Experimental Psych- SPSS Analysis

• To determine the research question and hypotheses, one must identify the variables in the study first
• The research question should be constructed in a way that would reflect causal inferences; Hence, using the terms-- leads to, results in, cause, affect, effect, etc.
• The variables should be operationally defined to make them measurable
• The hypothesis should be stated in an IF-Then statement
• Between-groups designs


  • Posttest-only control group design
  • Pretest-posttest control group design
  • Solomon-four group design
• Within-subject designs


  • Crossover design
  • Latin-square design
• Factorial designs: more than 1 IV
• The purpose of the present investigation is to study the effects of teacher's verbal reinforcement on pupils' classroom demeanor
• Conceptual RQ


  Research question that reflects causal inferences
• Statistical RQ


  Research question that can be tested statistically
• If-Then Hypothesis


  Hypothesis stated in an if-then format
• Statistical Hypotheses


  Hypotheses that can be tested statistically (e.g. null and alternative hypotheses)
• Directional/Non-directional


  Hypotheses that specify the direction of the effect (directional) or do not specify the direction (non-directional)
• Variables


  • Independent variable (IV) and levels
  • Dependent variable (DV)
• Confounding, Extraneous, Control variables should also be identified
• Operational definition of the variables and their level of measurement should be provided
• Research Designs


  • Between-group
  • Within-group
  • Factorial
• Potential threats to validity and ways to address them should be considered
• Data collection should generally be based on the research design and definitions of variables, but the researcher is free to have their own strategy
• Data analysis should be conducted using appropriate statistical tests
• Results, discussion, and conclusion should be reported
• The choice of statistical test depends on the experimental design used
• Between-groups designs


  • Posttest-only control group design
  • Pretest-posttest control group design
  • Solomon-four group design
• Within-subject designs


  • Crossover design
  • Latin-square design
• Tests for group differences

  • Independent-samples t-test
  • Paired-samples t-test
  • ANOVA (Analysis of Variance)
  • Factorial ANOVA
• If the experiment uses a between-group design, participants will be randomly assigned to two groups: an experimental group that receives verbal reinforcement, and a control group that does not receive verbal reinforcement
• The demeanor (number of negative behaviors) of the pupils in the two groups will be compared to determine if there is a significant difference
• The data can be encoded in Excel or SPSS, with the Reinforcement variable coded as 1 for with reinforcement and 2 for without reinforcement, and the Demeanor variable coded as the number of negative behaviors
• To determine if there is a significant difference, the p-value (Sig.) should be less than 0.05 to reject the null hypothesis and accept the alternative hypothesis
• If there is a significant difference, the interpretation would be: "There is a significant difference in the demeanor of pupils in class with verbal reinforcement (M=15.7, SD=2.7) and without verbal reinforcement (M=19.8, SD=3.3); t(54)=-4.96, p=0.000."