Advanced Research Methods- Chapter 10: Nonexperimental and Quasi-experimental Strategies

Nonexperimental Design

Unable to fully control all confounds and treats to IV.

Quasi-experimental Design

Makes attempts, but full control of confounds is not guaranteed.

Structure of Nonexp. and Quasi

Similar to true experiment (compares scores from different conditions). Independent variable is NOT manipulated.

Nonequivalent group design

Different groups of participants are formed under circumstances that do not permit the researcher to control the assignment. Cannot use random assignment or matching. Threats to IV: Individual differences, no assurance of equal groups.

Differential research design

A nonexp. design with nonequivalent groups. Compares preexisting groups of specific participant variables. Individual differences are the primary interest. a.k.a ex post facto design.

Post test only nonequivalent control group design

A nonexp. design with nonequal. groups. Preexisting groups. Treatment and control group. No random assignment. Only one observation per group.

Pretest-Posttest nonequivalent control group design

A quasi-ex. design with nonequivalent groups. Compares 2 non equal. groups. A pretest and posttest per group. Looking at change between observations.

Nonequivalent control group designs

Ex: Different factory schedules.

Pre-post design

Series of observations are made over time before and after treatment. No control group. Threats to IV: history, instrumentation, maturation, and statistical regression.

Pretest-posttest design

A nonexp. pre-post design. Simplest version of a pre-post design with one pretest and one posttest. No IV control.

Time series design

A quasi exp. pre-post design. A series of observations before and after treatment. Interrupted time series: interviewing event is not manipulated by the researcher. Trends!

Cross-sectional developmental research design

A between subjects nonexp. nonequal. group design. Different groups. Groups measured at one point in time and then compared. Strengths: Quick, cheap. Weaknesses: outside factors, cohort effects.

Longitudinal developmental research design

A within subject nonexp. design ( a one-group pretest-posttest design). Same group measured repeatedly over time. Differences in observations are compared. Strengths: No cohort effects, individual changes. Weaknesses: Long, participant attrition, practice effects.

Cross sectional longitudinal design

Compares results from separate samples that were obtained at different times.