Psychology 1A Experimental Methodology Component (a.k.a.

. Psychology and the Scientific Method) Lecture 1 Why do I need to learn this stuff? So that you can apply the principles of the scientific method to evaluate and think critically about psychological science. So that you become better consumers of science.
At the end of this course, you should be able to: 1) Appreciate diverse but complementary theoretical and experimental approaches to major psychological issues. The following areas will be explored: cognitive, social and personality development of the individual; how human beings interact and influence each other in social situations; individual differences and their measurement; normal and altered states of consciousness; and the impact of stress on health. 2) Better understand the application of psychological theories and principles to everyday life. The practical implications of key theoretical perspectives will be emphasised to give you an insight into the value and usefulness of psychology to society. 3) Recognise the role of the scientific method in psychology. Psychology 1A presents not only the outcome of research, but an appreciation of research methods in psychology and some insight into how the research process works. 4) Plan targeted information searches. Information literacy will be emphasised, such that you will learn to recognise the need for information, and be able to locate, evaluate and apply relevant information effectively. 5) Develop communication skills within a scientific context . A key element of the Psychology 1A course is writing components of a research report, in the context of which you will engage critically with information: sourcing relevant literature, integrating theoretical and empirical information , synthesising logical arguments and drawing conclusions. 6) Think more critically about the information presented throughout this course. This includes an ability to identify assumptions, critically evaluate research methodology and results, assess conclusions, as well as to compare and contrast competing theoretical perspectives. Problem solving and reflective learning skills will be emphasised.

Letter K example demonstrates Availability Heuristic: A rule of thumb for estimating probabilities based on the ease with which instances or occurrences can be brought to mind. SHOW ME THE EVIDENCE! How do we get the evidence? Three main goals of psychological research: To measure and describe To explain and predict To apply and control The Scientific Method: Identify a research area/question Propose a theory Derive a hypothesis to test Define groups to test Define variables to measure Definitions

Theory: Principle proposed to explain how a number of separate factors are related. OR: a systematic way of organising and explaining observations, which includes a set of propositions, or statements, about the relationships among various phenomena. Hypothesis: Prediction about relationship between two or more variables. OR: a tentative belief about the relationship between two or more variables. It predicts the findings that should be observed if the theory is correct. Variable: any phenomenon that can differ, or vary, from one situation to another or from one person to another; in other words, a variable is a characteristic that can take on different values. Continuous variable: a variable that can be placed on a continuum (e.g. degree of optimism, intelligence, shyness). Categorical variable: a variable comprised of groupings or categories (e.g. gender, species). Independent variable: the variable(s) we manipulate. Dependent variable: the variable we measure; because it can vary depending on what happens in the experiment. Confounding variable: can there be some other factor or influence which affects the data you collect from the experiment. The Research Process Conduct the research (variety of methods) Write up the results Communicate the results (conferences, scientific journals, press/media) Why should I participate in research? To experience research and researchers first hand To apply knowledge gained from lectures To see theories, methods and measures from the text book in action Lecture 2 The scientific approach and experiments in psychology Characteristics of scientific approach: The production of public knowledge The use of systematic methods Simplicity (parsimony) Precision
Confidence in theory is increased Predictions confirmed

Theory about some aspect of behaviour

Predictions derived from theory (hypotheses)

Research Predictions not confirmed

Confidence in theory is reduced Theory is modified and new predictions tested Theory is rejected

Theories: Definition: an explanation using an integrated set of principles that organizes and predicts observations. Dimensions of good scientific theory: Power: correctly explains a wide variety of phenomena Testability: makes clear predictions that could be proven wrong Simplicity: has few theoretical concepts and relationships Fertility: suggests brand new ideas to explore

Scientific Hypotheses: Hypotheses are statements about the relations between variables. They are testable predictions often implied by a theory. They carry clear implications for testing the stated relationship. A variable is anything which varies anything which takes on different values. Variables can be measured and/or manipulated. Falsifiability: It is not a natural way of thinking. People who are evaluating hypotheses tend to confirm rather than to disconfirm the hypotheses. Falsifiability (or refutability or testability) is the logical possibility that an assertion can be shown false by an observation or a physical experiment. That something is "falsifiable" does not mean it is false; rather, it means that it is capable of being criticized by observational reports.

Operational definitions: Definitions must be precise A detailed description of the procedures or operations used to measure or manipulate the variables Why? o Eliminate confusion o Place concepts in the public realm o Allow hypotheses to be tested and/or replicated

Lecture 3 Experimentation in Psychology 1. Experimentation is used in psychology to infer cause-effect relationships. 2. A control condition is needed in an experiment in order to properly deduce or draw conclusions from the results of an experiment. 3. Confounding variables, random error and order effects must be attempted to be controlled for in an experiment. 4. Experiments can use one-way or two-way designs. 5. E.g. of descriptive approach: survey. Problems with this approach include social desirability effects, and no causal inference can be made, since no variables are being manipulated. Causality 1. The cause has to precede the effect in time. 2. The cause must be related to the effect. 3. Other explanations of the cause-effect relation must be eliminated. Experimentation Isolate cause and effect Manipulate factor of interest (independent variable; only factor allowed to vary!) Measure the effect of the manipulated factor on the dependent variable. All other factors must be held constant in order to firmly deduce a cause-effect relationship. Experiment and control conditions Without a control condition, you cant be sure that the outcome you observed would not have occurred anyway, even without the experimental manipulation. Experiments require both an experimental condition, and a comparison, or control condition. Selecting participants for experiments: Random assignment

Random allocation prevents individual differences having systematic effects on experimental results.

Confounding Confounding occurs when an experiment contains a variable that systematically varies with the independent variable. I.e. independent variable and irrelevant variable change together. Is it the independent or irrelevant variable causing the effect?

Random error Random allocations of participants to different conditions can prevent confounding BUT the irrelevant variable may still exert some influence on the dependent variable. This influence should be distributed evenly between the experimental and control conditions. Thus effects are RANDOM and not systematic, and do not undermine the logic of the experiment. Ways to control irrelevant variables Random allocation of participants (sample size must be large enough) Repeated-measures design using the same group of participants in different conditions. Using matched-samples Order effects Any change in a participants performance that can be attributed to the fact that the participant experienced more than one level of the independent variable. Controlled for by counter-balancing, i.e. some participants take part in experiment in one order, some in the other. Matched samples (very tedious)

One-way designs Only one independent variable Simplest type compares only 2 groups (in 2 different conditions) Forms the building blocks for more complex designs Two-way designs Two independent variables Two levels of each independent variable Lecture 4 Descriptive approaches Description is the starting point for science. Naturalistic observation: no independent variable is manipulated; the researcher observes naturally occurring behaviour, but does not experiment with it. o Advantages: is necessary in studies which are difficult to experiment on and manipulate variables; useful in initial phases of research.

Disadvantages: impossible to isolate cause; cause-effect cannot be determined; time consuming; observed aware of observer? Participant observation: observation and recording of behaviour of others while actively participating with them in the situation. o Advantages: it can be used in situations that otherwise might be closed to scientific investigation. o Disadvantages: observer could lose objectivity; personal biases could interfere with description. Longitudinal studies: Studies that assess changes that occur across a period of time (age changes). o Advantages: major points of change are observed; genuine changes and stability is observed. o Disadvantages: time consuming; expensive; participant attrition (getting participants to come back at a later date to be assessed again) Cross-sectional studies: Studies that assess differences across different age groups. o Advantages: inexpensive and less time consuming; low attrition rate (participants not required to come back at later dates) o Disadvantages: cannot observe changes in individuals. Age-cohort effect o Cross-sectional studies investigate different cohorts with different experiences. o Cohort differences confounded with age differences. Longitudinal studies do not suffer from this. Case studies: an intensive description and analysis of a single individual, organisation, or event based on information obtained from a variety of resources such as interviews, documents, archival records, etc (e.g. feral kid Oxana from tutorial). o Advantages: excellent source of ideas and hypotheses; important in describing rare events that sometimes can provide empirical evidence in support of one theory over another. o Disadvantages: the cause of an event cannot be determined; difficulty in generalising to other individuals and situations. Survey techniques: o Limited data from large samples (opposite of case studies) o Usually questionnaire based o Random samples are taken (population representative and non-representative samples) o Address questions of how many, how much, who, why. o Advantages: quick and efficient; very large samples; obtain public opinion almost immediately. o Dangers of surveys: Haphazard samples volunteer bias, availability bias (need random samples) Question wording effects how you ask might affect the answer you receive. o

Lecture 5 Correlational approaches Correlational studies are descriptive (are a subset of them) and they examine the extent to which two or more variables are related, and if so how strongly. o Variables are NOT manipulated in correlational studies, instead variables of interest are selected and the possible relationships between them are examined. Hence, causation or the cause-effect relationship between variables CANNOT be determined, it can only be suggested. Correlational studies are important in the early stages of research and by examining strong relations between variables, predictions can be made, but the cause cant be determined. The scatter plot is used as a graphical representation of the data obtained in a correlational study. o The independent variable is placed on the x axis o The direction of the slope determines whether it has a +/- relationship o The degree of scatter determines the strength of the correlation

I.e. the density/tightness of data (amount of scatter in one area) Examples: Positive relationship IQ vs. school grades Negative relationship socioeconomic status vs. high school drop out No relationship eye colour vs. attitude o Correlation coefficient (measured by R = x ; where -1 <= x <= 1): a statistical measure of the extent to which 2 variables vary together. Strength = number Sign = direction Predictions can be made once we know the association between variables o Predictor variable criterion variable o Correlation indicates possibility of cause-effect relation BUT does NOT prove causation! Why not? Directionality problems o A might cause B or B might cause A Third variable o C could cause either A or B to affect each other o

Lecture 6 Critical consumers: Reliability and validity in research Ideal measurement scenario: measured score = true score o In reality: measured score + measurement error = true score Reliability is essential for good measurement o More measurements, more participants, etc. Priming effect: o When participants are primed to respond to something a lot quicker due to pre-exposure. Inter-observer or inter-rater reliability: o The degree to which different observers agree upon an observation or judgment Internal (split-half) reliability: o The degree to which items aim to assess or measure the same thing; relates to the internal design of the experiment and what it aims to investigate. o Look at correlation between 2 halves Test-retest reliability: o The extent to which scores on a test remain stable over time. Validity: correctness/truth of an inference. Internal validity (in design): o IV and DV are causally related o All rival hypotheses must be eliminated External validity (applicability): o Validity of the inference about whether the causal relationship holds over people, settings, time, etc. Construct validity: o Validity of inferences we make about higher order constructs from the operation we use to represent them. Depth of processing How to assess? Reflected in the quality of operational definitions o Content validity: Degree to which the items or tasks adequately sample the target domain. o Criterion validity:

Extent to which the procedures or behavioural measures can be used to infer or predict some criterion. Link between reliability and validity: o Note: Measures can be reliable but not valid But valid measures must be reliable

Lecture 7 Ethical dilemmas in psychological research The Hawthorne effect o Being aware that someone is observing or measuring your behaviour, which may change the way you behave (measure is then reactive; typically change in behaviour is an improvement) o I.e. our tendency to change our behaviour when being observed Ethical dilemma o Conflict between knowledge gained from study vs. potential harm to participant o Do the potential benefits of the study exceed the potential harm? o All research has to get ethical approval Key tenets of codes o 1. Protection from harm Physical, mental, or long-lasting o 2. Informed consent Legal requirement Participants must be given an accurate account of potential risks Participants must be allowed to withdraw o 3. Protection of Privacy Confidentiality of data unless otherwise agreed upon o 4. Valid research design Potential to yield results Experiment which violated all 4 tenets: Tuskegee Syphilis experiment o No consent o No information o Documented natural progression of syphilis o Observation, no treatment Experiment which violated privacy: Urinal study o Concealed observation (periscope) Experiment which deceived and harmed participants o Milgrams obedience study misinformation intentionally given (to teacher) Is it ethical to experiment on animals? o Ulrich: We cannot defend our scientific work with animals on the basis of the similarities between them and ourselves, and then defend it morally on the basis of differences Scientific misconduct e.g. Burt Affair o Plagiarism o Fabricating data (fudging) o Selective exclusion of data o Incomplete/inaccurate reporting of procedures o Partial or duplicate publication