Psych 303- Exam 3

How are confounds related to null effects?

Confounds can actually counteract some true effect of an independent variable (ex: experimenter is encouraging in stress condition but not control situation).

What is a moderator?

If there is a statistically significant interaction, then the second independent is a moderator.

What validity are control variables most important for?

Internal Validity.

What is the most important validity when interrogating an experiment?

Internal validity.

What is Attrition?

Occurs when people drop out of a study before it ends.

What is the between-subjects post-test?

When participants are randomly assigned to independent variable groups and are tested on the dependent variable once.

What is a matched-groups design?

When participants who are similar on a measured variable (ex: IQ) are grouped into sets and the members of each matched set are then randomly assigned to different conditions.

What is conceptual replication?

When researchers use different procedures to study the same research question.

What is a testing threat?

When scores change over time because participants have taken the test more than once.

When is a matched-groups design the most ideal?

When the experiment has a small number of participants.

What is selection effect?

When the kinds of participants at one level of the IV are different from the kinds of participants at the other level of the IV

What is a main effect?

When the overall effect of one independent variable on the dependent variable is found by averaging over the levels of the other independent variable.

What is a Quasi-Experiment?

When the researchers do not have full experimental control for practical or ethical reasons

What is an insensitive measure?

When the researchers have not used an operationalization of the dependent variable with enough sensitivity (thermometer example from book).

What is an instrumentation threat?

When the scores change over time because some problem with the measurement technique.

What is a within-groups design?

When there is only one group of participants, and each participant is presented with all levels of the independent variable.

What is unsystematic variability?

When there is variability within the experiment that occurs RANDOMLY across all levels of the IV.

Do post-test only designs meet the three rules for causation?

Yes.

How do we establish Causality?

1) Covariance-A changes with B 2) Temporal Precedence- When the causal variable comes before the effect variable in time 3) Internal Validity- Ruling out alternative explanations

What are the 2 ways to prevent testing threats?

1) Do not give a pre-test 2) Provide different versions of the test

How to evaluate external validity when making causal claims?

1) Does the causal relationship generalize other people and other situations?

What are the 2 advantages of pretest/posttest?

1) Ensures no selection effect 2) You can truly measure change

What are the 3 advantages of within-subject repeated measures?

1) Equivalence 2) Greater statistical power 3) Fewer participants are needed

What are 2 solutions in regards to individual differences?

1) Change the design (within groups instead of independent groups) 2) Add more participants

What are the 2 types of within-groups designs?

1) Concurrent measures design 2) Repeated measures design

How to evaluate construct validity when making causal claims?

1) How well is the dependent variable measured? 2) How well was the independent variable manipulated?

What are the 6 threats to internal validity that apply to a one group, pre-test post-test design?

1) Maturation 2) History threats 3) Regression to the mean 4) Attrition 5) Testing Threat 6) Instrumentation Threat

What 3 things can occur when there is a high within-group variability (too much unsystematic variability)?

1) Measurement error 2) Individual differences 3) Situational noise

What are the 3 threats to internal validity that can apply to any experiment?

1) Observer bias 2) Demand Characteristics 3) Placebo Effect

What are the 2 disadvantages of within-subjects repeated measures?

1) Order effects 2) Demand characteristics-When an experimenter contains cues that lead participants to guess the hypothesis These both mean the participants being exposed to only one condition can change how they react to the other condition.

What 3 things might prevent an experiment from detecting a true difference that exists between experimental groups?

1) Weak manipulations 2) Insensitive measures 3) Confound

How to evaluate statistical validity when making causal claims?

1) What is the effect size? 2) What is the statistical significance?

When can selection effect occur?

1) When the experimenters let participants choose what level of the IV they want to be in 2) When the experimenters assign one type of person to one condition (ex: all males) and another type of person to another condition.

What is maturation?

A change in behavior that emerges more or less spontaneously over time (ex: a cold just got better over time).

How do we help prevent regression threats, history threats, and maturation?

A comparison group.

What is a systematic variability?

A confound- Basically when the the experimenters create a problem.

What is the solution to the three threats to internal validity that apply to any experiment?

A double-blind placebo control study.

What is a control group?

A level of an independent variable that is intended to represent a neutral condition. Ex: A sugar pill

What is a confound?

A threat to internal validity.

What is a repeated-measures design?

A type of within-groups design when participants are measured on a dependent variable more than once.

What is a control variable?

Any variable that the experimenter holds constant on purpose.

What kind of confound is selection effect?

Design confound.

What are manipulation checks?

Extra dependent variables added to the experiment to test whether the experimental manipulation worked.

(T/F) All experiments do not need a comparison group.

False- they do so the researchers can compare one condition to another.

What is an independent variable?

The variable that you manipulate.

What is a dependent variable?

The variable that you measure.

How to prevent instrumentation threats?

Use a posttest-only design.

What is an interaction?

Used to find whether the effect of the original independent variable depends on another independent variable.

What is a history threat?

When a historical or external event occurs to everyone in the treatment group at the same time as the treatment.

What is regression to the mean?

When a performance is extreme at time 1, the next time performance is measured (time 2) it is less likely to be extreme.

What is Direct replication?

When a researcher repeats the procedures of a study as closely as possible to see if they obtain the same results.

What is replication-plus-extension?

When a researcher replicates the OG study but adds another variable to test an additional research question.

What is a one group pre-test post-test design?

When a researcher uses one group of participants, measures behavior (pre-test), exposes them to treatment, then measures using a posttest.

What is a Design confound?

When a second variable just so happens to vary systematically along with the intended independent variable.

What is a null effect?

When a study finds that the independent variable didn't make a difference in the dependent variable.

What is a concurrent measures design?

When participants are exposed to all levels of an independent variable at the same time (ex: taste test between coke & pepsi).

What is a between-subjects pretest/posttest design?

When participants are randomly assigned to at least two groups and are tested on a key dependent variable twice- one before and once after exposure to the independent variable.