Research Methods (10)

• The difference between the two designs is the degree to which the research strategy limits:

  • Confounding Variables

  • Controls threats to internal validity

• Nonexperimental - little to no attempt to minimize threats

• Quasi-Experimental - some attempt to minimize threats to internal validity and is almost, but not quite, a true experiment

• Both often look like experiments

  • Studies produce groups of scores that can be compared for significant differences

  • However, the different scores are not produced by manipulating an IV, but instead the groups are defined in terms of a participant variable (i.e. gender)

What is difference between a Non-Experimental and a Quasi-Experimental Design? (longe)

10.2 - Non-Equivalent Group Designs

module done…

Cannot use random assignment because the groups already exist

Why can’t researchers have control over the assignment of individuals in a group?

Yes.

• Individual differences prevent a clear cause-and-effect explanation

  • Groups have different participant characteristics

    • No random assignment = No assurance of equivalent groups

Are there any threats to Internal Validity?

• Compared pre-existing groups

• Goal: establish differences between the pre-existing groups

• Also called an ex-post facto design, looks at the differences “after the fact”

What is a Differential Research Design?

• One group is the treatment, second group (of similar, but nonequivalent participants) used for control group

• No random assignment

  • Does participation in a mindfulness-based stress reduction program lead to reduced levels of anxiety compared to the control group?

What is a Nonequivalent Control Group Design (post-test only)?

• Compared two nonequivalent groups

• One group is measured twice

  • Once before treatment and one after

• Second group is measured at the same two times, no treatment

• Design attempts to limit threats to internal validity, thus classified as a Quasi-Experimental Design

What is a Quasi-Experimental Design? (pre test - post test nonequivalent)?

10.3 - Pre-Post Designs

module doen

One group is measured before and after a treatment

What are Pre-Post Designs?

• History

• Instrumentation

• Testing Effects

• Maturation

• Statistical regression

What are the threats to the internal validity?

• Series of observations are made over time for one group of participants

• No attempt to control threats to internal validity

What is a Pretest-Posttest Design?

• Series of observations for each participant before a treatment or event, and a series of observations after a treatment or event

• A study in intervening events are not manipulated is sometimes called an interrupted time-series design

• The value of pretest and posttest observations help:

  • Allow researchers to see trends in data

    • Identifying practice, fatigue, maturation, etc., since they make noticeable changes before a treatment

  • Allows researcher to observe post-treatment trends

What is a Time-Series Design? (long?)

• Often applied to single individuals or single organizations

  • More often classified as single-case or single-subject designs

How can Time-Series Designs be applied in a Single-Case?

10.4 - Developmental Research Designs

MODULE DONE

• Developmental research designs are used to examine changes in behavior related to age

• Uses different groups of individuals

  • Each group has a different age

  • Different groups measured at one point in time and then compared

  • This design is an example of a between-subjects non-experimental design

• Pros:

  • Data can be collected in a short period of time

    • Observing how behavior changes as people age without waiting for them to grow older

• Cons:

  • Factors other than age may differentiate groups

    • Such as cohort effects (which generation were they born in?)

What is a Cross-Sectional Developmental Research Design? Pros and Cons?

• Examines development by observing or measuring a group of cohorts over time

  • EX) Measuring IQ in a group of 40 year olds, then measuring the same individuals again in ages 60 and 80

• An example of a within-subjects nonexperimental design

• Pros:

  • No cohort or generation effects

  • Assesses individual behavioral changes

• Cons:

  • Time-consuming

  • Participant attrition may create bias

  • Potential for practice effects

What is a Longitudinal Developmental Research Design? Pros and Cons?

10.5 - Applications, Statistical Analysis, and Terminology

Module Done

• Quasi-Independent Variable (IV)

  • Variable used to differentiate the groups of participants or group of scores being compared

• Dependent Variable (DV)

  • Variable that is measured to obtain the scores within each group

Review: What is a Quasi-Independent Variable (IV) and Dependent Variable (DV)?

• Two group designs have the advantage of simplicity

  • Easy to set up

  • Results are easy to understand

• Two-group design does not provide the full functional relationship between variables

• Two-group designs analysis

  • Compares numerical scores with a t-test for two means, or a single-factor analysis of variance for multiple means

  • Uses a chi-square test for non-numerical data

What is the Application and Analysis?