Research in Psychology An Introduction
Research in Psychology An Introduction PSYCH 300
Popular in Course
Popular in Psychlogy
This 4 page Class Notes was uploaded by Adeline Williamson on Monday September 21, 2015. The Class Notes belongs to PSYCH 300 at Ohio State University taught by Kristy Boyce in Fall. Since its upload, it has received 24 views. For similar materials see /class/209984/psych-300-ohio-state-university in Psychlogy at Ohio State University.
Reviews for Research in Psychology An Introduction
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/21/15
Chapter 9 QuasiExperimental Designs 0 QuasiExperimental Designs 0 Applied research experiments conducted in natural settings to improve the conditions in which people live and work Ex research by government on new school funding research on the effectiveness of daycare programs 0 The high degree of control that is needed to make causeandeffect conclusions is not possible in real world so Quasiexperiments they are like true experiments but they lack the degree of control found in true experiments 0 Quasi vs true experiments True experiment Quasiexperiment Intervention ie treatment is yes yes im lemented Experimenter includes yes yes a com arison Experimenter has a high degree of control over the experimental yes N0 situation ex random assi nment o QuasiExperiments o Obstacles to conducting true experiments obtaining permission to do research 0 Ex trying to conduct research in school districts businesses government offices prisons ect Access to participants 0 Need to use random assignment but people in natural settings don t want to be randomly assigned 0 many times people want to be in experimental treatment group not control group 0 Can use a waiting list as control groupcomplete dependent variable at same time as experimental group but receive treatment IV after research is complete 0 waiting list is not the same thing as random assignment 0 Threats to Internal Validity 0 Potential confounds threats to internal validity that can hurt causeand effect conclusions History when an event occurs at same time as treatment event becomes alternative explanation 0 Ex conducting research on Sept 11 Maturation participants naturally change over time 0 Ex Study conducted during 1St week of quarter vs 9th week of quarter Testing taking a test changes subsequent testing 0 pretest vs posttest pretest could give you information about what the research is looking at Instrumentation instruments used to measure performance may change over time 0 Ex Instrument test could change from pretest to posttest Regression to the mean participants may perform very well or very poorly due to chance if they were tested again then they would regress to the mean score 0 Participant might be chosen for group because they have an extreme score during pretest c When retested likely person s score would regress closerto mean even without experimental treatment Subiect Mortality attrition subjects may drop out of study which ruins equivalence between conditions Selection if differences exist between experimental and control groups at beginning of study 0 because there is no random assignment it is possible that experimental and comparison group are different in important ways Interactions with selection when one group reacts differently than the other rou to above conditions 0 amp experimental group matures more ra idl than com arison group 0 H control group is more affected by external event than experimental group These 2 threats are not controlled even by true experiments Contamination when there is communication about the experimental intervention between the groups of participants 0 Ex when those in experimental group talk about treatment with those in control group 0 can be problematic in natural setting ie quasiexperimental research because participants may know or work with each other Hawthorne effect changes in people s performance when they know si ni cant others are watching them 0 h boss researcher teacher 0 can be problematic in natural setting because can have effects on actual life if research is done at yourjob then this could have an effect on yourjob QuasiExperimental Designs 0 Even without random assignment can still create internal validity when Use a group like the treatment group as a comparison Researchers obtain pretest and posttest measures from both groups 0 need pretest and posttest measures because can t assume that the treatment and comparison groups are equivalent on important variables 0 measures can prove this however 0 Nonequivalent control group design 39 01 X 02 39 O1 02 o 01 Pretest measure Observation 1 o 02 Posttest measure Observation 2 o X treatment like independent variable 0 Dashed line treatment and comparison groups not formed by random assignment Example comparing sections of research methods and developmental psych on reasoning skills Adding comparison group minimizes threats to internal validity due to history maturation testing instrumentation and regression Add tremendous advantage in ability to make causal claims by adding comparison group However can t pretest for all possible differences between groups so can t rule out all differences Can t rule out interactions with selection because they might not have same experiences in real life 0 HAVING A COMPARISON GROUP WILL HELP WITH INTERNAL VALIDITY BUT WILL NEVER REPLACE RANDOM ASSIGNMENT o Interrupted Timeseries Design Simple interrupted timeseries design researchers obtain periodic measures before and after a treatment has been introduced 0 0102 03 O4 05XO6 O7 08 09 010 o 5 observations pretest treatment 5 observations posttest No comparison group These designs are likely to be used when a new government reform business campaign or new product is being introduced 0 Example Connecticut used design to analyze whether increased police force on speeding actually reduced traf c fatalities Helpful to know that treatment is planned for the future so that you can start research with enough time to measure observations before change 0 OthenNise you can use archival records for pretest observations Obvious discontinuities in a time graph are needed to show strong evidence for treatment effectiveness O o Gradual changes cannot be differentiated from regular fluctuations Main problem ruling out alternative explanations for discontinuity Threat of history is a common problem 0 Threat of instrumentation can be problem because institutions change the way they record information when they make procedural changes Threats of maturation testing and regression are generally ruled out because there are multiple observations 0 Those changes would likely be gradual not discontinuous Time Series with Nonequivalent Control Group Design Procedure is identical to simple time series but includes comparison group to provide pre and posttreatment observations 0 0102 O3 O4 05XO6 O7 08 09 010 O 0102 O3 O4 05 O6 O7 08 09 010 0 0105 Pretest measures 06 010 posttest measures X treatment Line treatment and comparison groups not formed by random assignment Controls threat of history events should be happened to both groups at the same time 0 Example could include traffic fatalities from other comparable states 000 0 Program evaluation 0 O O Assesses effectiveness of human service organizations government agencies schools hospitals police that cannot use pro ts as a measure of success Goal Provide feedback to organizations in order to decide what services to provide and how to provide them effectively Interdisciplinary includes psychology sociology economics political science and education Largest scale application of research methods ideas Four questions of program evaluations Example Instituting a recycling program Needs Is an agency or organization meeting needs of the people it serves Process How is a program being implemented 0 Can use observational research Outcome Has a program been effective in meeting its stated goal Efficiency Is a program costef cient relative to alternatives 0 May use archival records experimental or quasi experimental techniques
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'