Experimental Psych Final

Term
1 / 133
What is the Hawthorne Effect?
Click the card to flip 👆
Terms in this set (133)
What is the Hawthorne Effect?
individuals modify aspect of their behavior when they know they are being watched
good-participant role
Participant gives us the answers we want because they know we want that answer, people pleasers
Bad Participant Role
Participants purposely try to sabotage our research
Honest Participant
Participant gives us what we want by giving truthful answers
Anxious Participant
Participants are scared to give out answer
What are the stages of creating a survey?
Stage 1- Survey design and planning
Stage 2- Piloting
Stage 3- Final survey design and planning
Stage 4- Data collection
Stage 5- Data coding, analysis, and reporting
What are two pros of surveys?
Cheap, easy to administer, can help develop a hypothesis, easy to quantify responses
What are the two cons of surveys?
Unverifiable and could be invalid data
List two types of survey questions?
Yes/No, Forced Choice, Multiple Choice, and Open-ended
Which of the following is not a survey tip?
a. Make sure language and measures are consistent
b. Do not ask for personal question unless it is needed
c. Place sensitive questions at the beginning of the survey
d. Include clear instructions
Place sensitive questions at the beginning of the survey
What does WEIRD stand for?Western, Educated, Industrialized, Rich, DemocraticWhich of the following is an example of probability sampling? a. Snowball Sampling b. Multistage Sampling c. Quota Sampling d. Convenience SamplingMultistage SamplingWhich of the following is an example of non-probability sampling? a. Stratified Sampling b. Cluster Sampling c. Purposive Sampling d. Equal Opportunity SamplingPurposive SamplingWhat is the central limit theorem?As n increases the distribution becomes shaped more normallyWhat are the two biases in sampling?Convenience sampling and Self-selected samplesList two fundamental requirements of an experimental design?a. Random selection and assignment, independent groupsWhat can a researcher do when random assignment does not work?The researcher can purposefully assign using within-group designsIn an experiment, when can you not used matched pairs?a. If you cannot measure the variable that may cause differences in scores (outside of IV influence)List one problem of repeated measuresCannot remove effects of IV, can you measure the DV more than once, and the practice effectWhat is natural pairs?Participants are matched based on biological or relational lifeWhich internal validity threat refers to any event other than the planned treatment that occurs between pre-post testing that affects the DV? a. Maturation Threat b. Regression Threat c. Testing Threat d. History ThreatHistory ThreatTrue or False: Selection threat is if groups are not equal before the experiment, these differences may disrupt internal validityTrueTrue or False: Regression threat is when participants leaving a study can alter the balance between groups as well as other components related to statistical sensitivityFalseTrue or False: A control of a testing threat is to increase time between the pretest and the posttest.TrueTrue or False: Regression to the mean often occurs because of poor sampling from the populationTrueTrue or False: Factorial designs are useful when researchers want to investigate potential interaction effectsTrueTrue or False: Spreading interactions depend on the other variablesFalseTrue or False: Factorial designs allow us to include more than 2 groupsTrueIf the first factor has 3 levels and the second factor has 4 levels how many groups will the experiment have? a. 7 b. 12 c. 10 d. 612There are two kinds of two group designs, independent and dependentTrueDependent group designs are considered truly experimentalFalseIndependent group designs have random assignment and selectionTrueMatched pairs are an example of independent testsFalseDependent tests require a lower threshold for variance because groups are made off of shared characteristics or similaritiesTrueInternal validity is the ability to demonstrate that your independent variable is causing the observed changes in the dependent variable.TrueRegression is beneficial to internal validity because it strengthens relationships.FalseRegression occurs with extreme on clinical populationsTrueMortality threats involve participants leaving your experimentTrueMortality threats are uncommon in longitudinal studies.FalseFactorial designs are popular because of their simplicityFalseFactorial designs are just dependent testsFalseFactorial designs look for thresholds and limitsTrueFactorial designs combine dependent and independent testsTrueFactors is another word for independent variableTrueAn ethnographic study is conducted by the researcher immersing themselves into a culture to study it.TrueWhat kind of qualitative method includes observations of a single individual? Case studies Narrative studies Focus group Historiographiescase studiesIn assessing trustworthiness for qualitative studies there are 4 important factors, creditability, dependability, confirmability and ___________.transferabilityWhat is confirmability?Extent to which a qualitative research report is accurate, unbiased and can be confirmed by others.What is the difference between descriptive and analytical surveys?Descriptive: easy numbers, demographic data Analytical: attempts to test theory, explore and test associations between variablesHow many stages are included in the survey process?fiveWhich is not a stage of the survey process? Design and planning Data coding Data collection Pre-reportingPre-reportingWhy is the construction of questions in surveys so important?Poorly constructed questions can cause the participants to feel uncomfortable or lead them to an answer that may not be accurate.Researchers need to be mindful of survey delivery to make sure their sample is truly random.TrueWhat is the difference between a population and a sample? A population is the whole group, while a sample is just one part of the population A sample is the whole group, which a population is just one part of the sample There is no difference between a sample and a population A sample is one group while the population is another, they are compared to each otherA population is the whole group, while a sample is just one part of the populationWhat are the 4 multistage sampling stages?Counties, Segments, Households, and IndividualsA type one error is failing to reject the null when it is actually falseFalseWhat sets the critical threshold? Sigma Alpha Theta BetaAlphaWhat is the Principle of Parsimony?This is the principle that says explanations of phenomena should stay as simple as possible until a more complex answer is necessaryDon't put the _______ cart before the ________ horseDesign, HypothesisThe Principle of Parsimony is the idea that explanations of phenomena should remain simple until simple explanations no longer work.TrueWhat is true of a true experiment? Able to establish causality. An example of a true experiment is a independent 2-group experiment There must be two groups All of the above are trueAll of the above are trueWhy are perfectly matched pairs created in a repeated measure experiment?Participants scores are compared to their own scores after an independent variable has been introduced.Which dependent experimental designs should be used for a study on twins? Matched pairs Repeated measures Natural pairs A dependent design should not be used for twin studiesNatural pairsWhat is internal validity?The ability to say that the reason your DV has changed is only because of your IV.What is not a way we can establish cause and effect? Temporal precedence Co-variation Establishing no plausible alternative explanations ConfirmabilityConfirmabilityA history threat is the threat that the participant has prior knowledge of the experiment therefore changing their reaction to the IV.FalseWhat are some possible controls to prevent testing threats?Avoid pretests, alter the instrument, split-half administration, increase time between pre and post testInstrumentation threat is no longer an issue because we have climate controlled labs.FalseWhat makes factorial designs different from other designs we have looked at in class?More than two groups are allowedIn a factorial design, how do you find how many groups will be in a particular design?Multiply number of levels by number of factorsFactors are the same thing as an independent variable.trueIf you have 3 factors and 2 levels, how many groups would you have? 6 5 4 36event samplingObservation triggered by event of interesttime samplingObservation at different times of the dayLocation samplingSampling in different placesWhich one does not belong when looking to see if something is trustworthy? confirmability dependability credibility transferability all of the aboveall the aboveWhat is the Presentation effect?The four types of participants that you will have during the studies process which are the good, the bad, the honest, and the anxious participants. You have to look through them to find the honest participants because that is who you want to study.An analytic survey is used to describe a sample whether its size, behaviors, or opinionsFalsePro of yes/no surveysThey are easy to understand and a clear answerPro of forced choice surveysIt gives insight into their decision makingPros of multiple choice surveysIt gives more insight on how they feelPros of open minded surveysAllows responses to define central issues and variablesWhat are they different ways you can give a survey?Mail, personal interview, internet, phoneEqual opportunity sampleAnyone has an equal chance to get drawn for the studyConcerned with external validitySimple random sampling that gathers random participants and randomly assigns them to a random groupCluster samplingGathering whole groups of participants and then sorting them into specific groupsMultistage samplingStage 1 (Counties) à Stage 2 (Segments) à Stage 3 (Households) à Stage 4 (Individuals)Stratified random samplingCharacteristics you are focused on are proportionally represented in your sample and to the populationWhat do you need for a two-group design? a. 1 IV, 1 DV b. 1 IV, 1 DV, 2 Groups c. 2 IV, 2 DV d.None of the above1 IV, 1 DV, 2 GroupsWhen trying to think of a hypothesis you should always have a design or method on how you would conduct the experimentFalseWhen starting a two-group design, the general plan for it is selecting participants, assigning groups, trying to control extraneous and nuisance variables, and figuring out how to gather dataTrueIs this a good and safe experimental two-group design setup? You are studying if therapy is effective when trying to treat depression. Your independent variable is seeing a therapist and your dependent variable is depression symptoms. Your participants are clinically diagnosed with depression you make your participants get off any medication they are on and then split them into two groups one goes and sees a therapist and the other doesn't do anything.No, this is not a safe design because you are making the participants get off medication an not providing them with a safe alternative to keep their depression levels down.A dependent 2-group design is a true experimental designFalseMatch pairs design and quasi-experimental design is a category in dependent 2-group designTrueMatched pairs is a random assignment set-upFalseWhich is not an independent 2-group design a. Pre/posttest b. Between subjects c. Matched pairsMatched pairsIn a dependent 2-group design random assignment is impossible because there may not be enough participants to draw information fromTrueHistory threat is when the participants themselves change over timeFalseMaturation threat is when any event other than the independent variable occurred before or during the experiment that messes with the results and it usually only effects one groupFalseTesting threat is changes in scores because of the research design or instrument, like taking the exact same pre-posttestTrueInstrumentation threat is problems with the instrumentation that could be caused by time.TrueRegression threat is when participants that were outliers start to get closer to the mean over timeTrueInteraction effect is when interaction occurs when the impact of one of the independent variables depends on the level of another independent variable.TrueCrossover interactions are "only when"FalseSpreading interactions are "it depends"FalseThis study is looking into different types of influencers (athletic, model, food influencer) and the effect on middle school and high school students body image. How many conditions or groups would this experiment have? a. 2 b. 4 c. 6 d. 86In a study you are researching the effect online vs in-person classes have on traditional and non-traditional students. How many conditions or groups would this experiment have? a. 2 b. 4 c. 6 d.84Which of the following is not a Naturalistic sampling technique? Time-sampling Event-sampling Overview-sampling Location-samplingOverview-samplingWhich of the following are disadvantages of Naturalistic Observation? (If more than one, circle all answers) Hawthorne effect High ecological validity Presentation effect High external validityHawthorn and Presentation effectThe time-sampling technique samples observations at different points during the or at regular fixed intervalsTrueEvent-sampling enhances trustworthiness of data in terms of external validityFalseLocation-sampling is helpful for longitudinal studies or when there is concern about environmental effectsFalseWhich type of survey are used to describe a sample: size, behaviors and/or opinions? Analytical survey Questionnaire Descriptive surveyDescriptive surveyAnalytical surveys attempt to test a theoryTrueSurveys, questionnaires, tests and inventories can only be used for experimental researchFalseWhich of the following is not a pro of survey use? Easy to administer Easy to quantify responses Can help develop a hypothesis Very expensiveVery expensiveThe most basic two-group design has how many levels of the IV? 1 2 3 42Which of the following is NOT a goal of experimental research design? Establish temporal precedence of IV to DV Demonstrate covariance of effect Maintain internal validity to eliminate competing causal explanations Make sure the IV is competing with the DVMake sure the IV is competing with the DVA research proposal is a written statement of what? Introduction Results Experimental designExperimental designA larger test statistic means that it is less likely our results were arrived at by chanceTrueOur test statistics (t, F, r) are derived by dividing the between-groups variability by the error variabilityTrueWhich of the following is not a counterbalance of within-group designs? Order of questions Order of treatments Order of patients Position of itemsOrder of PatientsWhich of the following is not an advantage of independent-group designs? More simple than correlated-groups measures Establishes causality Sometimes you cannot create a matched pairs design You will always have a matched pairs designYou will always have a matched pairs designMatched pairs, Repeated measures, and Natural pairs are known as within-subjects comparisonTrueInternal Validity is the most important property of any experiment.TrueRegression to the mean often occurs because of poor sampling from the populationTrueIn selection threat, if groups are not equal before the experiment, these differences may disrupt internal validityTrueWhich of the following is not a threat to internal validity? Mortality/attrition threat Selection threat Regression threat Progress threatProgress threatWhich of the following is not a type of generalization? Population generalization Environmental generalization Social generalization Temporal generalizationSocial generalizationWhich of the following is not a potentially significant effect in factorial designs? Main effect Reaction effect Interaction effectReaction effectAn interaction effect occurs when the impact of one IV depends upon the level of another IV.TrueThe variable that the interaction effect depends upon is referred to as a moderating variableTrueWhen do you want to investigate potential interaction effects? when this effect/variable has already happened when this effect/variable is present Before this effect/variable is going to happenwhen this effect/variable is present