165

Psychology (Fach) / Research Methods (Lektion)

Diese Lektion wurde von Julia2308 erstellt.

Lektion lernen

zurück  |  weiter 1 / 4

• Scientific Method - an approach to asking questions - a method of obtaining answers - empirical approach > direct observations > systemtic control through experimentation - sceptical attitude - the strongest evidence is converging evidence from multiple studies > empirical reasoning
• Research Hypotheses - a tentative explanation for a phenomenon - often stated as a conditional statement (if theory X is correct, we expect behavior Y to be..)  
• Research Process - Develope a research question - Generate a research hypothesis - Form operational defintion  - Choose a research design - Evaluate the ethical issues - Collect and analyze data; form conclusions - Report reseach results  
• Null Hypothesis Significance Testing - Goal: determine whether mean differences among groups  in an experiment are greater than differences  expected simply because of chance (error variation) - Ho: assume groups do not differ; assume IV did not have an effect
• Null Hypothesis Significance Testing Probability theory - estimate likelihood of observed outcome, while assuming null hypothesis is true > ''statistically significant'': outcome has small likelihood of occuring under Ho > reject Ho
• Null Hypothesis Significance Testing Alpha/Level of Significance – outcome at p ≈ .05 has about a 50/50 chance of being repeated (at p < .05) in an exact replication – as probability of outcome decreases (e.g., p = .025, p = .01), likelihood of observing a statistically significant outcome (p < .05) in an exact replication increases – APA recommends reporting exact probability of outcome
• Null Hypothesis Significance Testing -Ho: accept? - NEVER accept Ho > do not conclude that the IV did not produce an effect - cannot make a conclusion about the effect of an IV > some factor in experiment may have prevented us from  observing an effect of the IV > most common factor: too few participants
• Null Hypothesis Significance Testing Criticism - Ho can almost always be rejected, if sample size is large enough - statistical significance ≠ practical significance > practical significance might not appear statistically significant if the test lacks power - publication bias and the file drawer problem
• Publication Bias - a bias with regard to what is being published, among what is available to be published - bias towards reporting significant results > 3 times more likely to be published when results are significant > most common reason for non-publication is a researchers declining to submit results for publication   
• File drawer problem - publication bias: bias towards publishing research with significant results oder research supporting the null hypothesis > especially file drawer problem: when null findings < sig. results publishing  
• Anzeige
• Null Hypothesis Significance Testing Experimental Sensitivity - likelihood an experiment will detect the effect of an IV when in fact, the IV has an effect > affected by experimental methods and procedures > sensitivity increases with good research design and methods > high degree of experimental control > little opportunity for bias  
• Null Hypothesis Significance Testing Power - Likelyhood that researchers will reject Ho, when Ho is false > low statistical power increases Type II errors > Power = 1 - β(tte) -> power increase, type II error decrease - three factors affect power of statistical tests > level of significance (alpha) - increase power by using a larger alpha (less conservative test) > size of the effect of the IV > sample size (N)- increasing n is the easiest way to boost the power of a test - 80% power - can be used to assess how much n to have in your experiment
• Type I Error - alpha α - incorrect rejection of a true null hypothesis - false positive  
• Type II Error - Β - failure to reject a false null hypothesis - false negative  > worse usually (medical setting: test did not recognize an illness)  
• Prospective Power Analysis   - estimate effect size of IV > examine previous research involving the IV - refer to “Power Tables” > identify sample size needed to observe effect of IV - use adequate sample size > most studies in psychology are “underpowered” because of  low sample size -  a priori power analysis > universally accepted as useful    
• Restrospective Power Analysis   - determine the power of a study based on the effect  size, sample size, and significance level - post-hoc power analysis -> controversial: Power approach Paradox    
• Power Approach Paradox - in the post hoc power analysis setting: > a study with a null results is thought to show MORE evidence that the null hypothesis is actually true, wehn the p-value is smaller, since the power to detect an actual effect would be higher  
• Verifiability principle - positivists’ claim that “truth” is revealed and conclusively warranted merely by the amassing of factual observations - disputed by Popper
• Falsifiability Principle (Popper!) - Popper: scientific knowledge evolves by a cyclical process involving the testing and retesting of plausible suppositions that are stated as falsifiable hypothesis
• Why questions How questions (1) - call for explanations that take the form of causal inferences or purposive inferences about the reasons (i.e., motives or intentions) for the particular behavior (2) - “what is happening” questions (i.e., how things work), or what we call descriptive research orientation
• Social constructionism - panpsychic - ''reality'' is a construction of the sum total of human mind - sciences (biology, physics) is an exception: replicable, empirical > the natural world is an exception as it exists independent of the human experience  - antirealists vs./+ metaphysical realists _________________________________________________________ - major difference between SC and C/P - BE > SC dismiss the use of scientific methods (experiment) and the hypothetico deductive model of science as irrelevant to studying purposive behaviour - model of choice: narrative analysis based on the interpretive model: ask people to explain personal experience and feelings in narrative terms - challenge the assumption that there is a “real” social world out there that some psychological theories reflect or match more accurately than others - knowledge and truth -> socially constructed realities - truth is anything that stands up to experience, enables us to make accurate predictions (however vague), or brings about or prevents certain outcomes
• Contextualism/Perspectivism - typically dismisses metaphysical realism as a pointless debate - except that knowlegde is limited - emphazies the context in which an action, utterance, or expression occurs and argues that in some important respect, the action utterance or expression can only be understood relaitve to the context - empirical research is necessary, but not just for deciding whether a theory is true or false (as all theories are cryptic) __________________________________________________________ - embraces the idea that more than one model or conception is needed to comprehend human nature - knowledge does not exist in a vacuum but is situated in a sociohistorical and cultural context of meanings and relationships - purpose of research: “which are the crucial perspectives”, not whether one hypothesis or theory is true or false, as “all hypothesis and theories are true, as are false, depending on the perspective from which they are viewed'' - two perspectives on cognition                                                                       > adding model & averaging model - doctrine: methodological pluralism and theoretical ecumenism: researchers need to employ different operations (multiple operationalism) in order to converge (triangulate) on the phenomena of interest to them (no universal laws without boarders) - ciriticism: idea of change is viewed as intrinsic > close to nihilistic perspective on the meaning of life (without objective meaning, purpose, or instrinsic value) - contextualists dismiss the idea that there is a reality that exists independent of our observations - Instead of exhausting ourselves by arguing for or against “real truth”, contextualists see the debate aside as pointless and settle for a less exalted idea of “truth” as dependable explanation ________________________________________________________ - 
• Evolutionary epistemology (organic evolution) - inspired by Darwins Theory of evolution through natural selection and adaption  - analogy between biological progression and adaption and the advancement scientific knowledge > successful theories and knowledge in science evolve in a competition for survival, similar to the way that evolution and adaption drive the transformation of organisms - Ultimately there is a priory reality waiting to be discovered and described by a process of elimination > epistemologists (like contextualists) believe that scientific progress is a matter making the  'best guesses' about reality, to explain a phenomenon  
• Three points of view: What philosophical researchers know- - Social constructionism - Contextualism/Perspectivism - Biological epistemology
• Peirce's four ways of knowing - Peirce established the doctrine of pragmatism – an emphasis on practical consequences and values as the standards used to evaluate the caliber of truth - four traditional foundations of knowledge: > method of tenacity > method of authority > priory method > scientific method
• Peirce's four ways of knowing Method of Tenacity - Verlaesslichkeit, Altbacken '' immer schon da gewesen '' - most primitive orientation - people cling to certain assumptions and beliefs simply because they have been around for so long (earth is fixed) > false consensus phenomenon: people overestimate the extent to which other hold the same beliefs as them)
• Peirce's four ways of knowing Method of authority - somewhat better measure of tenacity, but limited - relying on experts/authorities for the truth  
• Peirce's four ways of knowing Priory Method - individual powers of reason and logic to think for ourselves in a way that seems ostensibly unimpeded (angeblich unbehindert) - i.e math exercises - subject to one's own biases and tastes
• Peirce's four ways of knowing Scientific Method - people need a fourth way of knowing, one by which they can base their beliefs on something reliable and tangible (konkret) - solves dilemmas of disagreement - this method surpasses the other three, because it allows the combination of logic and reasoning and objective inverstigation of the world  
• Four supports of conviction: (extraempirical factors) - Rhetoric, Perceptability and Aesthetics  - Rhetoric: > “Rhetoric” is defined as persuasive language to prose, warrant, defend, or excuse certain beliefs. > Whorfian hypothesis: the language that we use is our window into the world. - Perceptability and Vizualization > The use of images in everyday reasoning and thinking in science and in the social world (the use of analogies etc.) > Linguistic regularities, beliefs, and communication strategies that cannot easily be explained in formal terms can often be explained (and sustained) in a conceptual framework of perceptible representations - Aesthetics > The perceptible images are evaluated on their beauty; probably a basic psychological component (theories in science can also be “beautiful” or “elegant”)  LIMITATIONS: > The factors of perceptibility, rhetoric, aesthetics, and empirical content are imperfect support. Make use of empirical methods to create comparable findings.
• Anzeige
• Neuroscience - most micro (nervous system) - biological and biochemical factors in behavioural science
• Cognition - more micro - thinking and reasoning in bahavioural science  
• Social Psychology - more macro - interpersonal and group factors in behavioural science  
• Sociology - most macro - societal systems in behavioural science  
• Broad research orientations: Descriptive, Relational and Experimental research orientations - Descriptive research orentations:   > answers how things are > frequently considered as a necessary first step > correlation (not causal!) between predictor and criterion - Relational reseach orientations:                                                                        > focus on atleast to variables to infer causality > Varied replications – replications that vary slightly from one another > Construct – refers to an abstract idea that is used as an explanatory concept > Operationalize – defined in an empirical way > Pseudosubject – confederate of the experimenter > working hypothesis/ ad hoc hypothesis - Experimental research orientations:                                                                 > focuses on the identification of causes (What?) > “experimental” and “control” groups to get infer causality
• Empirical princples as probabilisitc assertions - Empirical principles – empirically-based generalizations; a) Empirical – because they are based on controlled empirical investigations b) Principles – because we do not think of them as universal laws, but rather as generally accepted scientific truths about how behavior is likely to manifest itself in the situation specified - Deductive reasoning: reasoning from general to specific in a probabilistic way > Hempel model/ covering law model – scientific principles can be cast as deductive arguments that contain at least one premise that is universally true - Inductive reasoning: reasoning from specific to general in a probabilistic way  
• Empirical princples as probabilisitc assertions - Empirical principles – empirically-based generalizations; a) Empirical – because they are based on controlled empirical investigations b) Principles – because we do not think of them as universal laws, but rather as generally accepted scientific truths about how behavior is likely to manifest itself in the situation specified   - Deductive reasoning: reasoning from general to specific in a probabilistic way > Hempel model/ covering law model – scientific principles can be cast as deductive arguments that contain at least one premise that is universally true - Inductive reasoning: reasoning from specific to general in a probabilistic way    
• Hypotheses - theoretical statements or theoretical propositions - used interchangeable with''theory'' - historicall presumed to be testable, i.e falsifiable  
• Theories - hypothetical formulations, conjectural - used interchangeably with hypotheses - far-reaching theories: macro-level - focus on a specific situation/ activity: micro-level - presumed to be testable/falsifiable  
• Operational defintion - the technical name for an empirically based definition - defined on empirical by certain specifiable observational grounds procedures  - Campbell’s and Fiske’s strategy of multiple operationalism: operational defintions generally “underdetermine” (i.e., only partly define) perfectly valid concepts, and therefore it is necessary to measure a given concept by many different operations vs. Typology 
• Typology > alternative of trying to condense the definition of a psychological concept into a single sentence (or into a single measuring operation) is to formulate a typology (i.e., a systematic classification of types), to boil down a complex concept - a systematic classification of types when trying to condense the definition of a psychological concept > method: facet analysis (factor analysis)  
• Typology > alternative of trying to condense the definition of a psychological concept into a single sentence (or into a single measuring operation) is to formulate a typology (i.e., a systematic classification of types), to boil down a complex concept - a systematic classification of types when trying to condense the definition of a psychological concept > method: facet analysis (factor analysis)  
• Occam's razor > to cut away what is superflous > a researcher's parsimonious (but coherent) hypotheses building  
• Positivism -  statements authenticated by sensory experience are more likely to be true  - “bucket theory of the mind” because it assumes that knowledge proceeds from observations to generalizable knowledge - daws on a verification principle - is now said to be “false''   - (Critism) Hume: “All knowledge resolves itself in probability” and thus it is impossible to prove beyond doubt that a generalization is incontrovertible true''
• Falsificationism - according to the “fallibilists”, the proposition in question must be stated in such a way that it can, if incorrect can be rejected by some finite set of observations - it is argued that theories might have boundaries that will never be explored - falsificationism perspective does not simply mean that because certain predictions fail a test, the theory is automatically wrong; - but if one theory is more falsifiable than another theory, and if that theory has survived more empirical testing than the other one, it is presumed to be a better theory  
• Conventionalism - hybrid: Positivism + Falsificationism - plays on the role of language (also known as Duhem-Quine thesis) > meaning that theories evolve on the basis of certain linguistic conventions (like simplicity), not merely on the basis of their ability to withstand empirical disconfirmation > What happens when a “refutation” occurs (?):  It is usually taken as a signal that the theory needs tinkering or adjustment, not that it must be discarded > Sometimes, however, the modifications are so fundamental that it is difficult to recognize the old theory, or impossible to use the new theory to explain phenomena accounted for by the old theory
• Conventionalism - hybrid: Positivism + Falsificationism - plays on the role of language (also known as Duhem-Quine thesis) > meaning that theories evolve on the basis of certain linguistic conventions (like simplicity), not merely on the basis of their ability to withstand empirical disconfirmation   > What happens when a “refutation” occurs (?):  It is usually taken as a signal that the theory needs tinkering or adjustment, not that it must be discarded > Sometimes, however, the modifications are so fundamental that it is difficult to recognize the old theory, or impossible to use the new theory to explain phenomena accounted for by the old theory  
• Amalgation of Ideas ( Hypothesis Building guidlines: ) - requirements for scientific theories and hypotheses: mixture of falsificationism, conventionalism, and practicality > finite testability: must be stated in a way, that if it is false, aspects can be disconfirmed by a finite set of observations  > falsibiable > theories can add to or replace outmoded models > if its not supported it may not be right/ if it is supported it may not be right > there are always (better?) alternatives  
• Statistical significance - wenn the p value of the test is below α (usually 0.05). - Significance test = Size of effect x Size of study > the larger the study in terms of the total number of observations (N) or sampling units, or the larger the effect size the larger the value of the significance test (e.g., t, F, X²) and therefore the smaller (and usually more coveted the p-value
• Effect size - Two families of effect size: > Correlational family: (phi, rpb, R², partial-eta squared…) > Difference family: (Cohen’s d, Hedge’s g…) - Interval estimates around effect sizes > null-counternull interval: based on the actual p value rather than on previously specified alpha > the ''null” anchoring one end of the interval is the effect size that is associated with the null hypothesis (and is typically zero) > counternull refers to the non-null magnitude of the effect size that is larger than the obtained effect size and is supported by the same amount of evidence that is supported by the null value of the effect size > this interval (null value to counternull value) can alert a researcher to whether a conclusion of “no effect” might be an error (type 1 error).     

zurück  |  weiter 1 / 4