strategy forces us to think about the class of intended applications on theoretical grounds. This means that we have to ask: what are the boundary conditions of our theory? In the above classroom example, following a bottom-up strategy, we would build on our preliminary understanding of the cases in one context (e.g., a public school) to search for similar and contrasting cases in other contexts (e.g., a private school). We would then re-evaluate our theoretical description of the data and explore what makes cases similar or dissimilar with regard to our theory. This enables us to expand the class of intended applications alongside with the theory.

Of course, none of these strategies is superior per se . Nevertheless, they rely on different assumptions and may thus be more or less adequate in different contexts. The statistical strategy relies on the assumption of a universal population and invariant measurements. This means, we assume that (a) all samples are drawn from the same population and (b) all variables refer to the same behavioral classes. If these assumptions are true, statistical generalization is valid and therefore provides a valuable tool for the testing of empirical theories. The bottom-up strategy of generalization relies on the idea that contexts may be classified as being more or less similar based on characteristics that are not part of the model being evaluated. If such a similarity relation across contexts is feasible, the bottom-up strategy is valid, as well. Depending on the strategy of generalization, replication of empirical research serves two very different purposes. Following the (top-down) principle of generalization by deduction from scientific laws, replications are empirical tests of the theory itself, and failed replications question the theory on a fundamental level. Following the (bottom-up) principle of generalization by induction to similar contexts, replications are a means to explore the boundary conditions of a theory. Consequently, failed replications question the scope of the theory and help to shape the set of intended applications.

We have argued that quantitative and qualitative research are best understood by means of the structure of the employed models. Quantitative science mainly relies on variable-based models and usually employs a top-down strategy of generalization from an abstract population to individual cases. Qualitative science prefers case-based models and usually employs a bottom-up strategy of generalization. We further showed that failed replications have very different implications depending on the underlying strategy of generalization. Whereas in the top-down strategy, replications are used to test the universal validity of a model, in the bottom-up strategy, replications are used to explore the scope of a model. We will now address the implications of this analysis for psychological research with regard to the problem of replicability.

Modern day psychology almost exclusively follows a top-down strategy of generalization. Given the quantitative background of most psychological theories, this is hardly surprising. Following the general structure of variable-based models, the individual case is not the focus of the analysis. Instead, scientific laws are stated on the level of an abstract population. Therefore, when applying the theory to a new context, a statistical sampling model seems to be the natural consequence. However, this is not the only possible strategy. From a logical point of view, there is no reason to assume that a quantitative law like ∀ i : y i = f ( x i ) implies that the law is necessarily true, i.e.,: □(∀ i : y i = f ( x i )). Instead, one might just as well define the scope of the theory following an inductive strategy. 8 Formally, this would correspond to the assumption that the observed law is possibly true, i.e.,: ◇(∀ i : y i = f ( x i )). For example, we may discover a functional relation between “engagement” and “distraction” without referring to an abstract universal population of students. Instead, we may hypothesize under which conditions this functional relation may be valid and use these assumptions to inductively generalize to other cases.

If we take this seriously, this would require us to specify the intended applications of the theory: in which contexts do we expect the theory to hold? Or, equivalently, what are the boundary conditions of the theory? These boundary conditions may be specified either intensionally, i.e., by giving external criteria for contexts being similar enough to the ones already studied to expect a successful application of the theory. Or they may be specified extensionally, by enumerating the contexts where the theory has already been shown to be valid. These boundary conditions need not be restricted to the population we refer to, but include all kinds of contextual factors. Therefore, adopting a bottom-up strategy, we are forced to think about these factors and make them an integral part of our theories.

In fact, there is good reason to believe that bottom-up generalization may be more adequate in many psychological studies. Apart from the pitfalls associated with statistical generalization that have been extensively discussed in recent years (e.g., p-hacking, underpowered studies, publication bias), it is worth reflecting on whether the underlying assumptions are met in a particular context. For example, many samples used in experimental psychology are not randomly drawn from a large population, but are convenience samples. If we use statistical models with non-random samples, we have to assume that the observations vary as if drawn from a random sample. This may indeed be the case for randomized experiments, because all variation between the experimental conditions apart from the independent variable will be random due to the randomization procedure. In this case, a classical significance test may be regarded as an approximation to a randomization test (Edgington and Onghena, 2007 ). However, if we interpret a significance test as an approximate randomization test, we test not for generalization but for internal validity. Hence, even if we use statistical significance tests when assumptions about random sampling are violated, we still have to use a different strategy of generalization. This issue has been discussed in the context of small-N studies, where variable-based models are applied to very small samples, sometimes consisting of only one individual (Dugard et al., 2012 ). The bottom-up strategy of generalization that is employed by qualitative researchers, provides such an alternative.

Another important issue in this context is the question of measurement invariance. If we construct a variable-based model in one context, the variables refer to those behaviors that constitute the underlying empirical relational structure. For example, we may construct an abstract measure of “distraction” using the observed behaviors in a certain context. We will then use the term “distraction” as a theoretical term referring to the variable we have just constructed to represent the underlying empirical relational structure. Let us now imagine we apply this theory to a new context. Even if the individuals in our new context are part of the same population, we may still get into trouble if the observed behaviors differ from those used in the original study. How do we know whether these behaviors constitute the same variable? We have to ensure that in any new context, our measures are valid for the variables in our theory. Without a proper measurement model, this will be hard to achieve (Buntins et al., 2017 ). Again, we are faced with the necessity to think of the boundary conditions of our theories. In which contexts (i.e., for which sets of individuals and behaviors) do we expect our theory to work?

If we follow the rationale of inductive generalization, we can explore the boundary conditions of a theory with every new empirical study. We thus widen the scope of our theory by comparing successful applications in different contexts and unsuccessful applications in similar contexts. This may ultimately lead to a more general theory, maybe even one of universal scope. However, unless we have such a general theory, we might be better off, if we treat unsuccessful replications not as a sign of failure, but as a chance to learn.

Author Contributions

MB conceived the original idea and wrote the first draft of the paper. MS helped to further elaborate and scrutinize the arguments. All authors contributed to the final version of the manuscript.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.


We would like to thank Annette Scheunpflug for helpful comments on an earlier version of the manuscript.

1 A person × behavior matrix constitutes a very simple relational structure that is common in psychological research. This is why it is chosen here as a minimal example. However, more complex structures are possible, e.g., by relating individuals to behaviors over time, with individuals nested within groups etc. For a systematic overview, compare Coombs ( 1964 ).

2 This notion of empirical content applies only to deterministic models. The empirical content of a probabilistic model consists in the probability distribution over all possible empirical structures.

3 For example, neither the SAGE Handbook of qualitative data analysis edited by Flick ( 2014 ) nor the Oxford Handbook of Qualitative Research edited by Leavy ( 2014 ) mention formal approaches to category formation.

4 Note also that the described structure is empirically richer than a nominal scale. Therefore, a reduction of qualitative category formation to be a special (and somehow trivial) kind of measurement is not adequate.

5 It is possible to extend this notion of empirical content to the probabilistic case (this would correspond to applying a latent class analysis). But, since qualitative research usually does not rely on formal algorithms (neither deterministic nor probabilistic), there is currently little practical use of such a concept.

6 We do not elaborate on abductive reasoning here, since, given an empirical relational structure, the concept can be applied to both types of models in the same way (Schurz, 2008 ). One could argue that the underlying relational structure is not given a priori but has to be constructed by the researcher and will itself be influenced by theoretical expectations. Therefore, abductive reasoning may be necessary to establish an empirical relational structure in the first place.

7 We shall not elaborate on the metaphysical meaning of possible worlds here, since we are only concerned with empirical theories [but see Tooley ( 1999 ), for an overview].

8 Of course, this also means that it would be equally reasonable to employ a top-down strategy of generalization using a case-based model by postulating that □(∃ i : XYZ i ). The implications for case-based models are certainly worth exploring, but lie beyond the scope of this article.

  • Agresti A. (2013). Categorical Data Analysis, 3rd Edn. Wiley Series In Probability And Statistics . Hoboken, NJ: Wiley. [ Google Scholar ]
  • Borsboom D. (2005). Measuring the Mind: Conceptual Issues in Contemporary Psychometrics . Cambridge: Cambridge University Press; 10.1017/CBO9780511490026 [ CrossRef ] [ Google Scholar ]
  • Braun V., Clarke V. (2006). Using thematic analysis in psychology . Qual. Res. Psychol . 3 , 77–101. 10.1191/1478088706qp063oa [ CrossRef ] [ Google Scholar ]
  • Buntins M., Buntins K., Eggert F. (2017). Clarifying the concept of validity: from measurement to everyday language . Theory Psychol. 27 , 703–710. 10.1177/0959354317702256 [ CrossRef ] [ Google Scholar ]
  • Carnap R. (1928). The Logical Structure of the World . Berkeley, CA: University of California Press. [ Google Scholar ]
  • Coombs C. H. (1964). A Theory of Data . New York, NY: Wiley. [ Google Scholar ]
  • Creswell J. W. (2015). A Concise Introduction to Mixed Methods Research . Los Angeles, CA: Sage. [ Google Scholar ]
  • Dugard P., File P., Todman J. B. (2012). Single-Case and Small-N Experimental Designs: A Practical Guide to Randomization Tests 2nd Edn . New York, NY: Routledge; 10.4324/9780203180938 [ CrossRef ] [ Google Scholar ]
  • Edgington E., Onghena P. (2007). Randomization Tests, 4th Edn. Statistics. Hoboken, NJ: CRC Press; 10.1201/9781420011814 [ CrossRef ] [ Google Scholar ]
  • Everett J. A. C., Earp B. D. (2015). A tragedy of the (academic) commons: interpreting the replication crisis in psychology as a social dilemma for early-career researchers . Front. Psychol . 6 :1152. 10.3389/fpsyg.2015.01152 [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
  • Flick U. (Ed.). (2014). The Sage Handbook of Qualitative Data Analysis . London: Sage; 10.4135/9781446282243 [ CrossRef ] [ Google Scholar ]
  • Freeman M., Demarrais K., Preissle J., Roulston K., St. Pierre E. A. (2007). Standards of evidence in qualitative research: an incitement to discourse . Educ. Res. 36 , 25–32. 10.3102/0013189X06298009 [ CrossRef ] [ Google Scholar ]
  • Ganter B. (2010). Two basic algorithms in concept analysis , in Lecture Notes In Computer Science. Formal Concept Analysis, Vol. 5986 , eds Hutchison D., Kanade T., Kittler J., Kleinberg J. M., Mattern F., Mitchell J. C., et al. (Berlin, Heidelberg: Springer Berlin Heidelberg; ), 312–340. 10.1007/978-3-642-11928-6_22 [ CrossRef ] [ Google Scholar ]
  • Ganter B., Wille R. (1999). Formal Concept Analysis . Berlin, Heidelberg: Springer Berlin Heidelberg; 10.1007/978-3-642-59830-2 [ CrossRef ] [ Google Scholar ]
  • Guttman L. (1944). A basis for scaling qualitative data . Am. Sociol. Rev . 9 :139 10.2307/2086306 [ CrossRef ] [ Google Scholar ]
  • Hogg R. V., Mckean J. W., Craig A. T. (2013). Introduction to Mathematical Statistics, 7th Edn . Boston, MA: Pearson. [ Google Scholar ]
  • Hughes G. E., Cresswell M. J. (1996). A New Introduction To Modal Logic . London; New York, NY: Routledge; 10.4324/9780203290644 [ CrossRef ] [ Google Scholar ]
  • Klein R. A., Ratliff K. A., Vianello M., Adams R. B., Bahník Š., Bernstein M. J., et al. (2014). Investigating variation in replicability . Soc. Psychol. 45 , 142–152. 10.1027/1864-9335/a000178 [ CrossRef ] [ Google Scholar ]
  • Krantz D. H., Luce D., Suppes P., Tversky A. (1971). Foundations of Measurement Volume I: Additive And Polynomial Representations . New York, NY; London: Academic Press; 10.1016/B978-0-12-425401-5.50011-8 [ CrossRef ] [ Google Scholar ]
  • Leavy P. (2014). The Oxford Handbook of Qualitative Research . New York, NY: Oxford University Press; 10.1093/oxfordhb/9780199811755.001.0001 [ CrossRef ] [ Google Scholar ]
  • Maxwell S. E., Lau M. Y., Howard G. S. (2015). Is psychology suffering from a replication crisis? what does “failure to replicate” really mean? Am. Psychol. 70 , 487–498. 10.1037/a0039400 [ PubMed ] [ CrossRef ] [ Google Scholar ]
  • Miles M. B., Huberman A. M., Saldaña J. (2014). Qualitative Data Analysis: A Methods Sourcebook, 3rd Edn . Los Angeles, CA; London; New Delhi; Singapore; Washington, DC: Sage. [ Google Scholar ]
  • Open Science, Collaboration (2015). Estimating the reproducibility of psychological science . Science 349 :Aac4716. 10.1126/science.aac4716 [ PubMed ] [ CrossRef ] [ Google Scholar ]
  • Popper K. (1935). Logik Der Forschung . Wien: Springer; 10.1007/978-3-7091-4177-9 [ CrossRef ] [ Google Scholar ]
  • Ragin C. (1987). The Comparative Method : Moving Beyond Qualitative and Quantitative Strategies . Berkeley, CA: University Of California Press. [ Google Scholar ]
  • Rihoux B., Ragin C. (2009). Configurational Comparative Methods: Qualitative Comparative Analysis (Qca) And Related Techniques . Thousand Oaks, CA: Sage Publications, Inc; 10.4135/9781452226569 [ CrossRef ] [ Google Scholar ]
  • Scheunpflug A., Krogull S., Franz J. (2016). Understanding learning in world society: qualitative reconstructive research in global learning and learning for sustainability . Int. Journal Dev. Educ. Glob. Learn. 7 , 6–23. 10.18546/IJDEGL.07.3.02 [ CrossRef ] [ Google Scholar ]
  • Schurz G. (2008). Patterns of abduction . Synthese 164 , 201–234. 10.1007/s11229-007-9223-4 [ CrossRef ] [ Google Scholar ]
  • Shrout P. E., Rodgers J. L. (2018). Psychology, science, and knowledge construction: broadening perspectives from the replication crisis . Annu. Rev. Psychol . 69 , 487–510. 10.1146/annurev-psych-122216-011845 [ PubMed ] [ CrossRef ] [ Google Scholar ]
  • Smith P. (2020). An Introduction To Formal Logic . Cambridge: Cambridge University Press. 10.1017/9781108328999 [ CrossRef ] [ Google Scholar ]
  • Suppes P., Krantz D. H., Luce D., Tversky A. (1971). Foundations of Measurement Volume II: Geometrical, Threshold, and Probabilistic Representations . New York, NY; London: Academic Press. [ Google Scholar ]
  • Tooley M. (Ed.). (1999). Necessity and Possibility. The Metaphysics of Modality . New York, NY; London: Garland Publishing. [ Google Scholar ]
  • Trafimow D. (2018). An a priori solution to the replication crisis . Philos. Psychol . 31 , 1188–1214. 10.1080/09515089.2018.1490707 [ CrossRef ] [ Google Scholar ]
  • Watanabe S. (2018). Mathematical Foundations of Bayesian Statistics. CRC Monographs On Statistics And Applied Probability . Boca Raton, FL: Chapman And Hall. [ Google Scholar ]
  • Wiggins B. J., Chrisopherson C. D. (2019). The replication crisis in psychology: an overview for theoretical and philosophical psychology . J. Theor. Philos. Psychol. 39 , 202–217. 10.1037/teo0000137 [ CrossRef ] [ Google Scholar ]

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

  • We're Hiring!
  • Help Center

paper cover thumbnail

Comparing Qualitative Research Designs

Profile image of Seda Khadimally

The rigor of a well-designed qualitative research study heavily relies on the extent to which those conducting the study are informed about the theory and practice of it. The more knowledgeable the researchers are about different research design methods, the more salient their research findings will be. As much important as it is to know the philosophical underpinnings or the researchers’ positioning themselves prior to a qualitative research design, it is especially critical to have a profound understanding of the major concepts vested in it. Such understanding includes—but is not limited to—when it is appropriate to employ qualitative research, types of this research method (i.e., phenomenology, grounded theory research, case study, ethnography, Delphi method, etc.), and other such necessary information rooted in how to effectively and efficiently design a qualitative study germane in the literature-specific. With the intent to shed light to these and a number of more topics pertinent to the conduct of qualitative research design, this paper will discuss the content of three particular research designs and their approaches to the type of question the design would answer, type of problem it would address, sample size and selection, type of data collected, how these data are analyzed, and how the results of these analyzed data are presented. Examining the insights of each design with a comparative look, the paper will elaborate on details pivotal for the reader to understand, compare, and contrast, in hopes that they can ultimately design a salient qualitative research study.

Related Papers

International Journal of Social Sciences Interdisciplinary Research

Prashant Astalin

comparing qualitative and quantitative research methods pdf

Changsong Wang

Education for Health: Change in Learning & Practice

Kelly Devers

Richard Baskas, Ed.D. Candidate

IJIP Journal

Qualitative research attempts to broaden and/or deepen our understanding of how things came to be the way they are in our social world. If the research question involves exploring how people experience something, or what their views are, exploring a new area where issues are not yet understood or properly identified. A qualitative research design is probably the most flexible of the various experimental techniques, encompassing a variety of accepted methods and structures. Here, five of the major qualitative research designs namely ethnography, phenomenology, case study, grounded theory, and narrative research has introduced. Descriptionsof all five qualitative research designs are given separately. The design of qualitative research provides for the learner to understand the difference between phenomenology and grounded theory or between ethnography and case study and also narrative research and it gives the knowledge about itself. Common sense and research both involve an attempt to understand various aspects of the world. However research, but arguably not common sense, involves an explicit, systematic approach to finding things out, often through a process of testing out preconceptions and researchers working in the social sciences: psychology, sociology, anthropology etc., interested in studying human behaviour and the social world inhabited by human beings, found increasing difficulty in trying to explain human behaviour in quantifiable, measurable terms. Qualitative research attempts to broaden and/or deepen our understanding of how things came to be the way they are in our social world. If the research question involves exploring how people experience something, or what their views are, exploring a new area where issues are not yet understood or properly identified (e.g. before developing questionnaire items), assessing whether a new service is implementable, looking at "real-life" context, or a sensitive topic where you need the flexibility to avoid causing distress, your team probably needs to discuss using qualitative methodology.

International Journal of Qualitative Methods


For students and novice researchers, the choice of qualitative approach and subsequent alignment among problems, research questions, data collection, and data analysis can be particularly tricky. Therefore, the purpose of this paper is to provide a concise explanation of four common qualitative approaches, case study, ethnography, narrative, and phenomenology, demonstrating how each approach is linked to specific types of data collection and analysis. We first introduce a summary and key qualities of each approach. Then, using two common research contexts, we apply each approach to design a study, enabling comparisons among approaches and demonstrating the internal consistency within each approach. Given the nuance and complexity of qualitative research, this paper provides an accessible starting point from which novice researchers can begin their journey of learning about, designing, and conducting qualitative research.

Gloria Thakane Leutle

Studies in Comparative International Development

Gerardo Munck

Nurul Gasmi

To understand educational research, you now have the map (the steps that exist in the process of research) and the different paths you can take (quantitative and qualitative). Now we will explore some distinguishing features along the qualitative research design. These features are the research designs you can use to collect, analyze, and interpret data using quantitative and qualitative research. Some of the research designs may be familiar; others may be new, such as how these paths can converge with two designs called mixed methods research and action research. The discussion of designs will provide a more advanced understanding of educational research on your journey.


Yunita Putri

American Journal of Otolaryngology

chandra sen

Proceedings of the Royal Society B: Biological Sciences

IML Donsldson

Discrete Mathematics

Andras Recski

Dyes and Pigments

hossein hosseini

International Journal of Computer Applications

Aayush Mitan Kumar DESAI

Violeta Koleva

A study on foreign language teaching anxiety of non-native speakers

Burcu Ünyılmaz

alfredo quintos calluchi

SSRN Electronic Journal

Simon Wilkie

Langenbeck's archives of surgery / Deutsche Gesellschaft für Chirurgie

Hans-Jörg Mischinger

Translational Cancer Research

Ylva Skånér

International Journal of Scientific Research in Computer Science, Engineering and Information Technology

Nithin Ramakrishnan

Adsorption Science & Technology


Taşra’nın Dış Duvarı

Ahmet Bozkurt

Human Esmaeili

Journal of Synchrotron Radiation

Farhad Saeidi

Waldecir Gonzaga

Optics Express

Alexander Tolmachev

The Journal of Thoracic and Cardiovascular Surgery

British Food Journal


Revista Internacional de Medicina y Ciencias de la Actividad Física y del Deporte

Juan Garcia

Occupational and environmental medicine

Faith Davis

Uwe Kersting

See More Documents Like This


  •   We're Hiring!
  •   Help Center
  • Find new research papers in:
  • Health Sciences
  • Earth Sciences
  • Cognitive Science
  • Mathematics
  • Computer Science
  • Academia ©2024


  1. Qualitative vs Quantitative Research: Differences and Examples

    comparing qualitative and quantitative research methods pdf

  2. Qualitative vs Quantitative Research: What's the Difference?

    comparing qualitative and quantitative research methods pdf

  3. Qualitative vs Quantitative Research: Differences and Examples

    comparing qualitative and quantitative research methods pdf

  4. Qualitative vs. Quantitative Research

    comparing qualitative and quantitative research methods pdf

  5. Qualitative vs. Quantitative Research: Definition and Types

    comparing qualitative and quantitative research methods pdf

  6. Qualitative vs. Quantitative Research

    comparing qualitative and quantitative research methods pdf


  1. What is the Difference between Quantitative and Qualitative Research?

  2. Qualitative and Quantitative Research Methods

  3. Quantitative Research

  4. Comparing Groups on a Quantitative Response Variable Part 18

  5. Comparing Groups on a Quantitative Response Variable Part 19

  6. Quantitative and Qualitative Methods: A Quick Overview


  1. (PDF) Comparing Qualitative and Quantitative Approaches

    Comparing Qualitative and Quantitative Approaches. The discussion will compare the characteristic of both kinds of research, including their. purpose, research question and hypothesis, data ...

  2. PDF Research Design and Research Methods

    quantitative, and mixed methods research as three major approaches to research in the social sciences. The first major section considers the role of research methods in each of these approaches. This discussion then provides a basis for comparing qualitative and quantitative research as the two traditional alternatives in social science research.

  3. PDF Comparing the Five Approaches

    interpretations, and the research report. Qualitative researchers have found it helpful to see at this point an overall sketch for each of the five approaches. From these sketches of the five approaches, we can identify fundamental differences among these types of qualitative research. Finally, we compare the five approaches relating the dimensions

  4. PDF Comparison of Quantitative and Qualitative Research Traditions ...

    Which research methods or tools will be used to collect and analyse data? This article takes into account these components when explaining the differences between quantitative and qualitative research approaches. Quantitative and qualitative research designs differ in terms of their epistemo logical, theoretical and methodological underpinnings.

  5. PDF Module 1 Qualitative Research Methods Overview

    Qualitative research is a type of scientific research. In general terms, scientific research consists of an investigation that: • seeks answers to a question. • systematically uses a predefined set of procedures to answer the question. • collects evidence. • produces findings that were not determined in advance.

  6. PDF CHAPTER 1 The Selection of a Research Approach

    tors will often lead to embracing a strong qualitative, quantitative, or mixed methods approach in their research. The philosophical beliefs are important to include in a study because a reader can best interpret the research know-ing the biases and the particular stance being taken by the researcher. We

  7. Comparison of Quantitative and Qualitative Research Traditions

    There has been much discussion about quantitative and qualitative approaches to research in different disciplines. In the behavioural and social sciences, these two paradigms are compared to reveal their relative strengths and weaknesses. But the debate about both traditions has commonly taken place in academic books.

  8. PDF Quantitative and Qualitative Approaches to Comparative Education

    Among quantitative types of research, Picciano mentions descrip-tive studies, correlational research, causal comparative research and ex-perimental studies. Qualitative research methods include ethnography, historical research and case study research. To explain the differences between these methods, Picciano compared them along the lines of pur-

  9. A Practical Guide to Writing Quantitative and Qualitative Research

    Unlike in quantitative research where hypotheses are usually developed to be tested, qualitative research can lead to both hypothesis-testing and hypothesis-generating outcomes.2 When studies require both quantitative and qualitative research questions, this suggests an integrative process between both research methods wherein a single mixed ...

  10. Qualitative vs. Quantitative Research

    When collecting and analyzing data, quantitative research deals with numbers and statistics, while qualitative research deals with words and meanings. Both are important for gaining different kinds of knowledge. Quantitative research. Quantitative research is expressed in numbers and graphs. It is used to test or confirm theories and assumptions.

  11. Qualitative vs. Quantitative Research: Comparing the Methods and

    One key piece of information to help determine whether to use a qualitative vs. quantitative research method is which phase of development the study is in. For example, if a project is in its early stages and requires more research to find a testable hypothesis, qualitative research methods might prove most helpful.

  12. Quantitative and Qualitative Approaches to Generalization and

    In contrast to quantitative methodologies, there have been little attempts to formalize qualitative research strategies (compare, however, Rihoux and Ragin, 2009). However, there are several statistical approaches to non-numerical data that deal with constructing abstract categories and establishing relations between these categories (Agresti ...

  13. PDF The Usefulness of Qualitative and Quantitative Approaches and Methods

    This paper will discuss the advantages and disadvantages of qualitative and quantitative research approaches and methods, evaluating their usefulness as well as any ethical considerations in relation to problem-solving instruction in science education curriculum with indication of the dominant approaches in the area.

  14. PDF Comparison of Qualitative, Quantitative, and Mixed Methods Research

    COMPARISON OF QUALITATIVE, QUANTITATIVE, AND MIXED METHODS RESEARCH QUALITATIVE METHODS Emerging methods Open-ended Interview, observation, document, and audiovisual Text and image analysis Themes, patterns interpretation Phenomenology, grounded theory, ethnography, case studies, narrative

  15. PDF Quantitative vs. Qualitative

    Introduction 0 Two main traditions 1 in research: Quantitative and Qualitative 0 Quantitative research = inferential research 0 Qualitative research = interpretive research 0 Both different in terms of goals, applications, sampling procedures, types of data, data analysis, etc. 0 Although different, they can be complementary of one another i.e., in mixed methods 2


    qualitative research, where close questions are part of a quantitative one. Even observations can be analyzed by numerical means (83% of the subjects have smiled when landing on page X). Given that, the term "qualitative or quantitative research methods" is flawed - the qualitative and quantitative terms are

  17. (PDF) Comparing Qualitative Research Designs

    A qualitative research design is probably the most flexible of the various experimental techniques, encompassing a variety of accepted methods and structures. Here, five of the major qualitative research designs namely ethnography, phenomenology, case study, grounded theory, and narrative research has introduced.