Psych 311 Unit 11 Study Guide

Helpfulness: 0
Set Details Share
created 4 years ago by wouhib
updated 4 years ago by wouhib
show moreless
Page to share:
Embed this setcancel
code changes based on your size selection

Quasi-Experimental Research

  • Purpose
    • Attempts to explain a relationship (cause-and effect), but always has limitations
  • Data
    • Measure before/after scores for the treatment group and a different group that does not receive treatment
  • Validity
    • Makes some attempts to minimize threats to internal validity

Nonexperimental Research

  • Purpose
    • To describe a relationship between two variables (but not to explain the relationship)
  • Data
    • Measure scores for two different groups of participants (or one group at two different times)
  • Validity
    • Does not typically attempt to minimize threats to internal validity


card image

Nonequivalent Group Design

  • A research study in which different groups of participants are formed under circumstances that do not permit the researcher to control assignment to groups.
  • Random assignment is not possible. Therefore, groups are nonequivalent.

Non-Equivalent Group Designs: Differential Research

  • Compares pre-existing groups
  • Uses only participant characteristics (gender, race, or personality) to automatically assign participants to groups (not random assignment)
  • Dependent variable measured for each participant to obtain group scores

Non-Equivalent Group Designs: Differential Research

  • Goal
    • To determine whether groups are consistently different from each other
  • Classification
    • Nonexperimental research design

Non-Equivalent Group Designs: Differential vs. Correlational

  • Differential
    • Observes two naturally occurring variables without manipulation
    • One variable is participant differences – used to create groups
    • Second variable measured within each group
    • Mean differences compared between groups
  • Correlational
    • Observes two naturally occurring variables without manipulation
    • Treats all participants as a single group
    • Measures two variables for each individual

Non-Equivalent Group Designs: Differential vs. Correlational Comparison

  • Similarities
    • Both allow researchers to establish the existence of relationships and to describe relationships between variables
    • Neither explains cause-and-effect relationships
    • Similar interpretations of data
  • Differences
    • Produce different kinds of data
    • Run different statistical analyses

Non-Equivalent Group Designs: Threat to Internal Validity

  • Assignment Bias
    • When assignment produces groups with different participant characteristics
    • For example, comparing two high schools (preexisting groups) with significantly different participant characteristics
    • Participant differences are potentially confounding variables (could explain group differences)

Non-Equivalent Control Group Designs: Posttest Only

card image
  • Compares two nonequivalent groups of participants
  • One group is observed after receiving treatment
  • Other is observed but receives no treatment
  • Considered non-experimental design

Non-Equivalent Control Group Designs: Pretest-Posttest

card image
  • Compares two nonequivalent groups of participants
  • One group is measured twice (once before, once after)
  • Other is measured twice but no treatment is given
  • Considered quasi-experimental

Non-Equivalent Control Group Designs: Threats to Internal Validity

  • Assignment Bias
    • An issue for this type of non-equivalent design as well
  • Differential Effects
    • Potential confounding variable due to groups having different histories
    • One group could be influenced by different outside events
    • Other differences could be caused maturation, instrumentation, testing effects, and regression

Within-Subjects Pre-Post Designs

  • Definition
    • One group of participants is observed (measured) before or after a treatment or event
  • No control group
    • Goal is to evaluate treatment influence

Within-Subjects Pre-Post Designs: One-Group Pretest-Posttest

  • Each individual in a single group of participants is measured once before treatment and once after treatment\
    • O X O
  • Classified as nonexperimental design

Within-Subjects Pre-Post Designs: Similarities & Differences

  • Within-Subjects Pre-Post
    • Involves measurement before and after
    • No control group
    • Purpose – to compare pre-test scores with post-test scores
  • Pre-Test Post-Test Nonequivalent Control
    • Involves measurement before and after
    • Control group
    • Purpose – to compare groups
    • Uses counterbalancing to control time effects

Within-Subjects Pre-Post Designs: Threats to Internal Validity

  • Time Effects
    • History
    • Instrumentation
    • Testing effects
    • Maturation
    • Statistical regression
  • Addressing Threats
    • Some types of Pre-Post address threats, while others do not

Within-Subjects Pre-Post Designs: Time Series Design

  • Series of observations for each participant before a treatment or event and a series of observations after the treatment or event
  • Pretests & posttests reveal trends occurring before & after treatment
  • Series of observations helps to minimize threats to validity
  • Considered quasi-experimental design
  • O O O X O O O

Within-Subjects Pre-Post Designs: Time Series Designs (Terms)

  • Treatment
    • Manipulated by the researcher
  • Event
    • An outside occurrence that is not manipulated by the researcher (sometimes called interrupted-time-series)

Within-Subjects Pre-Post Designs: Time Series Designs

  • Examples
    • Examining changes in legal drinking age or speed limits (before/after)
  • Benefits
    • Can assess impact of outside events (except those occurring concurrently with treatment)

Developmental Research Designs

  • Definition
    • Used to examine changes in behavior related to age
    • Classified as nonexperimental design

Developmental Research Designs: Cross-Sectional

card image
  • Uses different groups of individuals, each group representing a different age
  • Different groups are measured at one point in time and compared
  • Classification
    • Considered between-subjects nonexperimental design
    • Also considered nonequivalent group design (groups created by an existing variable – age)

Developmental Research Designs: Cross-Sectional

  • Strengths
    • Can observe how behavior changes as people age without waiting for them to age
    • Time and cost savings
  • Weaknesses
    • Cannot speak to an individual’s development over time
    • Factors other than age may differentiate the groups (cohort effects or generation effects)
    • Cohorts – individuals who grew up at roughly the same time and under similar circumstances

Developmental Research Designs: Longitudinal

card image
  • Examines development by observing or measuring a group of cohorts over time
  • No treatment administered
  • Age is the “treatment” (observations followed by developmental aging)
  • Classification
    • Considered within-subjects non-experimental design (one-group prettest-posttest)

Developmental Research Designs: Longitudinal Strengths and Weaknesses

  • Strengths
    • Absence of cohort effects
    • Can discuss how an individual changes with age
  • Weaknesses
    • Time consuming and expensive
    • May need to train additional researchers
    • Participant attrition (aka participant mortality): when participants lose interest, move away, or die
    • Remaining participants may make sample biased
    • Potential testing effects over time

Developmental Research Designs: Cross-Sectional vs. Longitudinal

card image

Quasi-Independent Variable

  • Variable used to differentiate groups of participants or the groups of scores being compared in nonexperimental and quasi-experimental research