Help
Options
Didn't know it?
click below
click below
Knew it?
click below
click below
Don't Know
Remaining cards (0)
Know
retry
shuffle
restart
0:00
Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.
Normal Size Small Size show me how
Normal Size Small Size show me how
PSY-32 Exam 2
| Question | Answer |
|---|---|
| What are research ethics? | ethics: A set of principles guiding right conduct; The rules or standards governing the conduct of a person or the members of a profession. •American Heritage Dictionary |
| APA “top 5” standards for conducting ethical research | •Freedom from coercion •Protection from physical and psychological harm •The risk-benefit rule •Informed consent •Debriefing |
| The risk-benefit rule | 1. the benefits must always exceed the costs. 2. minimal risk: any harm or discomfort while participating is no greater than that experienced in daily life or during routine physical or psychological tests. 2a. if more than minimal risk, safeguards needed |
| Informed consent | Consent process needs to contain sufficient information for participants to be able to make an informed decision about participating. |
| Informed consent 2 | Sufficient information: •Nature of the experiment •Overview of procedures •Time •Potential risks and benefits •Can’t withhold information that might seriously influence a participant’s willingness to participate |
| Deception | •refers to withholding information from participants or intentionally misinforming them about an aspect of the research •not divulging the exact hypothesis does not necessarily count as deception |
| Deception: Pros | •allows researchers to study individuals’ natural behavior. •allows opportunities to investigate behaviors and mental processes not easily studied using nondeceptive methods. |
| Deception: Cons | •contradicts the principle of informed consent. •the relationship between researcher and participant is not open and honest. •frequent use may make individuals suspicious about research and psychology. |
| Debriefing | •educate participants about the need for deception or discomfort. •educate participants about the goals of the research and hypotheses being tested. •should attempt to leave participants in as positive a state as started. |
| Debriefing, pt 2 | •learning experience: e.g. 1. importance of psychological science. 2. exposure to common methods and measures |
| Who is responsible for ensuring ethical research? | •The researcher! •Institutional review board (IRB). |
| Institutional review board (IRB). | 1. committee of individuals who review research protocols to ensure the ethical appropriateness of the planned research. 2. protocol includes: •a careful but nontechnical summary of the goals of the research. |
| Institutional review board (IRB) pt2 | •a detailed description of the procedures to be followed. •consent and debriefing forms. |
| Who is responsible for designing ethical studies? | The researcher, I think. |
| Five general principles of ethical decision making | •Beneficence and Nonmaleficence •Fidelity and Responsibility •Integrity •Justice •Respect for People’s Rights and Dignity |
| Beneficence and Nonmaleficence | strive to benefit; do no harm |
| Fidelity and Responsibility | establish relationships of trust; accept responsibility |
| Integrity | promote accuracy and honesty |
| Justice | equal benefits and quality for all |
| Respect for People’s Rights and Dignity | respect dignity and worth; everyone has right to privacy, confidentiality, and self-determination |
| privacy | right to decide whether personal information is communicated to others beyond the research team |
| confidentiality | right not to have personal information linked to your identity |
| How might one use the general principles to guide ethical decision-making in the design of a study? | Choose a topic that benefits society or the participants. (finish the rest later. Make basic cards first) |
| Three steps you should follow if a research participant displays signs of psychological or physical injury | 1. Stay calm and think clearly. 2. Don't leave the participant in an emergency until it is clear that someone else is on the scene and is capable of assuming your responsibility. 3. Notify the dean and describe the events that took place.(If Tufts student |
| Three steps, third step elaboration | 3. If the research participant involved is a Tufts student, notify the Dean of the school in which the student is enrolled, and describe the events that took place. |
| Three main sources of assistance at Tufts (in case of emergency) | the Tufts Police, Health Services, and the Counseling Center |
| Ethics summary | Bottom line: Rarely are there absolute right or wrong answers. 1. The process is important. 2. Steps for ethical decision making: a. Consider the facts of the situation (procedures, participants, etc.). |
| Ethics summary, part 2 | b. Identify the relevant ethical issues from all viewpoints (the science, society, participants, general principles). c. Consider alternatives to the methods under consideration and the consequences of not doing the proposed research. |
| Ethics summary, part 3 | d. Decide on what to do. |
| Cause | •a contextually dependent event that makes something else exist. •In an experiment, the cause is the variable that the researcher manipulates: the independent variable (IV) |
| Effect | 1. the difference between what did happen and what would have happened to the same group of individuals. 2. In an experiment, the effect of the IV is observed in the variable the researcher measures: the dependent variable (DV). |
| Effect part 2 | 3. identifying a true effect requires that participants simultaneously be and not be exposed to something. 3a. we estimate the true effect by using control conditions |
| John Stuart Mill’s three criteria for causality | •Variation: changes in one variable (cause) must correspond with changes in the other (effect). •Temporal sequence: cause must precede the effect. •Eliminating confounds: no other explanations must exist for the effect. |
| Independent variable | The variable in an experiment that is manipulated by an experimenter. The experimenter expects the independent variable or variables to influence the dependent variable. |
| Dependent variable | The variable in an experiment that is measured by an experimenter. The experimenter expects the dependent variable to be influenced by the independent variable. |
| Provide 3 examples of DVs and 3 of IVs | (hint: you need to come up with 3 scientific hypotheses first) |
| Experiment | •“A research design in which the investigator manipulates 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” |
| Matching | A method of assigning participants to experimental conditions. Researchers who use matching try to identify similar individuals in order to place them in different conditions. |
| Matching, why it was done and why it’s not as often anymore | It was done to gain some degree of comparability between experimental conditions (i.e., to minimize the chance of a person confound). This strategy is inferior to random assignment. |
| Two key features of true experiments | 1. Random assignment of participants to groups. (if there are groups) 2. Manipulation of the IV (at least two levels). |
| Ways to Create Variation in the Independent Variable | •Presence versus absence, e.g. Drug v. No Drug •Amount of stimulus , e.g. 0 mg Drug v. 1 mg Drug v. 5 mg Drug •Type of stimulus, e.g. Happy film v. Scary film •Instructional manipulation, e.g. This puzzle is solvable v. This puzzle is not solvable. |
| Event manipulation vs Instructional manipulation | Event manipulation: manipulate characteristics of the context, setting, etc. Presence, amount, type. Instructional manipulation: different groups are given different instructions. e.g. This puzzle is solvable v. This puzzle is not solvable. |
| extraneous variable (AKA nuisance variable) | any variable other than the IV that may influence the DV. 1. Possible examples: age, IQ, race, sex, last time you had a fight with someone else, whether or not you just ate ice cream, personality traits. 2. Introduces noise |
| confounding variable | any variable other than the IV that may influence the DV *and* that systematically varies with the IV. Example: Number of people in the room (10, 2, or none) is confounded with experimenter behavior (surly for 10, friendly for 2 and none). |
| Manipulation Check | • a measure designed to see if an IV actually put participants in the psychological state that the experimenter wished to create. |
| Manipulation Check, relation to construct validity | Sort of like a test for construct validity: tries to find out if you really measured what you thought you were measuring. |
| Manipulation Check, why p1 | •If IV has no effect on the DV: •Maybe the IV really has no effect on the DV. •But, (see p2) |
| Manipulation Check, why p2 | could be because we simply didn’t do a good job manipulating the IV of interest: The manipulation check tells us whether we can rule out an unsuccessful manipulation of the IV as an explanation for no effect of the IV on the DV. |
| Random Assignment | •assigning participants at random to the conditions in your experiment ▫ every person has an equal chance of being assigned to each condition ▫ equates groups by randomly distributing the extraneous variables over the treatment groups |
| Random Assignment, p2 | ▫ this makes it very unlikely that any extraneous variable provides an alternative explanation for the effect of the IV on the DV |
| Between-subjects | Each participant serves in one and only one condition of an experiment . |
| Within-subjects (AKA repeated measures) | Each participant serves in more than one condition . . |
| Within-subjects advantages | No individual differences contributing to error variability. Fewer participants required |
| Within-subjects disadvantages | •Increased likelihood that participants will guess the hypothesis •Sequence Effects •Carryover Effects |
| Between-subjects advantages | |
| Between-subjects disadvantages | |
| Sensitivity | the statistical power to detect a significant difference between conditions when there is a real difference to be detected |
| Why do within-subjects designs have greater sensitivity than between-subjects designs? | Within subjects designs eliminate error variability due to individual differences. Compare to t: t = systematic variability/error variability |
| Sequence effects | •passage of time takes a toll on participant responses (e.g., perhaps because they get tired or bored). A problem for within-subjects design because participants undergo multiple treatments, and certain treatments occur at later times than others |
| Carryover effects | •response to one condition influences their response to a second condition. |
| Note on sequence and carryover effects | • Sequence and carryover effects can occur at the same time… |
| Sequence/Carryover effects summary | they may confound within-subjects designs if… •… participants receive the different levels of the IV in the same order AND •…condition order affects the DV . |
| May need to elaborate on that. Confusing | |
| If Sequence/Carryover effects are present | In that case, the pattern of differences between the means for your DV are subject to at least two plausible interpretations: 1. the IV explains the differences in the DV 2. a sequence/carryover effect explains the differences in the DV |
| Counterbalancing | giving different orders of within-subjects treatments to different (groups of) participants. Helps reduce sequence effects and carryover effects. |
| Add a note on order effects? | |
| Overall types of counterbalancing | 1. complete counterbalancing 2. incomplete counterbalancing: 2a. reverse 2b. partial 2c .Balanced Latin Square |
| complete counterbalancing | presenting every possible order of all the experimental conditions, with an equal number of participants randomly assigned to each of the different orders. |
| reverse counterbalancing | •pick one possible order (e.g., D H N) and then reverse it (N H D); half subjects randomly assigned to one order, half randomly assigned to the other. •average serial position the same, but some conditions always occur in the same serial position |
| partial counterbalancing | •randomly select a sample of orders from the population of orders. •some sequences (D to H, or H to N, etc.) may be repeated more frequently |
| partial counterbalancing, partial example | •with 6 conditions, there are 6! or 120 possible orders; randomly select 10 of them and assign subjects to the orders at random. |
| Balanced Latin Square | • generate a fixed number of orders meeting two criteria: 1. each condition appears exactly once in each serial position. 2. each condition must precede and be followed by every other condition an equal number of times |
| Latin Square Example | 1. A B D C, 2. B C A D, 3. C D B A, 4. D A C B |
| demand characteristic | • characteristic of an experiment that subtly suggests how people are supposed to behave |
| Within-Subjects Advantages version 2 | • There are no individual differences contributing to error variability when comparing conditions: the same people are in every condition! (age, gender, prior experiences, etc. are held constant) • Increases sensitivity; Fewer participants are needed |
| Within-Subjects Advantages version 2, example | Let’s say you want to study the combined effects of task difficulty (low, medium, and high), time pressure (low, high), and type of task on performance (verbal, mathematical, analytical) |
| Within-Subjects Advantages version 2, example 2 | If you need 30 Ps to get a good estimate of the mean for each condition, you’d need 3 X 3 X 2 X 30 = 540 Ps (Between-subjects may need something like 30 per condition. More like 10 for within.) |
| Figure captions | Goes below the figure. •Succinctly describe what’s depicted such that the reader doesn’t have to refer back to the text to understand what’s being shown. •Includes caption “Figure X.” (In italics) at the start. |
| Figures | •Figures include graphs, charts, drawings and pictures. •Useful for visualizing summarizing one’s procedures or findings. |
| What makes a good figure? | (Directly from the APA manual) •augments rather than duplicates the text •conveys only the essential facts •omits visually distracting detail •is easy to read •is easy to understand |
| More notes on figures | Don’t leave a figure hanging… • Must refer the reader to figures in the text: 1. Average performance in all three conditions was abysmal (See Figure 1). 2. Figure 1 depicts the mean rating of fun for each of the four groups. |
| Limitation of one-way designs | •Only manipulating one factor that may cause an effect on the DV, which is not very efficient… |
| Factorial Design | •A design that enables us to investigate the independent and combined influences of more than one independent variable |
| (probably elaborate on this) | |
| Factorial design research question example | •Is the effect of violent TV on aggressive behavior lessened by soothing music? |
| Factorial design elaboration 1 | The IVs are completely crossed: every level of every IV appears in combination with every level of every other IV. |
| How do we determine the independent and combined influences of these two IVs on the DV? | ANOVA |
| 2x2 factorial ANOVA, what it assesses | • Main effect of IV 1 (Type of TV) • Main effect of IV 2 (Presence of music) • Interaction between IVs (Type of TV * Presence of Music) |
| main effect | the overall influence of one IV, as reflected in the difference between the means |
| Interaction | • the effect of one IV depends on the level of another IV • two ways to graph and interpret the same interaction. (Depending on which IV is on the x-axis and which one is split.) |
| Interaction, choosing graphs | •no matter how you graph and interpret it, it’s the same interaction; either way is fine. •choose the one that makes most sense to you. |
| Interaction, graph labeling | •Label tells you: 1. how many IVs exist in the design. 2. how many levels each IV has. |
| Types of factorial designs, Between | • all IVs are manipulated as between-subjects variables |
| Types of factorial designs, Within | • all IVs are manipulated as within-subjects variables |
| Types of factorial designs, Mixed | • IVs are manipulated as between- and within-subjects variables |
| Two common patterns of interaction | •crossover interaction •spreading interaction |
| (need to figure out more about how professor expects us to describe these.) | |
| Crossover interaction | 1) no significant main effects of either IV. 2) the effects of each IV are opposite at different levels of the other IV. |
| Spreading Interaction | an IV has an effect under some conditions but has less of an effect (or no effect at all) under other conditions |
| maybe it’s actually just clear to me from these | |
| Template for describing an interaction p1 | • In the ________ condition (IV 1, Level A), _________ (DV) was ________ (lower / higher / the same) in the ________ condition (IV 2, Level X) compared to the ________ condition (IV 2, Level Y). |
| Template for describing an interaction p2 | • However… • In the ________ condition (IV 1, Level B), _________ (DV) was ________ (lower / higher / the same) in the ________ condition (IV 2, Level X) compared to the ________ condition (IV 2, Level Y). |
| Possible findings for a 2 x 2 Factorial Design (first half) | 1. A main effect for IV 1 only. 2. A main effect for IV 2 only. 3. Main effects for both only. 4. A main effect for IV 1 plus an interaction. |
| Possible findings for a 2 x 2 Factorial Design (second half) | 5. A main effect for IV 2 plus an interaction. 6. Main effects for both IVs plus an interaction. 7. An interaction only, no main effects. 8. No main effects, no interactions |
| Possible findings for a 2 x 2 Factorial Design, how do you know what effects you have? | You have to compute the relevant statistics. |
| Given a table of cell means for a 2 X 2 factorial design, be able to compute the relevant means for two main effects. | I know this very well. |
| Be able to graph and describe an interaction in words. | Ok, but maybe study this from the slides later. |
| What are multiple group experiments? | •one-way design in which there is one IV with three or more levels. •can be manipulated on a between-subjects or within-subjects basis . |
| What are three advantages to using multi-group designs? | 1. Increases external validity. 1a. Most real-world situations arenât that simple. 1ai. Good vs. Bad 1aii. Smart vs. Dumb 2. Helps you identify the shape of the relationship between the IV and the DV. 3. Helps you improve construct validity. |
| Multigroup experiments 2 | Really look at the examples, very useful. |
| Multigroup experiments 2, example summary | Other concepts can account for the ones you are using (construct validity), and the other concepts could fall on different scales from each other and the one you are using. |
| Multigroup experiments 2, example summary, elaboration | One could have a positive side on a scale while yours or the other does not. So, if you add a positive condition, you could find which one. Could be true for other combinations of variables and conditions as well. |
| Quasi-experiment | 1. Research designs in which the researcher has only partial control over their IV. 1a. often this means they lack random assignment to conditions. 1b. other times it means there is no control condition. 2. This means internal validity may be threatened. |
| Good reasons to conduct quasi-experiments, part 1 | 1. some things that are important to understanding behavior can’t be experimentally manipulated. 1a. gender, natural disasters, IQ. |
| Good reasons to conduct quasi-experiments, part 2 | 2. some topics that can be studied experimentally are very difficult to study experimentally. 2a. unavailable equipment or resources. 3. some topics that can be studied experimentally would be unethical to study experimentally. 3a. HIV, racial bias |
| POSTTEST-ONLY DESIGN WITH NONEQUIVALENT GROUPS | Simple “experiment” without random assignment. Participants assign themselves to groups. |
| Fixing POSTTEST-ONLY DESIGN WITH NONEQUIVALENT GROUPS, example | Does the online stat course increase understanding? • Tested two introductory probability classes: • one received stat course (treatment) • the other did not (control). |
| Fixing POSTTEST-ONLY DESIGN WITH NONEQUIVALENT GROUPS, example, issues with design | Problem: • Selection • The students who registered for the probability class that got SOCR might differ from the students who registered for the probability class that got the control |
| Fixing POSTTEST-ONLY DESIGN WITH NONEQUIVALENT GROUPS, partial solution, patching | Patching refers to testing for the influence of specific confounds. 1. conduct an internal analysis 2. add new conditions. |
| Patching, example of conducting an internal analysis | Limit analysis to math majors in both groups to see if the difference in mean performance between the online statistics course group and control groups remains (conducting an internal analysis) |
| Patching, example of adding a new condition | Collect more data from a new section of introductory probability; make sure iSIS enrollment system is set up to allow a specific proportion of math majors, same as in the control group (adding a new condition). |
| Why is patching still partial? | Even if you were to look just at math majors (or find out that the two groups had the same percentage of math majors), there are lots of other extraneous variables. E.g., if class who got SOCR has higher IQ or GPA, could explain their better performance |
| Why is patching still partial?, part 2 | Even if you can show that the two groups are matched on pretest performance, doesnât rule out possible threat to internal validity. ⢠Maybe a difference in motivation for the example. So, one group learns more. |
| ONE-GROUP POSTTEST ONLY DESIGN | No control group, no baseline. |
| ONE-GROUP PRETEST-POSTTEST DESIGN | No control group, basically. |
| SINGLE N DESIGNS | • Quasi-experimental designs that allow you to identify possible threats to internal validity by measuring the dependent variable multiple times before and after a manipulation. • Can be used with one participant (hence the name) or groups of participants |
| TYPES OF SINGLE N DESIGNS | •A-B Design (AKA Interrupted Time Series) •A-B-A Design •A-B-A-B Design •A-B-A-B-A Design |
| TYPES OF SINGLE N DESIGNS, note | they are all interrupted time series, technically. You can also do as many alternating repetitions as you want. |
| INTERRUPTED TIME-SERIES DESIGN | •quasi-experimental design in which a single participant or group of participants is tested repeatedly before and after an intervention •intervention can be a manipulation or a natural event. |
| INTERRUPTED TIME-SERIES DESIGN, more | Phase A. Allows you to establish stable (i.e., reliable) baseline. Phase B is after the intervention. |
| INTERRUPTED TIME-SERIES DESIGN, assumption | the pattern of pre-intervention responses would continue if you hadn’t introduced the intervention |
| INTERRUPTED TIME-SERIES DESIGN, problems, part 1 | PRONE TO HISTORY EFFECT. Baseline can be unreliable. (It would not actually continue even without intervention.) POSSIBLE REGRESSION TO THE MEAN. (if high variability in baseline, suspect regression to the mean.) |
| INTERRUPTED TIME-SERIES DESIGN, problems, part 2 | Other possible threats: maturation, mortality, instrumentation, testing. (selection is not possible because single n) |
| A-B-A design | remove treatment and take more measures at baseline. However, behavior may not return to baseline due to history effects or a treatment being permanent. |
| A-B-A design special case | the treatment is not withdrawn during the second baseline as in the A-B-A design. instead, the treatment is applied to an alternative but incompatible behavior during the second baseline. More powerful, because eliminates history effects to a degree. |
| A-B-A design example | positive reinforcement for reading, then removed at end of session. In special case, math reinforced instead in second baseline, and it rises after being at baseline for a long time. |
| A-B-A-B, A-B-A-B-A, etc. | can further extend the number of A and B phases in order to further reduce the possibility of threats to internal validity |
| Multiple-baseline designs and history effects | reduces chance of history effects, because multiple history effects would have to happen. |
| The 7 classic threats to internal validity | 1. History, 2. Maturation, 3. Regression Toward the Mean, 4. Testing, 5. Attrition (AKA Mortality), 6. Instrumentation, 7. Selection. |
| History | An event occurring between treatment and outcome that could affect the DV. |
| History, example | Effect of drug education program in schools using a pretest-posttest design. Heath Ledger dies between pretest and posttest. |
| Maturation | Changes in biological and psychological conditions that occur with the passage of time. • age, learning, fatigue, etc. |
| Regression to the mean | Extreme scores will tend to move towards the mean in their distribution. actual score = true score + random error. Metaphor: true score is âskillâ or âabilityâ, error is âluck.â ability doesnât usually change, luck does. (Not necessarily, bu |
| Regression to the mean, example | pre-test, choose bottom 1/3 of students for headstart, and post-test. Scores improve. But just regression to the mean? (To fix this problem a bit, pre-test headstart students specifically after the initial test.) |
| Testing | Changes in a participant’s score on repeat administrations of a test (aka practice effects) • Example: Effect of amphetamine on recall, list of 16 words for you to remember. Will remember better just because you heard the items repeated in the list. |
| Attrition (aka mortality) | Sometimes people don’t show up for, or don’t finish, an experiment. The longer the study, the greater the risk of attrition! 2. In studies with two or more conditions, attrition can take one of two forms. |
| Two forms of attrition | 1.homogeneous attrition. 2. heterogeneous attrition. |
| homogeneous attrition | 1. rate of drop-out is the same in different experimental conditions. 2. threat to external validity. |
| heterogeneous attrition | 1. rate of drop-out is different in different experimental conditions. 2. threat to internal validity. |
| Instrumentation, part 1 | Problem with the way the researcher measured the dependent variable. •Machinery may drift. •Human raters may get more reliable with practice. •Human raters may get bored or fatigued. |
| Instrumentation, part 2 | •The task may be way too easy (ceiling effect) or way too difficult (floor effect) |
| Selection | Different criteria used for assigning participants to conditions |
| Selection, example | • Effects of coffee on attention in morning classes: • 1st 10 through the door: Coffee • 2nd 10 through the door: Decaf |
| McGrathâs âthree-horned dilemmaâ | No study can address all three: 1. Precision 2. Generalizability to situations 3. Generalizability to people. (What is precision? Internal validity?) |
| McGrath dilemma for true (lab) experiment | Precision: Strength, Generalizability to Situations: Weakness, Generalizability to People: Weakness |
| McGrath dilemma for correlational (field) study | Precision: Weakness, Generalizability to Situations: Strength, Generalizability to People: Moderate |
| McGrath dilemma for quasi-experiment | Precision: Moderate, Generalizability to Situations: Moderate, Generalizability to People: Moderate |
| McGrath dilemma for population survey | Precision: Weakness, Generalizability to Situations: Weakness, Generalizability to People: Strength |
| McGrath dilemma, what to do? | Your best bet is to: •Know the advantages and disadvantages of a given method •Maximize a method’s advantages while minimizing its disadvantages •Construct a research program using multiple studies that compensate for one another’s weaknesses |
| Risk-benefit analysis | A comparison of the potential negative effects (or risks) in a study to its potential positive effects (or benefits). To receive accreditation by the APA, colleges and universities must have an IRB perform a risk-benefit analysis of all human studies. |
| Selection bias | choosing research participants from a nonrepresentative sample by using imperfect (i.e., biased)sampling techniques rather than true random sampling. This typically represents a threat to external validity. |
| Patching | a research method in which a researcher adds new conditions to a quasi-experiment to help establish the size of an effect, to test for the influence of a conceivable confounds, or both. |
| Participant expectancies | the form of participant reaction bias that occurs when participants consciously or unconsciously try to behave in ways they believe to be consistent with the experimenter’s hypothesis. Threat to internal validity. |
| See also demand characteristics for participant expectancies? | |
| Participant reactance | The form of participant reaction bias that occurs when participants attempt to assert their sense of personal freedom by choosing to behave in ways they believe to be in opposition to the experimenter’s expectations. Threat to internal validity. |
| Participant reaction bias | the bias that occurs when research participants realize they are being studied and behave in ways in which they normally would not behave. Most forms of participant reaction biased threaten the internal validity of an investigation. |
| Participant reaction bias, three forms emphasized in P & B | (1) participant expectancies, (2) participant reactance, and (3)evaluation apprehension. |
| Natural experiment, part 1 | A kind of quasi-experiment in which the researcher makes use of archival data documenting the consequences of a natural manipulation such as a natural disaster or a change in traffic laws in a particular state. |
| Natural experiment, part 2 | The best natural experiments typically involved arbitrary or near-chance events that affect a large group of people. |
| Hawthorn effect | the increases in productivity that may occur when workers (participants?) believe that their behavior is being studied or believe that they are receiving special treatment. |
| Hawthorn effect, explanation | Because participants who are receiving an experimental treatment are more likely to believe these things than are participants in a control condition, Hawthorne effects may be mistaken for treatment effects and thus are a threat to internal validity. |
| Freedom from coercion | The ethical principle of respecting participants’ rights to drop out of a study if they choose to do so. This mainly consists of making it clear to participants that they have the right to stop participating without any fear of negative consequences. |
| Evaluation apprehension | the form of participant reaction bias that occurs when participants attempt to behave in whatever way they think will portray them most favorably. Evaluation apprehension is a threat to internal validity. |
| Median split design, part 1, basic category | An approach to selecting people for inclusion in a laboratory study. |
| Median split design, part 2, definition | Experimenters pretest participants and sort into two groups depending on who scored in either the top half or the bottom half (i.e., people above or below the median) on an individual difference measure of interest. |
| Median split design, part 3, the qualification/limitation | This approach is usually inferior to the extreme groups approach. |
| Extreme groups approach | an approach to selecting people for inclusion in a laboratory study. Experimenters recruit people to take part in a study only if such people receive extreme (i.e. very high or very low) scores on an individual difference measure of interest. e.g. top or |
| Person-by-treatment quasi-experiment | a research design in which the researcher measures at least one independent variable and manipulates at least one other independent variable. Always factorial. |
Created by:
ncommons
Popular Psychology sets