122
Introduction to Research Methods (Fach) / RM (Lektion)
Anzeige
- The four canons of science - Determinism - Empiricism - Parsimony - Testability
- Determinism events follow certain rules -> orderly and predictable events have causes; can be discovered by scientific method human behaviour is not predetermined, is determined
- Empiricism make systematic observations don't trust - test!
- Parsimony pick simplier theory, if they both equally good Occam's razor Meyer's law
- Occam's razor it is inappropriate to make more assumptions than you absolutely have to
- Meyer's law it is a simple task to make things complex, but a complex task to make them simple
- Testability scientific theories should be testable (confirmable or disconfirmable) using currently available research techniques operationalism
- Operationalism the process of defining a 'fuzzy' concept so as to make the theoretical concept clearly distinguishable or measurable and to understand it in terms of empirical observations
- Falsifiability criterion it is possible to prove that a theory is false, but it is not possible to prove that it is true Karl Popper
- Hypotheses Specific predicitions about specific events that are derived from one or more theories help determine when/if specific theories are true
- Anzeige
- Theories Theories are general statements about the relations between two or more variables. good theory: deterministic, empirically grounded, parsimonious, testable have boundary conditions (laws don't)
- Laws Universal statements of the nature of things that allow reliable predictions of future events.
- Validation (hypothesis testing) Attempting to gather evidence that supports or confirms a theory or hypothesis
- Falsification (hypthesis testing) Attempting to gather evidence that invalidates or disconfirms a theory or hypothesis
- Qualification (hypothesis testing) Attempting to identify the boundary conditions under which a theory is and is not true
- Deductive Reasoning from general to specific The process of reasoning from one or more general statements (premises) to reach a logically certain conclusion. conclusion supposedly certain Theory -> Hypothesis -> Observation -> Confirmation
- Inductive Reasoning from specific to general Reasoning in which the premises seek to supply strong evidence for (not absolute proof of) the truth of the conclusion conclusion merely probable Obervation -> Pattern -> Tentative hypothesis -> Theory
- Valid is... actually supporting the intended point or claim; acceptable as cogent cogent: clear, logical and convincing
- Validity relative accuracy or correctness of a statement Internal External Construct Conceptual
- Internal Validity The extent to which a set of research findings provides compelling information about causality. High -> Variations in independent variable caused changes in dependent variable. Lab experiments are high in internal validity. Requirements: Covariation (changes in one variable must correspond with changes in the other), Temporal Sequence (changes in the first variable mus precede the changes in the second), Elimination Confounds (Third-variable problem) Am I correct in concluding that A causes B?
- External Validity The extent to which a set of research findings provides an accurate description of what typically happens in the real world. High -> Findings of the study will apply to other people, other physical environments, and perhaps other cultures. Generalizability with respect to people and situations. Am I correct when I generalize my laboratory findings to the real world?
- Construct Validity The extent to which the independent and dependent variables in a study truly represent the abstract, hypothetical variables of interest to the researcher. High -> Observable, study variables represent abstract concepts well. Quality of operational definitions Am I correct saying that my manipulation makes children aggressive rather than scared?
- Conceptual Validity How well a specific research hypothesis maps onto the broader theory that it was designed to test. High -> Research hypotheses or research programs map well onto theory. Broader than construct validity. Does the theory under investigation really lead to the suggested hypothesis? Am I correct that my studies falsified one theory rather than another?
- Independent Variable Manipulated by the experimenter; expected to influenve the dependent variable
- Dependent Variable Measured by the experimenter; expected to be influenced by the independent variable
- Reliability refers to the consistency or repeatability of a measure or observation
- Test-retest reliability The degree to which a test continues to rank order participants' scores in a stable manner over time (-> correlating responses on a measure at time 1 with scores on the same measure at time 2)
- Internal consistency The degree to which the total set of items or observations in a multiple-item measure behave in the same way. Statistical indicator: Cronbach's Alpha
- Interrater reliability The degree to which different judges independently agree upon a measurment, judgemant, or observation.
- Reliability and Validity Maximizing both validity and reliability is crucial to all forms of psychological research However: It is impossible to maximize all types of validity and reliability simultaneously Reliablity is a necessary but not sufficient condition for validity Realiability and validity are affected by the context and purpose of the measurement
- Anzeige
- Measurement scales Nominal (nomen - name) Ordinal (order) Interval (spaced at regular intervals) Ratio (meaningful ratios of values)
- Nominal (Categorical) Scales Objects or individuals are broken into categories that have no numerical properties Property: Identity Mathematical operation: none Examples: Gender; Religion; Schizophrenic, autistic, depressed
- Ordinal Scales Objects or individuals are cateogorized and the categories form a rank order along a continuum. Property: Magnitude Mathematical operation: Rank order Examples: ranking in a contest; educational background
- Interval Scales Units of measurement (intervals) between the numbers on the scale are all equal in size. Property: Equal unti size Mathematical operation: add, subtract Examples: Temperature; self-esteem scores; most psychological scales
- Ratio Scales In addition to order and equal untis of measurements, there is an absolute zero that indicates an absence of the variable being measured. Property: absolute zero Mathematical operation: add, subtract, multiply Expamles: Height; Weight; Age
- Self report measures easy and popular give people a list of statements and ask them to rate their agreement based on literature (scales) quick data collection often automatic data processing tools to check the underlying scales
- Observation Behaviour versus psychological construct best predictor of behaviour is the behaviour in the past less abstract/closer to real life greater validity high transparency Validity: Are people influenced by the fact that they know they are observerd? Yes, to a certain extent, but situation dependent, trust, get used to it easily, face validity
- Naturalistic observation Spontaneous behaviour, in natural 'habitat', 'unobtrusive'
- Participating observation Observer 'plays along'
- Lab observation Elicitation of specific behaviours in controlled environment
- Self observation Client/subject observes himself
- Selectivity You observe only what you focus on The rest gets lost
- Subjectivity Interpretation (does everybody see the same?) Is the observation sufficiently objective? - Systematic observation -> coding scheme -Training observers Reliability: - Overlapt between indepedent observers - Cohen's kappa
- Cohen's kappa Very global measure of reliability (% observed overlap - chance overlap) : (1 - chance overlap) 0.41 - 0.60 moderate agreement 0.61 - 0.80 substantial agreement 0.81 - 1.00 almost perfect agreement
- Codingscheme state (continue), e.g. duration of certain states (e.g. Sleeping waking, on-task/off-task) scoring of events (discrete), e.g. short lived behaviours/actions (e.g. hitting someone, giving the right answer) Measurement scales: nominal to ratio for each task question (protocol), is a segment (unit of analysis) for each segment, code level of reasoning or each child for each segment, code the degree of collaboration
- Video-observation "Paper & pencil": coding sheet Software
- Application of observation 1. Clinical context (diagnostics) analyzing problems in a context 2. Research suitable for studying change processes (learning effects) interactions between variables (what follows what) hypothesis generation but also testing
- ABC-scheme Antecedent - Target behaviour - Consequence
- Question effects Question order and response order Open vs. closed questions Assessment of no opinion Measuring a middle position Agreeing-response bias Tone of wording
- Psychological measurement from notion to numbers - measurement scales The challenges: - Judgement phase: how does the participant interpret the question - Response translation phase: making sure that we can make sense of the answers