Variables, Designs & Hypotheses

• What are the differences between experimental, quasi-experimental, and correlational designs?

  • Experimental: IV manipulated, causality inferred
  • Quasi-experimental: IV not manipulated, harder to eliminate confounding variables
  • Correlational: No manipulation, measures relationships between variables
• What types of variables and measurement scales are identified in the study material?

  Different types of variables and measurement scales
• What types of hypotheses are explained in the study material?

  Research, statistical, null, and alternative hypotheses
• What are the steps in conducting a research study?

  1. Research question
  2. Theory
  3. Generate hypotheses
  4. Design a study
  5. Ethical approval
  6. Collect data
  7. Analyze data
  8. Write up as a research report
• What does the independent variable (IV) represent in experimental methods?

  Something that we manipulate
• What does the dependent variable (DV) represent in experimental methods?

  What we are recording or measuring
• What can we infer when changes in the DV are due to changes in the IV?

  We can infer causality
• What characterizes quasi-experimental methods?

  • The IV cannot be manipulated
  • Examples include non-equivalent groups and pretest-posttest designs
  • Harder to eliminate confounding variables
• What are the key features of correlational methods?

  • No manipulations are made
  • Measures two or more variables
  • Determines the extent to which they are related
  • Cannot infer causality
• How many independent variables can an experiment have?

  An experiment can have more than one IV or factors
• What should each independent variable have in an experiment?

  Two or more levels
• In the Stroop task example, what is the independent variable?

  Congruency of word/colour stimuli
• How many dependent variables can an experiment have?

  An experiment can have one or more DVs
• What is important to specify when measuring the dependent variable?

  How we measure our DV (operationalization)
• What is a confounding variable?

  A variable we don’t manipulate that may influence results
• What are the characteristics of between-subjects designs?

  • Participants only take part in one level of the IV
  • Random assignment can account for individual differences
  • Less powerful, requiring more participants to detect a genuine effect
• What are the characteristics of within-subjects designs?

  • The same participant performs all levels of the IV
  • Known as repeated measures design
  • More powerful, requiring fewer participants to detect a genuine effect
• What are order effects in within-subjects designs?

  • Results may be influenced by the order of conditions
  • Can be due to practice or boredom
  • Best addressed through randomization of trials and counterbalancing
• What is a matched subjects design?

  • Used when a within-subjects design is not feasible
  • Participants are matched based on demographic characteristics
  • The pair is tested as one individual over two levels of an IV
• What characterizes correlational designs?

  • Variables cannot be manipulated
  • Examines existing variables to see how they co-vary
  • Does not imply causation
• What is a hypothesis?

  A theory-driven idea explaining a narrow set of phenomena
• What are the different types of hypotheses mentioned?

  • Experimental hypothesis
  • Statistical hypothesis
  • Null hypothesis (H0)
  • Alternative hypothesis (H1)
• What is the difference between experimental and statistical hypotheses?

  • Experimental hypothesis: Conceptual idea explaining an observation
  • Statistical hypothesis: Specific statement used to collect data and test the hypothesis
• What does the null hypothesis (H0) imply?

  Observations from samples imply they come from the same population
• What does the null hypothesis state for parametric statistics?

  All means are equal: H0: µ1 = µ2
• What is the prediction made by the null hypothesis in the Stroop task example?

  Mean reaction times will be the same for congruent and incongruent stimuli

  
• What does the alternative hypothesis (H1) predict?

  There will be a significant difference or relationship between variables
• What is a non-directional hypothesis?

  There will be a difference in mean reaction times for congruent and incongruent stimuli

  
• What is a directional hypothesis?

  Mean reaction times will be shorter for congruent stimuli than for incongruent stimuli

  
• What are the properties of null and alternative hypotheses?

  • Mutually exclusive: only one can be true
  • Exhaustive: covers all possible outcomes
  • In the Stroop task, means can either be the same (H0) or different (H1)
• What is null hypothesis significance testing?

  • Reject the null hypothesis when the probability of it being true (p) is lower than a specific criterion (α)
  • Involves generating a test statistic and setting a specific (α) criterion
  • Helps determine the probability value (p-value)
• What is the typical criterion (α) used for rejecting the null hypothesis?

  Usually set at .05