Research Ethics PDF

Lesson 2 experimental  An IRB is crucial in protecting individual rights by ensuring that each at-risk subject gives Research ethics informed consent to...

Lesson 2 experimental  An IRB is crucial in protecting individual rights by ensuring that each at-risk subject gives Research ethics informed consent to participate in a study, after Responsible Psychological Research being fully informed about the  Prioritizes ethical and study's nature. responsible treatment of subjects, ensuring their safety Research ethics and welfare.  Aims to advance  Robert Rosenthal argues that understanding of feelings, poorly designed research can thoughts, and behaviors for be unethical due to time and humanity's benefit. resource allocation, leading to  Researchers are legally inaccurate conclusions, responsible for any harm to damaging society, and wasting participants, including valuable resources on poor- unintentional harm. quality science.  Researchers may face lawsuits for damages if experiments cause harm CONSENT FORM  Federal government has formulated legal and ethical  Consent must be given freely, guidelines to protect without force or coercion. psychological research  Experiment participants can subjects. withdraw at any time.  Researchers must provide Institutional review board or irb detailed explanations of procedures and answer any  Determines if proposed study questions. puts subjects at risk.  Potential risks and benefits  Conducts risk/benefit analysis must be clearly communicated, to determine if individual risks including potential pain or outweigh potential benefits or injury. knowledge importance.  Data confidentiality must be  Requires understanding of assured. research design and expertise  Subjects cannot be asked to in research methods. waive their legal rights in case  of injury, as per federal  Refers to fairness in guidelines (45CFR 46.115). research burdens and benefits.  Examples include the BELMONT REPORT Tuskegee syphilis study,  "Belmont Report: U.S. Policy on which used poor, rural black Human Subject Research" men for research.  Statement of government  Requires researchers to policy on human subjects. select subjects fairly,  Includes three ethical avoiding disadvantaged or principles: respect for persons, easily manipulated beneficence, and justice. individuals.  Respect for Persons Deception and Full Disclosure  Upholds the autonomy of every individual in research  The relationship between decisions. researcher and participants  Provides protections for should be as open and honest vulnerable populations and as possible. In some individuals with diminished capacity. psychological studies, however,  Forms the basis for the true purpose of the study is informed consent. disguised.  Beneficence APA Standard: Deception in  Obligations to minimize Research harm and maximize benefits Psychological Research to individuals. Deception Policy  Forms the basis for  Psychologists only use risk/benefit analysis in IRB deceptive techniques in studies approval. justified by significant  Requires researchers to scientific, educational, or estimate potential risks applied value. truthfully and accurately.  Deception about research causing physical pain or severe  Justice  emotional distress is  Other identifying details prohibited. should be disguised if a  Deception integral to subject is recognizable. experiment design and conduct should be explained to participants early, preferably at Lesson 3 the end of participation. What are statistics?  Participants can withdraw their data at any time.  Statistics is the science of conducting studies to collect, Anonymity and Confidentiality organize, summarize, analyze, and draw conclusions from Research Anonymity and data. Confidentiality in Online Research  Researchers must protect participants' privacy by collecting data anonymously and identifying subjects by code numbers or fictitious names.  Most psychological research uses aggregated or group `Why do we study statistics? data and reports statistical results as average scores for  Statistics are crucial for each treatment group. interpreting large amounts of  Data must be stored data, organizing numbers into securely, kept confidential, interpretable forms, and saving and used only for purposes time and money for explained to the subject. participants, ourselves, and  Subjects' reactions should communities. not become gossip items to be shared with friends. Descriptive Statistics  Data sharing with colleagues should be treated with  Descriptive statistics is a discretion and subjects' method used to describe and identities protected. summarize data in a  meaningful way. It involves as the impact of tutoring on choosing a group, recording test scores. data, and using summary statistics and graphs to  Variable being measured describe group properties. This  May change based on other method allows for insights and variables visualization of both entire  Depends on other variables populations and individual samples, eliminating uncertainty in inferring Independent variables properties about large data  An independent variable in an sets. experiment, such as tutoring, remains constant despite other Inferential Statistics variables, while the dependent  Inferential statistics involves variable (test scores) may making predictions about a change. large group of data using a representative sample. This  Variable being technique uses a random manipulated sample to describe and infer  Doesn't change based on the population, often in the other variables form of probability. The  Stands on its own accuracy of inferential statistics relies on the sample data's Levels of independent variable representation of the larger population. Qualitative and quantitative variables Types of data and how to collect them Qualitative variables  Qualitative variables, such as hair color, eye color, religion, Dependent variables and gender, express attributes without numerical ordering,  The dependent variable is the often referred to as categorical variable being measured or variables. tested in an experiment, such Quantitative variables differences between the ranks do not exist.  Quantitative variables, such as age, heights, weights, and body  The interval level of temperatures, are numerical measurement ranks data, and and can be ranked, and can be precise differences between classified into discrete and units of measure do exist; continuous groups. however, there is no DISCRETE AND CONTINUOUS meaningful zero. VARIABLES  The ratio level of measurement  Discrete variables assume possesses all the characteristics values that can be counted. of interval measurement, and there exists a true zero. In  Continuous variables can addition, true ratios exist when assume an infinite number of the same variable is measured values between any two on two different members of specific values. They are the population. obtained by measuring. They Sampling techniques often include fractions and decimals. Probability sampling Measurement Scales  Probability Sampling Overview  Involves randomly selecting  Nominal level of measurement a small group from a larger classifies data into mutually population. exclusive (non overlapping)  Predicts the likelihood of all categories in which no order or responses matching the ranking can be imposed on the overall population.  Requires an equal, non-zero data. chance of selection for all  The ordinal level of individuals in the population measurement classifies data  Each person's chance of into categories that can be selection must be ranked; however, precise represented as a probability.  Systematic sampling Types of probability sampling  Systematic sampling, also Simple random sampling known as interval sampling, involves assigning a number to  Simple random sampling each member of a population ensures equal chance of and selecting them at regular selection from the population intervals to form a sample. using tools like random number generators. However, it is prone to bias, as smaller Advantages of probability sampling sample sizes reduce the likelihood of drawing a reliable  Probability sampling offers sample at random. cost-effective, user-friendly, Stratified random sampling and geographically dispersed benefits for large audiences,  stratified random sampling reducing bias, and limiting involves dividing populations variability, making it ideal for into distinct groups based on agile experience management specific characteristics, platforms and deadline-driven ensuring accurate results. This research. method separates individuals from each group, combining Limitations of probability sampling them into a single sample. Cluster sampling  Stratified and cluster sampling offer advantages but may not  Cluster sampling involves accurately represent dividing a population into population differences, clusters with similar overlapping characteristics, or characteristics, randomly target the intended audience. selecting entire clusters to save costs. It's commonly used for When to use probability sampling large or geographically dispersed populations, but  Probability Sampling in faces higher sampling error Quantitative Studies risk.  Ideal for quantitative studies requiring statistical analysis.   Used when full population  Non-probability sampling is a surveying is difficult or different approach, involving expensive. no equal chance of selection  Used in market research for from the overall population. insights into large  Some members may have zero populations. chance of being selected.  Involves understanding  Non-Probability Sampling in consumer usage, influencing Research purchasing decisions, and  Often used for exploratory and emerging industry qualitative research focusing categories. on specific expertise, experiences, or insights.  Probability Sampling in  Examples include snowball Business Improvement sampling, where researchers  Allows companies to refine connect with relevant people ideas and enhance business without a full list of by analyzing data from their participants. entire target market.  Risks and Benefits  Example: A coffee chain  Non-probability sampling is expanding customer loyalty easier and cheaper but has a program with additional higher risk of sampling bias due payment options and to subjective judgment of the rewards. researcher.  Assesses customer response  The sample size and end before making significant results don't necessarily changes. represent the entire population. Understanding Probability and Non- Probability Sampling  Probability Sampling Types  Includes simple random sampling, stratified sampling, cluster sampling, and systematic sampling. Lesson 4 aim to describe naturally Formulating the hypothesis occurring behaviors. WHAT IS A HYPOTHESIS? Examples of nonexperimental hypotheses Hypothesis Overview Represents end of research  Type of Design idea process. Main idea of experiment. Correlational – (related) Statement about predicted relationship between variables. Ex post facto – (report) Existing State Not a speculation, guess, or Non-equivalent groups - Participants hunch. aren’t randomly assigned NONEXPERIMENTAL and Longitudinal– (how) Studying the experimental HYPOTHESIS same participant Nonexperimental Hypothesis Cross-sectional– (will show) Different Overview set of participants THE CHARACTERISTICS OF AN Predicts relationships between events, traits, or behaviors. EXPERIMENTAL HYPOTHESIS Not a cause-and-effect Experimental Hypothesis statement. In true experiments, Every experiment has at least hypothesis predicts effects of one hypothesis. antecedent conditions on measured behavior. Complicated experimental designs that compare several NONEXPERIMENTAL designs treatments at the same time Do not restrict responses. may test various hypotheses Do not typically include hypothesis. simultaneously. Phenomenology, case studies, Experimental hypothesis is a naturalistic observation, tentative explanation of an qualitative studies, and surveys event or behavior, describing the effects of specific  conditions on a measured analytic statement due to its behavior, after extensive vagueness thought and discarding Hypotheses should be concise improbable explanations. enough to be proven wrong. Contradictory Statements Synthetic Statements contradictory statements, statements with elements that Synthetic statements can be oppose each other—because true or false. contradictory statements are Experimental hypotheses must always false be synthetic. example - “I have a brother Experiment tests yield and I do not have a brother” information for decision- Because analytic statements making. are always true and contradictory statements are Example: "Hungry students always false, we do not need to read slowly. conduct experiments to test Non- Synthetic Statements them: We already know what the outcome will be.  Nonsynthetic statements should Synthetic Statements be avoided at all costs. These fall Hypothesis is a synthetic into two categories: statement if it can be analytic or expressed in the "If … then" Contradictory. form. This form expresses potential Analytic relationships between antecedents and measured An analytic statement is one behaviors. that is always true; for Example: "If you look at an example, “I am pregnant or I appealing photograph, then am not pregnant.” your pupils will dilate." Analytic statements can be The statement can be true or generated by inadequate false. prediction statements. For instance, "The weight of dieters will fluctuate" is an Testable Statements commonly employed in science and mathematics to construct Experimental hypotheses must an overall explanatory scheme. be testable, allowing manipulation and measurement of behavior. Many interesting hypotheses are currently unscientific due to their inability to meet this criterion. Falsifiable Statements Researches hypotheses must be falsifiable, meaning failures to find the predicted effect are considered evidence that the hypothesis is false. Parsimonious Statements Simple explanations are preferred over complex interpretations, promoting equal adequacy. Fruitful Statements Ideally, a hypothesis is also fruitful; that is, it leads to new studies. The Deductive Model It is often difficult to know in advance which hypotheses will The deductive model, a reverse of the inductive model, be the most fruitful. involves reasoning from THE INDUCTIVE MODEL general principles to make predictions about specific The inductive model, a method instances, particularly useful of reasoning from specific when a well-developed theory cases to general principles, is has clearly stated premises. What is the Equity Theory? Equity theory is a motivation theory that says that employee motivation is mostly determined by their sense of fairness at work.  Walster, Walster, and Berscheid's Theory of Equity in Interpersonal Situations Individuals aim to optimize outcomes (rewards minus costs). Perceived inequity leads to distress in relationships. More distress, the harder it is to restore equity.