Quantitative Research Designs

Quantitative Research Designs

• The presentation discusses various quantitative research designs.
• Learning objectives include identifying different designs, explaining their characteristics, and discussing types.
• Experimental research has specific characteristics: comparison of groups (experimental and control groups).
• The experimental group receives a treatment, while the control group does not (or receives a different treatment).
• Manipulation of the independent variable can take different forms such as comparison of different methods, presence versus absence of a factor, or varying degrees of the same treatment.

Experimental Research: Comparison of Groups

• Experimental group receives a treatment or intervention.
• Control group receives no treatment or a different treatment.
• Comparison group serves as a benchmark for comparisons.

Experimental Research: Manipulation of the Independent Variable

• Independent variable manipulation can be established in different ways.
• This could involve comparing one form of variable to another (e.g., inquiry vs lecture methods).
• Or, comparing the presence or absence of a particular factor (e.g., use of slides in teaching).
• Also, varying the degree of a variable (e.g, comparing different amounts of teacher enthusiasm).

Experimental Research: Randomization

• Random assignment ensures every participant has an equal chance of being assigned to either condition to minimize bias.
• Randomization happens before the experiment begins, not as a result of the experiment conditions.
• Random assignment creates equivalent groups with differences due to chance only.
• Random assignment doesn't guarantee perfectly equivalent groups, large sample size is needed.
• Random selection involves selecting participants from a population fairly. Each member of the population has an equal chance of being selected into the sample.

Experimental Designs - Poor Experimental Designs

• One-shot case study: Treatment, then observation (X O).
• One-group pretest-posttest design: Pretest, treatment, posttest (O X O).
• Static-group comparison design: Treatment group (X O); control group (O).
• Static-group pretest-posttest design: Pretest (O), treatment (X), posttest(O) in one group; pretest(O) and posttest (O) for the other

Experimental Designs - True Experimental Designs

• Randomized posttest-only control group design: Treatment group (R X O); control group (R C O). This involves random assignment, and only observations after the treatment.
• Randomized pretest-posttest control group design: Treatment group (R O X O); control group (R O C O). This design has both pretests and posttests.
• Random Solomon four-group design: Treatment group 1 (R O X O), control group 1 (R O C O), treatment group 2 (R X O), control group 2 (R C O)

Quasi-Experimental Research

• Avoids random assignment, uses naturally occurring groups.
• Examples involve assigning naturally occurring groups, such as two classes.
• Natural manipulation of variables (e.g. examining the effect of an event, or comparing groups with varying attributes, like age or ethnicity).
• Some variables can't be manipulated in this research type due to ethical considerations.

Quasi-Experimental Research: Naturally Occurring Independent Variables

• Inherent subject variables (age, sex, race)
• Socially-caused subject attributes (social class, residence region).
• Disease and illness subject attributes (limb loss, mental illness).
• Environmental conditions (natural disasters).
• Other naturally occurring parameters such as the occurrence of a particular event (e.g., strikes in factories).

Correlational Research

• Investigates relationships among two or more variables without manipulating them.
• Doesn't determine causality, only explores associations.
• Purpose to clarify understanding of important phenomena by identifying relationships among variables.

Causal-Comparative Research

• Aims to find the cause or outcomes of existing differences between groups.
• It is associational research, and does not attempt to influence the participants.
• Explores different aspects of causal comparative studies (e.g., effects and causes).

Survey Research

• Collects information from a group of people to describe a population.
• Information is gathered through asking questions on a sample, not the entire population.

Survey Research: Types

• Cross-sectional survey: collects information from a sample at one point in time.
• Longitudinal survey: collects data at various points to examine changes over time (trend, cohort, panel designs).
• Trend study: uses different samples (new people each time) in a population that may change over time.
• Cohort study: uses different samples of a specific group at different points in time that doesn’t change, (e.g., all 1st year teachers from various years).
• Panel study: uses the same sample of individuals at multiple points in time during the study.

References

• Fraenkel, J. R., Wallen, N.E., & Hyun, H.H. (2015). How to design and evaluate research in education (9th ed.). McGrawHill.