Experiments & Causality

experiment

• a type of study designed specifically to answer the question of whether there is a causal relationship between two variables

  • AKA randomized controlled trials (RCTs)

  • Features of experiments:

    • The researchers manipulate the level of the independent variable (conditions)

    • The researchers exert control over variables other than the independent and dependent variables

  • Ex: Hypothesis: People with bigger bowls will eat more pasta than people with smaller bowls

• Control variables: any var

control variables

• any variable that an experimenter holds “constant” between groups

• Control group: group that does not get the IV

  • Serves as a source of comparison

random assignment

• Accomplished by ensuring that every person has an equal chance of being assigned to any group

• Does not eliminate the influence of extraneous variables, but randomly distributes them among groups

• The most important and powerful control technique

• Different from SRS; this is a method for assigning participants in a sample to different conditions

types of experiments

• one-way design

• between-subjects designs

• within-subjects designs

one-way design

• an experimental design in which only one independent variable is manipulated

  • The simplest one-way design is the two-group experimental design

  • An experiment requires at least two levels of the IV to compare the effects of one level on the DV relative to the other level

between-subjects designs

different people are compared to each other

• each participant is only ONE experimental condition

• posttest only design

• pretest-posttest design

posttest only design

• Participants undergo the manipulation (just one condition) and then complete measures (once)

  • Participants are randomly assigned to groups, IV introduced, and DV measured

pretest-posttest design

• Participants complete the measures, then the manipulation, then the measures again

  • Participants are randomly assigned, the DV is measured, the IV is introduced, and the DV is measured again

advantages of the pretest-posttest design

• Rules out selection bias (occurs when random assignment fails)

  • We can use pretest scores to ensure that the two groups are the same

• Using change scores is more powerful because it eliminates noise

  • Accounting for possible pre-existing differences within participants

within subjects designs

• the same people are compared to themselves

  • Each participant is in ALL conditions

• benefits:

  • Each participant serves as their own control

  • No possibility of selection effect

  • Noise reduced

  • Doubles your sample size

• drawbacks:

  • order effects

    • fatigue effects

    • oractice effects

    • carryover effects

fatigue effects

participants get worse at the measures (boredom, distraction, etc)

practice effects

participants get better at the measures (learning effect, etc)

carryover effects

effects of one condition contaminate subsequent responses

how to fix order effects

• counterbalanced design or counterbalancing

• informed consent

counterbalanced design or counterbalancing

• In a within-subject design, presenting the levels of the IV to participants in different sequences to control for order effects

• Experiment: set the stage→manipulate the IV→measure the DV→debrief participants

informed consent

• Discuss the purpose of the study

• Very general procedure, including the amount of time

• Risks and benefits of research

• Compensation

• How their identity and privacy will be protected

• Voluntary participation and permission to withdraw

• Contact information for questions

demand characteristics

features of the study that influence participants to become aware of (or think that they know) the purpose of the study

• These can lead to changes in behaviors or responses (reactivity)

how can we avoid demand characteristics

• Be general about what will happen and what participants will do, but avoid explaining the true, specific purpose

  • Say “ You will think about aspects of your life in the future and how they make you feel”

  • NOT “You will be primed with self-uncertainty about your future to see how that affects your satisfaction with life.

• Make it difficult to figure out what the study is about, or make participants think that it’s about something else

  • Use deception

  • Add fuller tasksand/or items

    • Measures that won't actually be analyzed but are used to mask the true purpose of the study

• Ask them about their thoughts during debriefing

  • “What did you think was the true purpose of this study?”

    • If their guess is spot on, you may need to remove the participant from analyses

straightforward manipulations of the IV

manipulate with written instructions, verbal materials, and/or visual materials

staged manipulations of the IV

• manipulate an entire setting; event manipulations

  • Often uses confederates

    • A person who appears to be another participant (or a random person), but who is part of the experiment

manipulation strength and type

• Selecting a manipulation should be based on: 

  • Ethical considerations (ex: how sad can I make participants)

  • Realism of the experiment → does this manipulation occur similarly out in the world?

    • Concerns about external validity

  • Money and resources

manipulation checks

• Attempts to directly measure (check the effectiveness of a manipulation)

  • Ex: my positive emotion ad actually elicited positive feelings

• If the manipulation is not valid or effective, you need to enhance or change your IV

• Measure of construct validity

floor effects

all low scores on a measure

Ceiling effect

all high scores on a measure

debriefing participants

• Explain the true purpose of the study

• Mitigate any negative consequences of manipulation

• Keep the secret

  • Necessary when deception (staged manipulation) is involved

  • Dont tell your peers who may be in the study

  • Wait until study is over