Quasi

• Quasi-experimental Design

  A researcher manipulates an independent variable but does not randomly assign participants to conditions
• Types of Quasi-Experimental


  • Nonequivalent group design
  • Time series design
  • Interrupted time series design
  • Single case time series design
• Nonequivalent group design


  • Different groups of participants are formed under circumstances that do not permit the researcher to control the assignment of individuals to groups
  • The researcher cannot use random assignment to create groups of participants
• Nonequivalent Group Study


  The general structure of a nonequivalent group study
• Types of Between Treatment Designs

  • Experimental
  • Quasi-Experimental
  • Non-experimental
• Experimental


  • Randomized
  • Manipulation
  • Comparative group
  • Control of extraneous variables
• Quasi-Experimental


  • Not randomized
  • Manipulation
  • Comparative group
  • Control of extraneous variables
• Non-experimental


  • Not randomized
  • No manipulation
  • Comparative group
  • Control of extraneous variables
• Between-subject variables


  Independent variables or factors in which a different group of subjects is used for each level of the variable
• Assignment bias precludes a clear cause-and-effect explanation in nonequivalent group designs
• Groups have different participant characteristics in nonequivalent group designs
• There is no random assignment to groups in nonequivalent group designs, so there is no assurance of equivalent groups
• Within-Subject Quasi-Experimental Design: Pre-Post Design


  A research study in which a series of observations is made over time for one group of participants
• Types of Within Treatment Designs

  • Quasi-Experimental
  • Non-experimental
• Pretest Post Test Group Design

  The dependent variable is measured once before the treatment is implemented and once after it is implemented
• Pretest Post Test Group Design


  • Effectiveness of an antidrug education program on elementary school students' attitudes toward illegal drugs
• Time-Series Design


  A modification of the pre-posttest design, with a stronger design involving a series of observations (not single for before and after)
• Interrupted Time Series


  If the event is not manipulated by the researcher, it can be called an interrupted time-series (e.g. legalization of marijuana)
• Time Series Design


  A series of observations for each participant before a treatment and a series of observations after the treatment, with the treatment manipulated by the researcher
• Discussion


  1. Regression to the mean

  2. Spontaneous remission

  3. History

  4. Maturation
• Threats to Internal Validity

  • History
  • Instrumentation
  • Order effects
  • Carryover effect
  • Practice effect
  • Fatigue effect
  • Context effect
• Threats to Internal Validity (cont.)

  • Maturation
  • Statistical regression or regression toward the mean
• Counterbalancing


  Controls for order effects by testing different participants in different orders (ABC, ACB)
• Combination Designs


  A treatment group that is given a pretest, receives a treatment, and then is given a posttest, compared to a control group given a pretest, does not receive the treatment, and then is given a posttest
• Combination Designs


  • Students in one school are given a pretest on their attitudes toward drugs, then are exposed to an antidrug program, and then given a posttest, compared to students in a similar school who are given the pretest, not exposed to an antidrug program, and then given a posttest
• Combination designs control for history (celebrity overdose) or maturation (improved reasoning), but still confound other factors
• Let's Practice


  Two professors decide to test the effect of giving daily quizzes on student performance, with Professor A giving quizzes and Professor B not giving quizzes, then comparing the performance of students in their two sections on a common final exam
• Confounds to this study

  • List five confounds
• Caffeine Makes people smarter

  • Group one: experimental
  • Group two: quasi-experimental
  • Group three: non-experimental
• Caffeine activity


  • Experimental: Can individuals manipulate a maze faster than individuals without caffeine
  • Quasi-experimental: Same
  • Non-experimental: Is there a difference between grades of people who use caffein or those who do not use
• Caffeine activity


  • Experimental: Randomly assign people
  • Quasi-experimental: Assign coffee group vs those who do not drink coffee
  • Non-experimental: Placebo group (caffeine free), no coffee and caffeinated coffee
• Factor


  An independent variable in an experiment, especially those that include two or more IVs
• Factorial design


  A research design that includes two or more factors
• Factorial Designs


  • A two-factor design has two IVs
  • A single-factor design has one IV
  • A 2 x 3 x 2 design is a three-factor design with a total of 12 conditions
• Experimental Factorial Designs


  • Two or more IVs (factors) are manipulated
  • May involve quasi-independent variables (e.g., age, gender) that are not manipulated
  • Creates a more realistic situation than examining a single factor in isolation
  • Can see how individual factors as well as the group of factors, acting together, influence behaviour
• Structure of a Two-Factor Experiment

  The levels of one factor determine the columns and the levels of the second factor determine the rows
• Main Effects


  The mean differences among the levels of one factor
• Interaction Between Factors


  Direct influence (one factor influences the other)
• Interaction Between Factors


  • Drug interaction: one drug modifying the effect of another drug
  • One drug can exaggerate the effects of another
  • One drug may minimize or completely block the effects of another
• Identifying Interactions


  Graphing the results of a two-factor study: Nonparallel lines (lines that cross or converge) indicate an interaction between the two factors