Lecture: How to examine scientific thinking

HOW TO EXAMINE SCIENTIFIC THINKING Scientific Thinking | Testable Ideas | Ethics in Research

Slide 1 Transcript

Welcome to the Research Methods course. In the module to follow, you will be introduced to many of the core concepts needed to conduct meaningful research activities and projects. Not everything will be contained in the activities you complete, so it is important to complete all the readings. You are encouraged to check out the additional resources available, and to do the optional activities to gain more complete familiarity with the research tools you will need. This course has another purpose, which is to prepare you for the graduate capstone project or thesis you will be completing as part of a degree program. To this end, there is a research proposal you will construct during the course.

So, we will begin with a brief refresher on the evolution of scientific thinking, how ideas become hypotheses and research questions, and some of the ethical considerations that apply to research activities, most notably with human participants.

Systematic Researchers of Record

Aristotle  al-Haytham Roger Bacon Galileo

Descartes Copernicus Kepler Newton

Slide 3 Transcript

Probably the first known systematic researcher of record was Aristotle in ancient Greece. He developed a method of induction, reasoning, observation, and measurement to develop his philosophy of disciplined inquiry. A bit later, an Islamic scientist named al-Haytham developed a method that included stating the problem, testing a hypothesis, interpreting a conclusion, and publishing results. This same method was used by Roger Bacon during the Renaissance. Galileo introduced the concept that by adding empirical data in repeated outcomes a theory may be substantiated. And others like Descartes, Copernicus, Kepler, and Newton brought these notions together so they operate fully as empirically grounded theory.

6 Principles of Scientific Thinking

Ruling out Rival Hypotheses

  Correlation Is Not Causation

  Falsifiability

Replicability Extraordinary Claims Require Extraordinary

Evidence

Occam's Razor

Slide 5 Transcript

There are 6 principles widely regarded as encompassing scientific thinking. The first is ruling out rival hypotheses. This systematic narrowing of potential explanations for something observed means eliminating alternate causes so there is only one explanation left standing. This is a fundamental aspect of medical research, for example. The second principle is that correlation does not indicate causation for what is observed. You probably have seen this expressed before and can intuitively understand that some things may appear to be connected but actually are coincidental, and one thing does not cause the other. The third principle is related to falsifiability. Here, the point is that something must be proven wrong to establish it is valid. This is what is fundamentally behind the concept for a hypothesis. You must reject the proposition that there is no difference to establish evidence there is a difference. A fourth principle is about replicability. This means a claim must be able to be tested or replicated again, which is central to the scientific method of inquiry. A fifth principle is that extraordinary claims require extraordinary evidence. For instance, when a claim is made that counters what is known and accepted, more convincing evidence must be produced to change perceptions. The last principle is about Occam’s Razor which is also called the law of parsimony, stated first in the 14th century. William of Ockham believed that where there are two or more explanations for something, and the simpler one is better since it makes fewer assumptions. These principles are the foundation for the scientific method.

Linear Trajectory for Conducting a Research Effort

Identify the Problem or Issue

Literature Review

Select Methodology

Conclusion & Recomended

Future Research

Completed in Capstone

Suitable for single variable research Less inclusion of outside Follows a critical path

Analysis

References

Slide 7 Transcript

Here is a familiar schematic, on the left, that shows the linear trajectory for conducting a research effort. It applies, more or less, to most research approaches. Note that the linear version suggests no outside influences beyond those already acknowledged and is perhaps more representative for experimental designs. The step-wise representation is likened to a funnel, with the end product reduced to what is deductively established as the outcome most likely to explain a hypothesis, for example.

Research Cycle for Conducting a Research Effort

Continuing search for explanations Accepts the relative temporary truth Facts are challenged and may change

Gather & Analyse

Think & Plan

Discover

Write & Publish

Share/ Impact

Slide 9 Transcript

The cyclical version suggests there is a less likely final solution and that the knowledge is continually morphing and evolving. This is especially applicable for research designs that build upon previous work, whether experimental or not. Both approaches accept that facts change, and what we thought we understood is continually evolving. In this regard, what is considered true today may be replaced with a new truth tomorrow. One problem in advancing our knowledge is that well established researchers and experts may become less open to alternate explanations as they become more convinced of the validity in their findings while pressing for deeper understanding. We also need researchers who readily embrace change and widen their efforts to cover their subject area more broadly to include additional factors and discover potential alternatives. Both groups advance our knowledge, although different research designs may be employed.

Comparison of Scientific Thinking

Qualitative Research Method Quantitative Research Method

Uses the scientific method to record observations as numeric data.

Allows for more objective analysis of observations.

Uses the scientific method to make observations, with conclusions drawn without reliance on statistical analysis.

Includes most of the research in behavioral sciences.

Uses the scientific method to record observations as numeric data.

Allows for more subjective interpretation of observations.

vs.

Slide 11 Transcript

What began as the scientific method of research was oriented toward a quantitative, or measurable, approach to investigating phenomena. The qualitative approach was more aligned with what had been the approach for centuries earlier, often seen as traditional philosophy. The fundamental difference between them being that the quantitative approach already starts with a presumption, while the qualitative approach is searching for the actual core question. More on this shortly. For literature reviews, the quantitative method focuses on what metrics are related to the research question, while qualitative literature reviews are more open and exploratory, searching for as many variables as possible to help explain something. The mixed methods, which will be discussed in another module, try and use the applicable parts of each approach to balance an inquiry. To this extent, the qualitative researchers seek to adopt some quantitative methodology to strengthen their designs and to fortify conclusions, and vice versa for the quantitative researchers.

Scientific Thinking 4 Goals of Science

1. Describe

Another categorization for types of research: Basic Research: Uses the scientific method to address theoretical issues about fundamental processes and underlying mechanisms.

Applied Research: Focus is on problems with practical implications that have the potential for immediate action.

Reciprocity: Use what is learned in basic research and move it to applied research, or test how applied research fits with theories to explain basic research.

2. Predict 3. Explain 4. Control

Slide 13 Transcript

The 4 goals shown here open the scientific investigation process, so it is tied less closely to the traditional scientific method. The qualitative approach has grown in acceptance and has expanded the processes to generate knowledge. To this extent, any of the 4 terms may apply to either approach – quantitative or qualitative. Privitera, in one of your textbooks, captures the terms basic and applied as the action figures for research. They apply to both approaches – quantitative and qualitative, although not equally. For example, a quantitative approach may predict, where a qualitative approach may describe. Another categorization of types of research is directed more toward intentions. Basic research is oriented toward building theory for prediction. Applied research, on the other hand, seeks to solve an immediate or current problem now. Some scholars have expressed this as why something happens versus what to do about it. We can use what is learned in theory and apply it to practical situations, or we can test how practical solutions fit with theories we use to explain that problem.

Testable Ideas

Literature Review Sources Examines recent (or historically significant) research studies, industry data, or research reports related to the proposed study

Avoids “re-inventing the wheel”, and determines who has reviewed this area in the past and what has already been tried (including supporting and non-supporting evidence)

Historical Data Historical Records Articles in Refereed Journals Substantial Sponsored Studies Verified Industry Reports Industry Analyses

Slide 15 Transcript

A review of the literature occurs before you develop a hypothesis or specific research question to test an idea. The purpose is to evaluate what has been established or studied in relation to your topic. Often, this provides guidance for you to start focusing on a particular aspect, often something that has not been resolved well, if at all. In considering sources, those that are reviewed independently or by peer experts have greater credibility. In this regard, open sources from blogs, industry statements, magazines, or other less scholarly sources have reduced credibility and should not be used to establish validity. In this way, you can import validity for your project or idea where the source closely aligns with the objective.

Testable Ideas

Hypothesis Theory Specific, testable claim or prediction about what you expect to observe given a set of circumstances (if—then) Requires numerical data to derive an outcome or prediction

Broad statement to account for existing body of knowledge (why) Provides unique predictions to extend that body of knowledge

Research Question Specific, testable claim or prediction Found more often in qualitative designs Starts with a question, uses literature to investigate

Slide 17 Transcript

A hypothesis is a statement about the relationship between variables - an if-then proposition. If particular conditions exist, then a predicted outcome is likely, to a certain degree of confidence. This requires numerical data or metrics to establish the conditions that support the conclusions and is often found in experimental designs. Among the challenges for researchers are to select variables that best operationally represent the concepts, determine how many will be considered, and design appropriate measures for them. The hypothesis guides the study, identifies relevant facts, suggests which form of research design is appropriate, and gives a framework to organize results. A research question is usually associated with qualitative designs and seeks to identify as many variables involved as possible. From this, more focused inquiries can follow. In this sense, the question is established first, then the literature review is conducted to expand the potential influences. Picture this more like an hour glass where many aspects are considered and move toward more specific questions, leading to expanded areas for further research. A theory is not something a graduate student typically undertakes. Theory does not establish truth necessarily, instead it is a generally accepted explanation for evidence and how it is understood. The process slowly builds support for a theory and generates several more hypotheses. Collectively, the results from associated studies further strengthen the theory.

Ethics in Research What is Ethics?

Norms for conduct that distinguish between acceptable and unacceptable behavior

The basic concepts and fundamental principles of decent human conduct

A system of moral principles

The rules of conduct recognized in respect to a particular class of human actions or a particular group, culture, etc.

Slide 19 Transcript

Often, the ethical principles that apply for scientific inquiry are established and monitored by professional associations or licensing and certification bodies. These seldom are crystal clear, and there is typically a degree of latitude and interpretation about what is acceptable or not. Still, there is an expectation that participating partners will strive to adhere to the spirit of the ethical principles and be guided by those within their profession. Some of the obvious concepts in one part of the world may not coincide with other parts or cultures. For this reason, it is very valuable to engage in research activities with participants from these differing cultures, since wider application and generalizability from study results can be useful in a wider context. One of the enduring aspects, when debating ethics, is the role of morals. Usually, moral principles apply more broadly within a society, whereas, professional ethics may be more limited in scope and applicability.

Ethics in Research Why Adhere to Ethics in Research?

Promotes the Aims of Research Prohibition Against

Knowledge

Avoidance of Error

Fabricating

Truth Falsifying

Misrepresenting Research Data.

vs.

Slide 21Transcript

While this may appear obvious, research results and conclusions must be credible. With the modern version of the scientific method, which was implemented formally in Germany around the 1850s, an approach was developed that provided an accepted process for replication and confirmation of the work by others. In laboratory settings, this can be approximated, however, in more natural settings the results can rarely be replicated. This gave rise to a growing interest in qualitative research which, while not an antithesis of the scientific method, approaches inquiry from a very different perspective. That said, one can begin to see where the concept of an immutable truth is elusive. Ethics in research also attempts to systematically reduce the influence of various types of error, so that conclusions become more robust and believable. Researchers can be tempted at times to misrepresent their actual findings for personal gain or to enhance reputations. To protect participants in research, and those who may be affected with research results (think pharmaceuticals), work across several disciplines and organizations may be involved. In collaborative efforts like these certain values are essential. These might include trust, accountability, mutual respect, and fairness. They carry over to matters of authorship, copyright, patents, and credit for contributions.

Ethics in Research

IRB Reviews For student or faculty research activity, the ERAU Institutional Review Board (IRB) must approve all research proposals that involve human subjects.

The IRB can move quickly on simple activities (video games)

Proposals Proposals are systematic investigations that may include:

Surveys and questionnaires Interviews and focus groups Evaluations of social or educational programs Cognitive and perceptual experiments

Full-scale review is required if procedures involve particular applications involving humans that present a risk, e.g.

Tobacco Sleep Biological Samples

Slide 23 Transcript

The IRB procedures are available at the ERNIE portal, and before you even consider doing research with human subjects you must become very well acquainted with the requirements. Such research is a systematic investigation and includes development, testing and evaluation, and is designed to contribute to generalizable knowledge. Here we see what activities are intended as systematic, meaning they have set procedures of collecting information. The IRB can move fairly quickly on more simple activities (such as nonviolent video games) and often a designated reviewer can accomplish the work. However, a full-scale review is required if procedures involve administration of things like drugs, tobacco, sleep deprivation, caffeine, food, alcohol, food coloring, biological samples, and such. Well, that’s it for this presentation. Please continue with the other activities in the module and have a great week.