Chapter 10

• Causal claim - boldest claim you can make

  (related to, is associated with, linked to, makes, influences or affects)
• Experiment - researchers manipulated at least one variable + measure another
• Manipulated variable - variable thats been controlled when researchers assign a participant to a particular variable
• measured variables - records of behaviour or attitudes

  (self reports, behavioural observations or psychological measures)
• independent variable - manipulated causal variable
• dependent variable - measured variable
• control variable - any variable held constant
• 3 rules for causation

  [1] covariance - do results show a causal variable is related to the outcome variable? are distinct levels of the IV associated with different levels of the DV
• 3 rules for causation

  [2] Temporal precedence - Does the study design ensure that the causal variables comes before the outcome variable in time
• 3 rules for causation

  [3] internal validity - does the study design rule out alternative explanations for the results
• comparison group - group whose levels of IV differ from those in treatment group in some intended + meaningful way manipulating the IV is necessary for establishing co-variance
• control group - level of IV intended to represent no treatment (neutral condition)
• placebo group - control group given inert treatment
• confounds - potential threats to internal validity
• design confound - experimenters mistake in designing the IV

  second variable varies systematically with the intended IV
• unsystematic variability - levels of variable fluctuates independently of experimental group (random) - can make it difficult to detect differences in IV
• selection effects - kinds of participants at one level of the IV are systematically different to those at the other level

  • can happen when experimenters let participants choose what group they want to be in
  • assign 1 type of participant to each condition
• selection effects avoided with random assignment
• independent groups design - separate groups of participants are placed into different levels of the IV
• within subjects - each person presented with all levels of IV
• posttest only design - participants randomly assigned to IV + tested on DV. satisfies all 3 criteria for causation
• Pretest/posttest design - participants randomly assigned to at least 2 groups + tested on DV twice. enables researchers to track performance over time
• repeated-measures design - participants measured on DV more than once
• concurrent measures design - participants exposed to all levels of an IV at the same time + a single attitudinal/behavioural preference is the DV
• Advantages of within groups

  • ensures participants in the 2 groups will be equivalent
  • matched groups can be treated as within groups
  • make more precise estimates of the differences between conditions
  • requires less participants
• Repeat measures - establishes temporal precedence
• order effects - being exposed to 1 condition first changes how participants react in later conditions
• order effects are a confound
• practice/fatigue effects - long sequence events allow participants to get better at the task/get tired/bored
• carryover effects - contamination carries over from one condition to the next
• counterbalancing - presenting levels of IV to participants in different sequences
• full counterbalancing - all possible condition orders are represented
• partial counterbalancing - only some of the possible conditions represented - present conditions in a randomised order for every subject - use a latin square
• disadvantage of within groups

  • potential for order effects
  • may not be possible/practical
  • demand characteristics
• demand characteristics - cue that leads participants to guess an experiments hypothesis
• manipulation check - extra dependent variable that researchers can insert into an experiment to convince them that their experimental manipulation worked
• pilot study - separate group complete the study before/after primary interest study to confirm the effectiveness