122

Introduction to Research Methods (Fach) / RM (Lektion)

Diese Lektion wurde von HuberthErbse erstellt.

Lektion lernen

zurück  |  weiter 2 / 3

• Judgement Phase Open-ended questions - useful for refining measurements (pilot testing) - difficult to code Structured self-report questions - easy to code - more difficult to get right
• Wording Questions Well - simple is better - informal language - no negatives - double-barreled questions should be avoided - forcing a choice is bad - avoid floor and ceiling effects - avoid loaded questions - make questions relevant to all participants - multiple items for the same construct - mix it up - all is relative (context) - ease into socially sensitive issues - ask sensitive questions sensitively (and cleverly) - quarantee anonymity
• The Response Translation Phase number of scale points choosing anchors to create equal appearing intervals unipolar vs. bipolar scales (-5 -4 -3 -2 -1 0 1 2 3 4 5, bipolar)
• Experiment Treatment group and control group Randomization - Randon sampling - Random assignment
• Random sampling external validity selecting people to take part in a study on a totally arbitrary basis
• Random assignment internal validity placing people in conditions on a totally arbitrary basis
• Threats to validity Confounds and artifacts - individual differences: selection, non respone bias - changes over time: history, maturation, regression towards the mean - studying people changes people: mere measurement, testing effects, mortality, participant bias, experimenter bias
• Confound Should have been held constant but was accidentally allowed to vary (and covary with the independent variable) Threat to internal validity
• Artifact a variable that: - (a) is held constant in a study or series of studies and that - (b) might represent a restricted context under which the effect will be observed   Should have been systematically varied, but was held constant Threat to external validity
• Selection bias Choosing participants from a nonrepresentative sample by using imperfect sampling techniques rather than true random sampling mostly threat to external validity
• Anzeige
• Non-response bias Occurs when a substantial proportion of those invited to take part in a study refuse to do so mostly threat to external validity
• History Changes that occur over time in a large group of people When conducting a pretest-posttest study in which all participants receive a treatment, changes due to history may masquerade as treatment effects Threat to internal validity
• Maturation Changes that occur over time in a specific person or group of people due to normal development or experience (e.g., growth, learning) When conducting a pretest-posttest study in which all participants receive a treatment, changes due to maturation may masquerade as treatment effects Threat to internal validity
• Regression towards the mean The tendency for people who receive extreme scores on a specific measure to score closer to the mean when given the test/measure at some point in the future Performance = (almost) true score + error/luck Threat to internal validity
• Mere measurement effect Tendency for participants to change their behaviour simply because they have been asked how they will act in future Threat to external validity
• Testing effects Tendency that performance is better on a test or personality measure the second time it is measured Reason: Probably reflect learning and attitude polarization Threat to internal validity
• Mortality/Attrition The failure of some research participants to complete an investigation Homogeneous attrition: Attrition rates are equal across experimental conditions -> primarily threat to external validity Heterogeneous attrition: Attrition rates are different across experimental conditions -> primarily threat to internal validity
• Participant bias When research participants realize they are being studied and behave in ways in which they normally would not (1) participant expectancies (try to behave in ways they believe to be consistent with the experimenter's hypothesis) (2) participant reactance (try to disconfirm an experimenter's hypothesis) (3) evaluation apprehension (people's concerns about being judged favorably or unfavorably by another person) Typically threat to internal validity Demand characteristics: Aspects of an experiment that subtly suggest how participants are expected to behave (contribute to participant expectancies)
• Experimenter bias When the investigator's expectations about participants lead to false support for these expectations Experimenters interpret ambiguous behaviours in ways that are consistent with their expectations Experimenter treat participants differently in different conditions
• Counter measures Random assignment Replication Blind or double-blind experiment
• True experiment A research design in which the investigator maipulates one (or more) variables and then assesses the impact of this manipulation An experiment isolates the effect of a variable by holding all other variables constant while this variable is manipulated   eliminate confounds
• Two ingedrients (of an experiment) Use of manipulation Random assignment to condition
• Advantages of (lab) experiments Experiments conducted in the laboratory make it easier to eliminate confounds than experiments conducted in the field Experiments allow researchers to observe things that they could not observe any other way Experiments minimize noise (extraneous variables that influence the DV but are evenly distributed across experimental conditions) Experiments are well suited not only to the isolation of variables but also to the systematic blending of variables: To analyze interactions between two or more independent variables
• Drawbacks of experimental designs Lack of realism Possible low external validity Solution: Experimental (psychological) realism
• One-way and factorial designs One way design: one independent variable Factorial design: two or more independent variables that are crossed (each variable appears in combination with every other independent variable)
• Factorial designs most common design: 2x2 Main effects Interaction effects (qualification)
• Interaction. The effect of an independent variable on the dependent variable is different at different levels of another independent variable. The presence of an interaction means that one cannot make a simple, blanket statement about the independent variables in a factorial study
• Quasi experiment A research design in which the researcher has only partial control over his or her independent variables. Assignment to conditions: No use of random assignment e.g. natural experiments, person-by-treatment quasi-experiments
• Why quasi-experiments? True experiments are not conducted, because - it is simply not possible - it is difficult - it should not be done (ethics) True experiments have been conducted, but - the external validity is in question
• Natural experiments No use of random assignment at all Use of naturally occurring 'manipulations'
• Anzeige
• Person-by-Treatment Quasi-Experiments At least one independent variable is measured and one other independent variable is manipulated Person-by-treatment designs are always factorial designs
• Patched-up designs Patching: Adding new conditions to a quasi-experiment to help establish the size of an effect, to test for the influence of conceivable confounds, or both. A set of quasi -experimental findings in which a researcher has made repeated use of patching - the addition of control groups - to refine and strengthen the design. Collecting multiple waves of data from two presumably comparable groups of research participants. If the two groups of participants receive highly similar scores on a dependent measure prior to a natural manipulation but begin to receive different scores immediately after the natural manipulation, this strongly suggests that the manipulation may be responsible for the observed post-manipulation differences.
• Case studies A systematic study of the experiences of a particular person or group of people Sources of inspiration for formation of theories and/or further research Extraordinary experiences Uncovering general psychological principles Detailed observations Conditions often hard, impossible or unethical to replicate
• What makes a case study scientific? Goal: development or refinement of theories of human behaviour Explaining extraordinary events in terms of established scientific principles Case studies are used to develop and sometimes test theories that are open to falsification Scientific statements and analyses are open to disconfirmation
• Single variable reserach Census Population survery: - a survey designed to provide a description of the members of a population by sampling a representative portion of that population. - Also: epidemiological research - Also: research on public opinion Convenience sample
• Multiple variable research Studies of correlational nature - Archival research - Obseravtional research - Questionnaires - Interviews Confounds - Person - Environmental - Operational
• Archival research Research in which investigators examine naturally existing public records to test a theory or hypothesis. e.g. Police reports, hospital records, newspaper circulation rates... High in external validity However, need to maximize internal validity
• Correlational studies Correlation is not causation!! Post hoc ergo propter hoc Reverse causality/bidirectionality problem Third Variable Problem
• Research ethics according to Goodwin (2002): Behaving in a way that is morally correct - to do what is right a. Treat human research participants with respect and in a way that maintains their rights and dignity b. Care for the welfare of animals when they are subjects of research c. Be scrupulously honest in the treatment of data
• Ethical guidelines Means-end assumption Risk-benefit analysis Internal review boards (IRBs) Informed consent Freedom from coercion Confidentiality Debriefing
• Four ways of knowing intuition logic authority observation
• Positive test bias the tendency for people who are evaluating hypotheses to attempt to confirm rather than to disconfirm these hypotheses
• behavioral confirmation the tendecy for social perceivers tp elicit behaviors from a person that are consistent with their initial expectancies of the person
• paradoxical incidents puzzling or nonsensical observations
• analyzing the practitioner's rule of thumb analyzing things that experts in a particular are do to achieve certain outcomes
• serendipity luck or good fortune
• hypothetico-deductive method begin with a set of basic assumptions (or observations) and derive one or more logical consequences from these basic principles
• focus group a small but representative sample of participants from the group a researcher wishes to understand meet together to discuss their experiences
• pseudo-experiment research design in which someone tests a claim about a variable by exposing people to the variable of interest and noting that these people fell, think, or behave as expected   no control group
• participant reaction bias the bias that occurs when people realize they are being studied and behave in ways that they normally wouldn't

zurück  |  weiter 2 / 3