Chapter 12 The Nuts and Bolts of QuasiExperiments

Chapter 12 The Nuts and Bolts of Quasi-Experiments Dawn M. Mc. Bride The Process of Research in Psychology Second Edition

Steps in the Research Process 2 Dawn M. Mc. Bride - The Process of

Quasi-Experiments l 3 A type of research design where a comparison is made, as in an experiment, but no random assignment of participants to groups occurs § Pretest-Posttest designs § Time series designs Dawn M. Mc. Bride - The Process of Research in Psychology, 2 nd Edition © 2013 SAGE Publications, Inc.

Quasi-Experiments l 4 Behaviors fore which random assignment would be difficult, impossible or unethical. § Field settings: Schools, mental health agencies § Parenting behaviors § Quality of Therapists Dawn M. Mc. Bride - The Process of Research in Psychology, 2 nd Edition © 2013 SAGE Publications, Inc.

Pretest-Posttest Designs l No control group: compare post-test with pre-test scores l May include control group, without random assignment: non-equivalent groups § l 5 Implement a new intervention with one classroom of 8 th graders and use another classroom as control If random assignment is used– then study is an experiment (pre-test may be used or not) Dawn M. Mc. Bride - The Process of Research in Psychology, 2 nd Edition © 2013 SAGE Publications, Inc.

Pretest-Posttest Designs 6 Dawn M. Mc. Bride - The Process of Research in Psychology,

7 Dawn M. Mc. Bride - The Process of Research in Psychology, 2 nd

Sources of bias: l History effects§ events that occur during the course of a study to all or individual participants that can result in bias. Ø Ø l 8 9 -11 Hurricane Regression towards the mean Dawn M. Mc. Bride - The Process of Research in Psychology, 2 nd Edition © 2013 SAGE Publications, Inc.

Pretest-Posttest Designs: Bias l Testing effects l Possible solution § Solomon four-group design (Shadish et al. , 2002). Ø Ø 9 one set takes the pretest and posttest one set that takes only the posttest Dawn M. Mc. Bride - The Process of Research in Psychology, 2 nd Edition © 2013 SAGE Publications, Inc.

Pretest-Posttest Designs 10 Dawn M. Mc. Bride - The Process of Research in Psychology,

Time Series Designs 11 l Measuring the behavior or attitude multiple times before and event or ttreatment and multiple times after over the same time period. l The patterns of scores over time before and after the event or treatment are compared to determine if a difference has occurred. Dawn M. Mc. Bride - The Process of Research in Psychology, 2 nd Edition © 2013 SAGE Publications, Inc.

Time Series Designs: Interrupted l l The “treatment” is an independent event, such as a historical event. § Iraq war Patterns of scores are compared over time: before and after the event occurs § l 12 Typically uses archival data: e. g. Consumer confidence Index; Census Data Requires sophisticated data analyses procedures Dawn M. Mc. Bride - The Process of Research in Psychology, 2 nd Edition © 2013 SAGE Publications, Inc.

FDA warning Oct 2013 antidepressants and suicide risk 13 Dawn M. Mc. Bride -

Noninterrupted Time Series Designs l The “treatment” is implemented by the researcher l The participants’ depression symptoms would be measured over a period of time § § 14 before they began the new therapy after therapy concludes Dawn M. Mc. Bride - The Process of Research in Psychology, 2 nd Edition © 2013 SAGE Publications, Inc.

Noninterrupted Time Series Design 15 Dawn M. Mc. Bride - The Process of Research

Times Series D. Sources of Bias l History § recession, natural events l Maturation § changes that occur to the participants during the course of a study that can result in bias Ø Developmental changes l Attrition 16 Dawn M. Mc. Bride - The Process of Research in Psychology, 2 nd Edition © 2013 SAGE Publications, Inc.