Lecture Notes on Research Ethics (Chap 2-5) PDF

Chapter 2: Research in Ethics - IRBs will also help researchers estimate the degree of risk involved by their Research Ethics - a framework of values within studies. which we conduct research....

Chapter 2: Research in Ethics - IRBs will also help researchers estimate the degree of risk involved by their Research Ethics - a framework of values within studies. which we conduct research. Important Reasons Why Poorly Designed - help researchers identify actions we Research Can Be Unethical consider good and bad, and explain the - Student’s teachers’ and administrators; principles by which we make time will be taken from potentially more responsible decisions in actual beneficial educational experiences situations. - Poorly designed research can led to unwarranted and inaccurate The well-being of the individual research conclusions that may be damaging to participant is no less important than the search the society that directly or indirectly of knowledge pays for the research - Allocating time and money to poor Research that is harmful to participants is quality science will keep those finite undesirable even though it may increase resources from better-quality science wisdom Informed Consent – voluntary agreement of A researcher is legally responsible for what an individual, or his or her authorized happens to research participants representative, who has the legal capacity to give consent, and who exercises free Institutional Review Board - composed of power of choice without any form of researchers, evaluate research proposals to constraint or coercion to participate in make sure that they follow ethical standards. research. - administrative body established to Assent - used to express willingness to protect the rights and welfare of human participate in research by persons who are research subjects. by definition too young to give informed consent but are old enough to understand Two different levels of IRBs: the proposed research in general, its - Psychology expected risks and possible benefits, and - Full University the activities expected of them as subjects. o Student projects in psychology ➔ need at least 3 randomly At risk – increase the chance of harm selected psychologists ➔ to compared with not participating in the sign up their research that it is study. ethical Minimal Risk - do not increase the o If outside the university ➔ full likelihood of injury. level university Research Ethics Board (Philippines) which can The Belmont Report proposed three ethical be found in multiple institutions principles: Purpose of IRB - Respect for persons - individuals have - IRBS protect the safety of the research the right of self-determination (basis of participants. informed consent). - to decide whether a proposed study - Beneficence - minimize harm and increases participants’ research can maximize potential benefits (basis of cause physical and or psychological risk/benefit analysis) discomfort. - Justice - fairness in both the burdens and benefits of research.  when the research study involves use of Deception - broadly means the use of deception or incomplete disclosure. The deliberately misleading communication with debriefing provides participants with a full participants about research purposes. explanation of the hypothesis being tested, - may be used when it is the best way to procedures to deceive participants and the obtain information. reason(s) why it was necessary to deceive them. - may not be used to minimize the participants’ perception of risk or - At a minimum, the debriefing exaggerate their perception of potential statement should include the following: benefits. o Label the form as “Debriefing - Psychologists do not conduct a study Statement” involving deception unless they have o Study title determined the use of deceptive o PI name and contact techniques is justified by the study’s information for follow-up significant prospective scientific, questions educational, or applied value and that o Student researcher’s name and nondeceptive alternative procedures contact information if are not feasible. applicable, for follow-up questions Types of Deception o Thank participants for taking - Active Deception - involves the time to participate in the intentionally providing inaccurate or study false information to participants (e.g., o Explain what was being studied one study team member tells (i.e., purpose, hypothesis, aim) participants that they will they will be in lay terms. engaged in a cooperative task with other participants, but instead, they will Anonymity - means that subjects are not be interacting with other members of identified by name. the study team). Confidentiality - means that data are securely - Passive Deception - (i.e., deception by stored and only used for the purpose explained omission), or incomplete disclosure of to the subject. information applies when information about the real purpose or nature of the Animal Welfare - The human care and research is withheld from participants. treatment of animals. The care and treatment of animals in research is regulated by the Which steps must researchers take if deception Animal Welfare Act of 1996 (USA) is used? - Subjects must be allowed to withdraw - Institutional Animal Care and Use from the experiment at any time and Committee (IACUC) to evaluate animal should never face coercion to remain. research before it is conducted. - The experimenter should provide a - In compliance with Republic Act 8485, debriefing (full disclosure) after either or the Animal Welfare Act of 1998, and personal participation or the Department of Agriculture completion of the entire study. Administrative Order No. 40 on Rules and Regulations on the Conduct of Debriefing – It is an essential part of the Scientific Procedures Using Animals, informed consent process and is mandatory UPLB established the Institutional  Animal Care and Utilization Committee Plagiarism - misrepresenting someone’s “ideas, (IACUC) under the Office of the Vice words, or written work” as your own. Chancellor for Research and Extension - a form of fraud in which an individual (OVCRE) to help safeguard the welfare claims false credit for another’s ideas, of animals used in scientific procedures, words, or written work. research and instruction of concerned units in the University (Philippines, and University of the Philippines) Chapter 3: Alternatives to Experimentation: American Association for Laboratory Animal Nonexperimental Designs Sciences (AALAS) - published detailed training manuals that include information on adequate Nonexperimental approaches - do not create housing, sanitation, and nutrition for all animals levels of an independent variable nor randomly used in research assign subjects to these levels. - They are used where experiments are APA’s Committee on Animal Research and not ethical or possible, or where we Ethics (CARE) has been influential in want to test hypotheses in realistic establishing national guidelines for animal conditions. welfare and includes standards for animal care Internal Validity - the degree to which a among its ethical principles researcher can establish a causal relationship between the independent and dependent variables. Animal Rights Movement - The basic premise of - An experiment has high internal validity the APA guidelines for animal research is that when we can demonstrate that only the animal research is acceptable to further the antecedent conditions are responsible understanding of behavioral principles and for group differences in behavior. promote human welfare. - An internally valid experiment allows us to draw cause-and-effect Animal Rights - the position that sensitive conclusions. species (those that can feel pain and suffer) - Laboratory experiments are often have equal value and rights to humans. higher in internal validity because of their control of extraneous variables. Scientific Fraud – involves falsifying or - Researchers create levels of the IV and fabricating data use procedures like matching and - Reasons: A researcher random assignment to conditions. What are the main lines of defense against External validity - the degree to which research fraud? findings can be generalized to other settings and - The peer review process filters individuals. submitted manuscripts so that only 15 – 20 % of articles are printed. - Replication, where researchers attempt to reproduce the finding of other, is the second line of defense. - Competition by colleagues for scarce resources, while a cause of fraud, is the third line of defense Degree of Manipulation of Antecedent Deviant case analysis - researchers examine Conditions - assignment of subjects to differences between deviant and normal antecedent conditions created for the individuals to identify etiological factors experiment - Cases of deviant and normal - Experiments are typically high in degree individuals are compared for of manipulation; nonexperiments are significant differences. These usually low. differences may have important implications for the etiology, or origin, Degree of Imposition Units - how much you of the psychopathology in question. limit a subject’s responses on the DV - This approach may also be applied to - extent to which the researcher nonclinical issues such as social trends constrains, or limits, the responses a and adult morale. subject may contribute to the data. - E.g. Likert scales Retrospective data - recollections of past events that are collected in the present. Phenomenology - a subject's description of - This information may be compromised personal subjective experience. by faulty memory, current mood, and - personal experience as a source of data the retrieval cues that are present when - low in manipulation of antecedent you are asked to recall an event. conditions and low in imposition of units (low-low) Field studies - nonexperimental studies conducted in the field (real-life settings). The Case Studies - a researcher compiles a experimenter does not manipulate antecedent descriptive study of a subject's experiences, conditions. observable behaviors, and archival records kept - nonexperimental designs used in real- by an outside observer. life settings and include naturalistic - Serve five major purposes according to observation, unobtrusive measures, Kazdin (2003). participant observer studies, and o They are the source of surveys. inferences, hypotheses, and - range from low-low to low-high theories - antecedent conditions are not o They are source of developing manipulated in field studies, but the therapy techniques degree of constraint of responses varies o They allow a study of rare considerably from study to study phenomena - When they conduct naturalistic o They provide exceptions, or observations, researchers attempt to counterinstances, to accepted remain inconspicuous so that the ideas, theories or practices presence of an observer does not alter o They have persuasive and the behaviors they observe. motivational value - Case studies range from low-low to low- Systematic observation - researcher uses a high. prearranged strategy for recording observations - Limitations of cases studies in which each observation is recorded using o representativeness of sample specific rules or guidelines so that o completeness of data observations are more objective. An o reliance on retrospective data important feature of coding systems is that they would give the same results to different researchers. Coding systems - do not manipulate antecedent conditions or alter the naturally occurring Naturalistic observation - examines subjects’ behavior of subjects, but they clearly limit the spontaneous behavior in their actual types of responses that will be recorded, environments and may obtain more bringing these studies into the low-high. representative behavior than experiments. Observers need to learn and practice the coding - method can achieve high levels of system thoroughly before they can actually use external validity the system in a research setting, and the - when they conduct naturalistic results should be reproducible by different observation, researchers attempt to trained observers. Many such coding systems remain inconspicuous so that behaviors have been developed for observational observed are not altered by the research in a wide variety of situations. Most of presence of the observer these systems allow observations to be quantified and permit statistical analyses. - a scheme or list of mutually exclusive labels, categories, and so forth—each of which characterizes a coherent dimension of interest—used for classifying information obtained by observing others. Strange Situation – an observational technique that has been used by developmental psychologists which is used to assess the quality of attachments that very young children have forged with their parents or other caregivers - Eight parts designed to mimic activities that occur in real-world settings: o The experimenter introduces the parent and baby to the playroom and leaves o The parent sist while the baby plays o A stranger enters, sits, and talks to the parent Field Experiments - experiments conducted in o The parent leaves, the stranger real-life settings offers comfort if the baby is - a research method in psychology that upset. The stranger leaves takes place in a natural, real-world o The parent returns, greets the setting. It is similar to a laboratory baby, and offers comfort if the experiment in that the experimenter baby is upset. The stranger manipulates one or more independent leaves variables and measures the effects on o The parent leaves the room the dependent variable. o The stranger enters and offers - the participants are unaware they are comfort being studied, and the experimenter o The parent returns, greets the has less control over the extraneous baby, offers comfort if variables. necessary, and tries to interest - Example: Hofling’s Hospital Study on the baby in toys Obedience Reactivity - subjects alter their behavior when o participants’ oral or written they know that they are being observed. descriptions of their Participant-observer study - field observation in experiences which the researcher is part of the studied o accounts from literature, group. poetry, visual art, television, - This approach contrasts with naturalistic theatre, and previous observation, where the researcher does phenomenological (and other) not interact with research subjects to research avoid reactivity. The main problems are invasion of privacy, not Chapter 4: Alternatives to Experimentation: telling people that you are studying their Surveys and Interviews behavior, and pretending to be a group member. Survey research - obtains data about opinions, Pretending to be a group member (e.g., a attitudes, preferences, and behaviors using researcher pretending to be a weight lifter) is a questionnaires or interviews. serious problem that requires careful planning. - survey approach allows researchers to study private experience, which cannot Archival Study - descriptive method where be directly observed researchers reexamine data that were collected - Anonymous surveys can increase the for other purposes. accuracy of answers to sensitive - Example: universities collect a wealth of questions. data through surveys like the - Surveys can allow us to draw inferences Graduating Senior Questionnaire (GSQ) about the causes of behavior and can and interviews. complement laboratory and field experiments. Qualitative Research - obtains data consisting of - The survey approach does not allow us words instead of numbers. to test hypotheses about causal - Data obtained through self-reports, relationships because we do not personal narratives, and expression of manipulate independent variables and ideas, memories, feelings, and control extraneous variables. thoughts. - Qualitative research is invaluable in Major Steps in Constructing Surveys studying contextual phenomena, - Identify specific research objectives. behavior that can only be understood - Decide on the degree of imposition of within its context. units (degree of response restriction). - Decide how you will analyze the survey Paradigm shift – a fundamental change in the data. way people think and work within a discipline, field, or area of study. Closed questions (structured questions) - can be answered using a limited number of Empirical Phenomenology - rely on an alternatives and have a high imposition of units. experimenter’s private experiences or other experiential data Open-ended questions - require that o the researcher’s self-reflection participants respond with more than a yes or 1- on relevant experiences 10 rating and have a low imposition of units. Nominal Scale – assigns items to two or more - People differ in their willingness to distinct categories that can be named using a answer, position preference, and yea- shared feature, but does not measure their saying and nay-saying. magnitude Ordinal Scale – measures the magnitude of the Willingness to answer - the tendency to guess dependent variable using ranks, but does not or omit items when unsure. Usually comes into assign precise values play whenever questions require specific knowledge about facts or issues Interval Scale - measures the magnitude of the DV using equal intervals between values with no Position preference - selecting an answer based absolute zero point. on its position Ratio Scale - measures the magnitude of the Manifest content - the plain meaning of the dependent variable using equal intervals words printed on the page. While we expect between values and an absolute zero. subjects to respond to the manifest content of questionnaires, they may ignore it when answering questions about their feelings or What should be considered when creating attitudes. surveys? Yea-saying (Acquiescence) - agreeing with an Subjects decide to refuse to answer surveys item regardless of its manifest content. during the start or first few questions Nay-saying - disagreeing with an item regardless Engage subjects from the start by asking of its manifest content. interesting questions they will not mind answering Context effects - changes in question interpretation due to their position within a First survey question should be: survey. - relevant to the survey’s central topic - This problem is especially likely when - easy to answer two questions are related and not - interesting separated by buffer items (unrelated - answerable by most respondents questions) - closed format Social desirability response set - representing Whenever possible, use commonly used ourselves in a socially appropriate fashion when response options. Avoid value-laden questions responding to a question’s latent content that might make a response seem (underlying meaning) embarrassing. Structured Interviews – questions are asked the Response styles - tendencies to respond to same way each time. This provides more usable, questions or test items without regard to their quantifiable data actual wording - a tendency to respond to a test item or Unstructured Interviews - The interviewer can interview question in some explore interesting topics as they arise. These characteristic manner regardless of the data may not be usable for content analysis. content of the item or question Population - consists of all people, animals, or random sampling for the cluster of objects that share at least one characteristic. population. o E.g. An educational institution Sample - a subset of the population of interest has ten branches across the (the population we are studying) country with almost the Probability Sampling – Involves random number of students. If we want selection, allowing you to make strong statistical to collect some data regarding inferences about the whole group. facilities and other things, we can’t travel to every unit to Nonprobability Sampling - involves non- collect the required data. random selection based on convenience or Hence, we can use random other criteria, allowing you to easily collect sampling to select three or four data. branches as clusters. Advantages of probability sampling - A probability sample is more likely to represent the population (external validity) than a nonprobability sample. - We know the exact odds of members of the population being included in our sample. This tells us whom the sample represents. The four main probability sampling methods are: - simple random sampling – every member of the population has an equal chance of being selected. Your sampling frame should include the whole population. - systematic random sampling - Every member of the population is listed with a number, but instead of randomly generating numbers, individuals are chosen at regular intervals. - stratified random sampling - the total population is divided into smaller groups to complete the sampling process. The small group is formed based on a few characteristics in the population. - cluster sampling - the cluster or group of people are formed from the population set. The group has similar significatory characteristics. Also, they The four main nonprobability sampling have an equal chance of being a part of methods include: the sample. This method uses simple - quota sampling - Quota sampling relies on the non-random selection of a  predetermined number or proportion of Chapter 5: Alternatives to Experimentation: units. This is called a quota. Correlational and Quasi-Experimental Designs - convenience sampling - includes the individuals who happen to be most Quasi - “seeming like.” accessible to the researcher. - purposive sampling - judgement Quasi experiments - superficially resemble sampling, involves the researcher using experiments, but lack their required their expertise to select a sample that is manipulation of antecedent conditions and/or most useful to the purposes of the random assignment to conditions. research. - They may study the effects of - snowball sampling - used to recruit preexisting antecedent conditions—life participants via other participants. The events or subject characteristics—on number of people you have access to behavior. “snowballs” as you get in contact with - A quasi-experiment might compare the more people. incidence of Alzheimer’s disease in patients who used ibuprofen since age 50 and those who did not. - We should use quasi-experiments when we cannot or should not manipulate antecedent conditions. In experiments, researchers randomly assign subjects to antecedent conditions that they create. An experiment might randomly assign subjects to either daily ibuprofen or aspirin use, and then measure their incidence of Alzheimer’s. Pearson correlation coefficient - is used to calculate simple correlations (between two variables) and may be expressed as: r (50) = +.70, p =.001. - Correlation coefficients have four properties. linearity, sign, magnitude, and probability. Magnitude - the strength of the correlation coefficient, ranging from -1 to +1. Probability - likelihood of obtaining a correlation coefficient of this magnitude due to chance.