CH 10 Flashcards
Two different types of Quasi-experimental designs are…
Natural experiments (ex post facto) subject variable designs (falling on a continuum between weak and strong experimental design. extraneous variables aren't always controlled.)
Natural experiments are…
a treatment is applied but by nature, and observed by the researcher. Investigator must wait for event to occur and measures are made after the fact. Effects of events may be assessed by comparison of behaviors before/after or subjects that did/not receive treatment.
Beddington College Study (1943)- students more liberal after college years. Other factors can impact. Basic flaw: cannot imply cause and effect.
Subject Variable designs are…
ex: IQ, weight, anxiety levels, gender, age, schizophrenia.
Can’t be manipulated by investigator, only selected. Co-variation between one or more variables and subject variable is likely.
Produces correlations, so no causal relationships can be inferred.
Procedure used to minimize confounds is…
if variable (IQ) is close related to performance (Test) you can match by IQ and compare performance based on IQ, and minimize confounds. Matching can reduce sample size, which means the effect size is larger, which means you'll have trouble rejecting the null.
Some things are difficult to measure in a precise or valid way. AN example is…
schizophrenia.
Institutional decline is…
taking a normal person and putting them in an institution, and will become more like the patients there. Normal people go down hill the longer they stay.
Regression to the mean is…
artifacts that can produce confounds. Subtle and difficult to spot. Most retesting will show regression from extremes to the average behavior.
Ex: high doses usually used on subjects. behavior becomes more variable due to high dose, including scores on range you’re testing.
define assignment measures
used to assign participants to experimental and control groups. those with scores below the cutoff score are assigned to one group, and those with scores above the cutoff are assigned to the other group.
define design components.
those structures and procedures used in constructing research designs.
define Regression discontinuity design.
assigns participants to groups based on their scores on an assignment variable and assesses the effect of a treatment by looking for a discontinuity in the groups regression lines.
