Research Methods
The systematic and objective identification, collection, analysis, dissemination and use of information for the purpose of decision making related to the identification and solution of problems.
Types Problem Identification Research that is undertaken to help identify problems that are not necessarily apparent on the surface and yet exist or are likely to arise in future. Problem Solving Research undertaken to help solve specific problems.
Basic vs. Applied Research: A Continuum Basic Research: Development of theory, principles and findings that generalize over a wide range of people, tasks, and settings. Applied Research: Development of theory, principles and findings that are relatively specific with respect to particular populations, tasks, products, systems and/or environments. 4 4
Characteristics of Basic Research Advantages Greater generality Conducted in controlled lab which prevents intrusion of potentially confounding variables third-variable sources of causation More confidence when drawing causal inferences between variables (the cause/effect relationship) Disadvantages Lab environments can often be simplistic and/or artificial And they may have little resemblance to the real world. 5 5
Characteristics of Applied Research Advantages More likely to resemble real-world conditions Can be more efficient in identifying design flaws and/or more effective alternative designs Disadvantages Harder to generalize from the specific test group studied to the larger “population of interest” Less control over potentially confounding events 6 6
Overview of Research Methods The Experimental Method What is it? Deliberately producing change in one or more independent variables (IVs) & measuring the effect of that change on one or more dependent variables (DVs) Manipulation performed while keeping all other variables under control The Goal: Control Showing that the IVs, and no other variables, are responsible for any measurable differences in the DVs Most direct way to ascertain cause/effect relationships “Quasi-experiment” – when not all other influential variables can be controlled. 7 7
Overview of Research Methods Descriptive Research What is it? The use of procedures used to characterize a population in terms of some relevant set of attributes. Criterion variables—Dependent variables including physical characteristics or performance measures Stratification variables—Predictors, including age, sex, and education Summary statistics include: Mean, median, mode, frequencies, percentiles Note: Cannot draw causal inferences from descriptive study results. 8 8
Overview of Research Methods: Evaluation Research What is it? Procedures used to assess the performance of people using a product or system. Widely used in government and industry. Attempts to answer relevant questions in a practical way. 9 9
A Model of the Research Process: Phases of Research Idea gathering Phase (Curiosity, interest, and enthusiasm) Definition of the Problem (Identify an area of interest, then clarify and refine) Design of the Procedures (Research design,subject selection, ethical considerations, type of observations, data analysis) Observation (“Doing the research”) Data-analysis (Statistical procedures should be appropriate to the question being asked and to the observation procedures being used. Should be specified in advance) Interpretation (How do the results help answer the research question; how does the data contribute to knowledge in the field) Communication/Publication (Purpose: (1) communication findings to appropriate audience; (2) allow replication of the study to strengthen confidence in the findings) 10 10
Significance of Research Research inculcates scientific and inductive thinking and it promotes the development of logical habits of thinking and organisation. Research provides the basis for nearly all government policies in our economic system. It has special significance in solving various operational and planning problems of business and industry.
It is important for social scientists in studying social relationships and in seeking answers to various social problems. For doctoral students, research may mean a careerism or a way to attain high position. To professionals in research, research may mean source of livelihood.
To philosophers and thinkers, it may mean the outlet for new ideas and insights. To analysts and intellectuals, research may mean the generalizations of new theories.
Research Methodology is a way to systematically solve the research problem. It may be understood as a science of studying how research is done scientifically.
Characteristics of Good Research Scientific method Multiple methods Interdependence Value and cost of information Systematic Logical 1- 15 15
Scientific Method is Devoid Of: Personal beliefs Perceptions Biases Values Attitudes Emotions This is the ideal but as you will see, it is very hard to achieve this is the psychological and social sciences. One of the reasons that the psychological and social sciences are sometimes referred to as the “soft sciences”.
Scientific Method – 5 Steps Asking the Question Problem Identification & Forming Hypothesis(es) Experimental Design & Methods Interpretation & Communication Testing the Hypothesis - Data analysis Collecting Relevant Information (data)
Basic Assumptions Underlying Science Determinism – belief that there are causes or determinants of behaviour. Reality – belief that there is an underlying reality or truth in nature. Rationality – events can be understood through the use of logical thinking. Regularity – events in nature follow same laws. Discoverability – it is possible to discover the uniformity or laws in nature. Determinism is the belief that behaviour is caused by specific events. Determinism assumes that there is uniformity in nature; that is, the nature is predictable, reproducible, real, and logical. What do these axioms mean? Axiom are self-evident truths. Science attempts to uncover the “reality in nature.” Rationality – events make sense, they are logical. For example, there are rational explanations for anxiety.
Approaches to Scientific Research: Essential Requirements Empirical. Gather evidence through observation and measurement. Measurement. The variables of interest are measured—through multiple techniques and perspective, if possible. Replicability. Variables are measured consistently, and in such a way as can be replicated by other researchers. Objectivity. It is important to approach research questions and interpret results without bias. 19 19
Characteristics of the Scientific Approach Control – eliminating the influence of extraneous variables that could affect the observations (confounds). Operational definition – defining the steps or operations used to measure the phenomenon (e.g., anxiety). Replication – Reproducibility of the results. Replication is where you demonstrate, or someone else demonstrates, that if the experiment were to be conducted in exactly the same way again, the results would be the same. Replication could be done by making intergroup, intersubject, or intrasubject observations. Define each (page 17). “Only through replication can we have any confidence that the results of our studies are valid and reliable.” Many grading systems now exist for grading confidence in the results of studies. The highest grades are assigned for results where the findings have been replicated across one or more studies; the sample sizes were large enough to minimize the risk of chance findings(e.g., outlier distorting the results).
Description Discover that the phenomenon exists Demonstrate that the phenomenon exists Describe its elements One method by which one might first discover & describe a phenomenon is through naturalistic observation. Another is through qualitative research methods. We will be talking about both these scientific research methods in subsequent lectures.
Explanation Why does the phenomenon exist What causes the phenomenon Causes are the antecedent conditions that result in the occurrence or manifestation of the phenomenon. Thinking back to the description of the person who was afraid to go on buses, can you think of some of the antecedent conditions for this person’s fear (e.g., thinking about going out in public; how crowded the space; ease of escape; opportunities for doing things to help keep one safe; etc.).
Scientific Theory Theory is the explanation for the phenomenon. Theories not only describe why or how the phenomenon occurred but also guides the way for further research. Theories are often referred to as models for a specific phenomenon. If the predictions stemming from a theory are proven “true” when subjected to scientific study than the theory proves useful in helping to explain the phenomenon under study. If not, the theory is very useful and must be revised.
Transtheoretical Model Precontemplation Preparation Contemplation Maintenance Action Relapse (Prochaska & DiClemente, 1992)
Where do the ideas come from? Everyday life Practical issues or needs Past research Theory
Sources of Research Questions Everyday life What are the most effective ways to teach research methods? What personal characteristics make a favorable impression in a job interview? What other questions come to mind? Generally, the more relevant your research question is to solving an everyday life problem, the more valuable the research.
Sources of Research Questions Practical Issues Why do some employees have very high absenteeism rates. Why do more car accidents happen on specific stretches of the road? Why are most heart attacks on a Monday morning? Like the previous example, research questions that solve real practical issues.
Sources of Research Questions Past research Knowledge develops in small steps. Rarely does one study answer all the questions to the research topic.
Sources of Research Questions Theories Summarize & integrate existing knowledge Suggests new relationships between factors Helps one make new predictions about a phenomenon based on the theory. What are some theories in psychology that have generated a lot of research?
Defining the Research Question Caveat – Not all ideas are subject to scientific enquire. Research question must be capable of being confirmed. The study must be feasible. Variables in the problem should express a relationship: E.g., What relationship exists between two or more variables? Ideas can be born of luxurious thinking, personal beliefs, conceptions, suppositions, assumptions, what ifs, guesses, etc. How do you turn an idea into a researchable question? With respect to not all questions are researchable, a research question must be one that is capable of being confirmed or refuted.
Defining the Research Question - continued Problem should be capable of being stated in a question form. Examples are: What is the effect of….? Under what conditions do….? Does the effect of….? A research question defines the area of interest but is not a declarative statement like a hypothesis. With respect to not all questions are researchable, a research question must be one that is capable of being confirmed or refuted.
Defining Research Problem and objectives It is important to define research problem appropriately because it serves as a guideline to the researcher for conducting the rest of research project 1- 32 32
Chain Restaurant Study One day I received a phone call from a research analyst who introduced himself as one of our alumni. He was working for a restaurant in Colombo and wanted help analyzing the data he had collected while conducting a marketing research study. When we met, he presented me with a copy of the questionnaire and asked how he should analyze the data. My first question to him was: What is the problem being addressed? 1- 33 33
Chain Restaurant Study When he looked confused, I explained that data analysis is not an independent exercise. Rather, the goal of data analysis is to PROVIDE INFORMATION RELATED TO THE PROBLEM COMPONENTS. 1- 34 34
Chain Restaurant Study I was surprised to learn that he did not have a clear understanding of the marketing research problem and that a written definition did not exist. So before going any further, I had to define the marketing research problem. Once that was done, I found that much of the data collected was not relevant to the problem. In this sense, the whole study was a waste of resources. A new study had to be designed and implemented to address the problem defined. 1- 35 35
Symptom A symptom occurs as a result of a problem. It is often a complicated process to distinguish a symptom from a problem 1- 36 36
Symptoms and Causes 1- 37 37
Formulating Hypotheses Stated in declarative form. Posits a relationship between variables. Ideally reflects a theory or body of literature. Is brief and to the point. Is testable.
Hypotheses Scientific hypothesis states the ‘predicted’ relationship amongst the variables. Null hypothesis is a statement of no relationship amongst the variables. You can only test the null hypothesis. “The scientific hypothesis does not specify the exact amount or type of influence expected (and so cannot be directly tested).” (page 97) Thus, support for the scientific hypothesis is obtained indirectly by rejecting the null hypothesis.
Examples of Hypotheses Research Idea Question Hypothesis Drug abuse and child abuse Is drug abuse related to child abuse? There is a positive relationship between drug abuse among adults and their physical and psychological abuse as children.
Does eating pizza make you fat? How will we operationally define our measures? Does eating pizza make you fat? The experimental hypothesis: The more pizza a person eats, the more weight she/he will gain. The null hypothesis: a person’s weight will be the same regardless of how much pizza they eat. Does drinking beer and eating cheese-fries give you nightmares? The experimental hypothesis: People who consume beer and cheese-fries before bedtime will report more nightmares than those who do not. The null hypothesis: People who consume beer and cheese-fries before bedtime will report about the same number of nightmares as those who do not. 41 41