Conceptual Foundations of Research Methods

The slides for this week’s lecture can be found here

Week 2 - Conceptual Foundations: At the heart of all research and evaluation tasks are assumptions about “how we know what we know” - So to begin the class we will discuss the idea of Epistemology and Ontology. We will relate these concepts to research paradigms and theories and (hopefully) develop an appreciation for why these concepts inform the selection of a research method in LIS. The [slides] for this week’s lecture can be found here

Theory and Methods

Students in Library and Information Science classes, especially research methods classes, often complain that the content is too theoretical. I have taught at and attended four LIS schools and I have heard variations of the ‘theory vs praxis’ debate play out at each and every one. Often when I hear someone complain that a concept, a lecture, or an assignment is “too theoretical” what I think they mean instead is “too abstract” - the content lacks concrete examples and meaning in a person’s life.

My goal this first week of LIS 570 is to try to convince you there is nothing so practical as a good theory. Being clear about the underlying assumptions, beliefs, and values that drive our everyday work - and being able to anchor those assumptions, beliefs, and values to a particular research paradigm in our field is incredibly liberating. It makes our thoughts and our actions all the more clear to ourselves and to others. This is, in many ways, the power of theory. In order to understand and make use of this power we will talk first, abstractly :), about research paradigms. Then we will discuss how research paradigms differ from one another - based largely on how they consider questions of ontology, epistemology, and use of methods. Then, I’ll introduce some of the main research paradigms you’re likely to see in LIS.

Research Paradigms

A paradigm is a set of basic beliefs, or a “worldview” that is shared by a group of people ¹. Paradigms of research are therefore made up of people whom agree about what constitutes an important problem, how it should be investigated, and what theories or beliefs about their worldview matter to that investigation. Think of a paradigm as a way to describe people who think the same way about research. The same way a political party is a group of people who (roughly) think the same way about social, economic, and environmental issues.

It is rare that someone will come right out and tell you what research paradigm they align themselves with, but by reading their articles and asking questions you can likely guess a person’s worldview. With some practice this will become second nature to you - just like identifying someone based on their clothing (however superficial that may be) tells you something about their values, beliefs, etc.

However well they may describe someone’s beliefs a paradigm is far from a perfect way to categorize a researcher. In the same way that a person may not strictly align themselves with a political party - they may agree on some of the basic, or most important parts of a political party’s platform. Just like our political beliefs place us into certain parties, so too do our research beliefs place us into certain paradigms.

What Paradigms Are Made Of

If research paradigms are likeminded people what are they like-minded about, exactly? In general, they agree on ontological, epistemological and methodological concepts related to the world around us.

Let’s use a simple example to understand what paradigms are made of:

In your class introductions I asked you to take a short quiz about what makes a sandwich a sandwich. You judged different photos of food which had different aspects that we typically associate with a sandwich (bread, cheese, etc.). At the end you were given an identity based on how you judged the *structure** and *ingredients** that you believe make up a sandwich.

My results from the quiz called me an “ingredient purist, structure purist” - and according to the quiz that aligns me with others who are “Lawful Good”

Another way to say this is that based on my beliefs I am part of the Lawful Good sandwich paradigm - I align myself with other “purists” based on what we think *should** constitute a sandwich - it should be two distinct pieces of a bread-like substance, there should be distinct filling, and that filling should be layered, etc, etc.

For sandwiches, what we think matters about their classification are things like structure and ingredients. When we think about research - what we think matters is… well… matter. That is, what we care about is the world around us, what it is made up of, and how we can gain access to knowledge about the world around us - its people, its environment, its organization, its information, etc.

Ontology: Hot-dogs are Sandwiches

What exists in the world or what constitutes reality is a question of ontology. So, when we say that a researcher or a research paradigm has an ontological commitment what we mean is that they have a belief about reality, what exists, and what constitutes existence.

Less abstractly, think about our sandwich quiz for a moment: If we believed that what constitutes a sandwich is “two distinct pieces of a bread-like substance” then our ontology - the reality of our worldview - does not consider a hot-dog a sandwich. But, if we are align ourselves with a worldview where two pieces, or even bread, are not required for a valid sandwich - then we might consider a hot dog to be a sandwich.

‘Are hot-dogs an example of a sandwich?’ is a seemingly insignificant question. There is nothing of great substance that hinges upon whether or not a hot dog is a sandwich. But, ontological commitments - what we consider to be reality very much does matter in the world of social science research. If we can’t agree on what constitutes reality - what does it mean to exist, and what is truth - then how can we possibly be expected to read, interpret, and validate one another’s work?

Let’s use another example, but this time from the world of research: Let’s say Francois is a researcher designing a survey based on the use of privacy features in a web-browser. At the beginning of Francois’s survey she needs to collect demographic information (age, education level, gender, etc) so that later she can analyze the data based on a “sample” of her survey respondents. This sampling will allow Francois to say something like “50% of respondents age 20-25 don’t use a private window when viewing their email at school or work”.

Now, let’s assume that Francois has a worldview that denies there is a social construction of things like gender. Instead, Francois believes that if she simply asks his survey participants to answer that their Sex is male or female - she captures all of the necessary information she needs. Besides being sexist and designing a bad survey, what Francois’s ontological commitments requires is that she make certain research design decisions that will impact how she answers her research questions. Going even further, Francois’s ontological position shapes the way she presents her research to the world, and perpetuates a binary reporting of data as either male or female.

Hot-dogs and sandwiches don’t really matter, but research that samples based on distinctions between sex and gender absolutely does. And in short, ontology matters. What we believe is reality, and what we believe is valid interpretations of that reality are at the core of designing, conducting, and reporting research results.

Epistemology + Methods: HOW do we know?

If ontology is our belief about what exists in the world then the next question we should ask is how do we know what exists in the world? Put another way, if this is a class that is about research then presumably we aren’t just sitting around talking about what we think, but we are trying to prove that what we think is true.

How we know what we know is the concept of epistemology. For philosophers, epistemology is the central question of all scientific research. It cuts to the heart of empiricism and in particular the proposition that a person can know something.

This is going to be a bit abstract at first, but understanding the role of epistemology in research is incredibly helpful (I promise). In the philosophical account of epistemology there are three criteria for having valid knowledge (that is, proving that you know something):

1. True
2. Justified
3. Belief

For shorthand let’s call this TJB. To have knowledge of any thing we need to believe it is true, we have to be justified in this belief, and what we believe has to, ultimately, be true.

• I believe that the freezing point of water is 32 degrees celsius.
• I am justified by a huge amount of scientific literature - from chemistry to physics - that supports my belief.
• Even if I am extremely skeptical I can use a thermometer and measure the temperature of water until it reaches a solid state. My belief is true. I know that the water freezes at 32 degrees celsius.

Much of what we “know” about the natural world is based on accepted scientific findings (e.g. the freezing point of water). Over time we become more convinced that our justified beliefs are true - repeated experiments show over and over that something like the freezing point of water is a certain temperature, or we know that a vaccine works because it has to pass a stringent clinical trial that proves it has efficacious properties.

The social world is a bit more complicated than simply sticking a thermometer in a glass of water. Obtaining direct knowledge about things like human behaviors, actions, motivations, etc. is nearly impossible. In social science research our justification for claiming to know something is where most of our time and effort is focused.

How we justify beliefs that we hold is by using techniques to gather or tools to collect data. These techniques and tools are called research methods - the title of this class. What we’re going to investigate over the course of our 10 weeks together is how we justify our beliefs through the collection, analysis, and presentation of data.

TL;DR

Let’s quickly review what we’ve covered so far:

• Ontology is our belief about what exists in the world, and what constitutes reality
• Epistemology is how we gain knowledge about the world, and how we know what reality contains. In the epistemological account: knowledge = true + justified + beliefs (TJB)
• Methods are the ways that we find out or acquire knowledge to support theories and justify beliefs. In social science research we depend heavily on techniques and tools to justify our beliefs.
• A research paradigm is made up of people who share a worldview about reality, and this worldview if formalized by commitments to ontology, epistemology, and methodological practice.

Major Research Paradigms in LIS

In LIS there are roughly two major research paradigms ²: Positivism and Constructivism (or Interpretivism). Over the next two sections I’ll explain the commitments of positivists and constructivists, and then summarize this neatly in a table.

Positivism (and post-positivism)

Positivism is a belief in a singular truth or reality. For a positivist, true facts about society and social relations can be discovered, and facts about social reality can be measured and known absolutely.

A worldview that believes in one (and only one) truth requires careful and exacting research methods. Therefore, most positivists use techniques like surveys, behavioral data like social media posts, and log-data such as circulation records to form beliefs about the world. Coupled with these methods of data collection are methods of data analysis such as statistics which attempt to predict or describe patterns in data.

An example of positivist research in LIS might be in measuring how many times a book is checked out, or how often a research article is cited ³. By measuring this an LIS researcher can, for example, demonstrate which book was most important to a population over a period of time. For example, a data scientist at the New York Public Library analyzed circulation records and then used descriptive statistics to show that in 2019 the most checked out book in New York City was ‘Becoming’ by Michelle Obama.

Another example of positivist research can be found in our LIS spotlight reading from last week ⁴. The authors of this article conducted a survey of public libraries in Tennessee to understand how voter information is provided to patrons, and the impact of these civic engagement activities on voting. The positivism in this article is a bit covert, but the authors believe there is a measurable effect of providing information to potential voters (and if we had perfect knowledge of each and every voter and each and every library outreach activity) we can reliably discover which outreach activity is most effective for which kind of voter. The key here is understand that without perfect knowledge - that is by sampling a population instead of directly measuring the entire population - the positivist has to rely on statistics to make informed guesses as to the truth. In the words of epistemology - the justification for their beliefs are found in statistics which provide an approximate measurement in place of a true measurement.

In short, positivists believe that there is a single reality, that knowing reality is possible through measurement (or using tools like statistics to infer measurements), and data are collected by direct interaction with people (e.g. surveys) or by indirect / secondary interactions (e.g. traces of their behavior, such as circulation records of a public library).

Constructivism and Interpretivism

Unlike positivists, constructivistss believe there is no single reality or single truth to be discovered about the social world. Instead, constructivists believe that relative truth and facts about reality are socially constructed - they are the ideas and interpretations of people who collectively agree to their meanings.

A constructivist does not deny things that are the result of positivist research - like the freezing point of water - simply because they believe temperature measurements are a social construct. The natural world can have truths even if they are best explained by social constructs. A constructivist simply sees the social world as being too complex and too unwieldy to be explained by just one truth. Constructivists believe that reality needs to be interpreted and translated into numerous worldviews - each of which may have different meanings and implications for holding knowledge.

Constructivists tend to favor the use of qualitative over quantitative research methods ⁵ - because most constructivists reject the idea that a statistic can give a plausible answer to something complex that they are interested in asking. Instead, constructivists ask questions about why things happen the way that they do, or how things procedurally occur.

Let’s take an example from a future LIS research spotlight that will make the viewpoint of constructivists less abstract. In ‘Archival Amnesty: In Search of Black American Transitional and Restorative Justice ⁶’ Dr. Tonia Sutherland uses the conceptual framing of truth and reconciliation commissions to advocate for restorative justice in memory institutions. Comparing historical events to contemporary archival practices she also argues for conceptually revisiting both appraisal and amnesty of records in Archival Science. The writing is wholly within a constructivist paradigm: History, critical document analysis, and archival theory are all qualitative methods used to justify her beliefs. Put another way - her article is not arguing for any one single notion of truth, but instead it is circling the idea of justice and using historical information and contemporary examples as justification for beliefs about what is just and right ⁷.

Other Research Paradigms

The two most dominant and general paradigms in LIS (positivism and constructivism) can be refined, made more specific, and also differentiated. In particular, constructivists tends to differ on questions of epistemology - with some preferring to emphasize the role of power (e.g. queer theory, feminism, marxism) and others the importance of utility or greatest good (e.g. pragmatism). In readings this week you will have the chance to explore some more examples of research paradigms that are applicable to LIS. However, each of these descriptions of a paradigm is imperfect and necessarily incomplete. To cover even one of these paradigms we could spend an entire quarter reading the literature of just one and still be just scratching the surface. What is important to take away from this look into th conceptual underpinnings of research and assesment is this:

• Beliefs about reality and truth shape our research - in the choices we make about topics, questions that are important, and the ways we go about finding answers to our questions.
• There are not right or wrong research paradigms - each has a strength and a weakness that practitioners acknowledge. There are, however, valid and consistent ways of practicing within a paradigm. The rest of our course is going to be largely focused on articulating questions, and using methods to answer questions that fit within a paradigm of positivism or constructivism.
• If you don’t recognize your own beliefs in one of the paradigms we’ve briefly covered here don’t worry - it is rare that someone perfectly identifies with a research paradigm (just like it is rare that someone fully agrees with a political party). Over time these paradigms will start to become clearer and you will feel yourself more naturally accepting claims of or reading authors in one paradigm or the other.

Readings

• Brown, M. E., & Dueñas, A. N. (2020). A medical science educator’s guide to selecting a research paradigm: building a basis for better research. Medical Science Educator, 30(1), 545-553. PDF … Note this publication comes from the field of medical sciences, but the explanations of ontology, epistemology, and method choice are exceptionally clear and second to none.

• Robert Wood Johnson Foundation, (2017) ‘Qualitative Research Guidelines Project’ [Available here][http://www.qualres.org/HomePhil-3514.html]. (Choose one paradigm to read in-depth, and skim others)

LIS Research Spotlight

• Duarte, M. E., & Belarde-Lewis, M. (2015). Imagining: Creating spaces for indigenous ontologies. Cataloging & Classification Quarterly, 53(5-6), 677-702. PDF

Suggested

• Baronov, D. (2015). Conceptual foundations of social research methods. Routledge. UW Library Link - Note: This book is a thorough, but very accessible introduction to the paradigms discussed in this week’s lecture. The UW Library has print copies available (and if you request an electronic copy they will send you one)

• Bird, Alexander, “Thomas Kuhn”, The Stanford Encyclopedia of Philosophy (Winter 2018 Edition) - On the concept of a paradigm

Exercise

You should be forming groups or starting to explore topics for Assignment 1. In the assignment description there are two exercises that you can complete to get started on your first draft.