PSYC 201W Lecture Notes - Lecture 11: Random Assignment, Time Series, Internal Validity
Document Summary
Lack of random assignment is key point that distinguished this from true experiments. Quasi"s are more vulnerable to internal validity threats cuz no experimental control. In correlational design, it is subjects that determine the range of values for a variable, in quasi this is determined by the researcher. In quasi research, research has more control than in an observational setting. Order of xs and os indicate the temporal order of design. Numerical subscripts are used to indicate specific observations when there are more than one. X o: a treatment occurs and dc is measured after. One group pretest-posttest design: o x o, dv measured before and after treatment. Simple interrupted time-series design: o1 o2 o3 x o4 o5 o6, dc is repeatedly measured at periodic intervals before/after a treatment. Not possible to randomly assign participant conds. Random assignment is used in a way that cant be assumed confidently to create equivalent groups.