schedule study to avoid a predictable external event Instrumentation Problems - limitations in measurement
concurrent events, not past incidents History Solutions - control/comparison group minimize/eliminate communication between investigators History Problems - events that occur outside of an intervention that are out of the investigators control protocol for intervention administration
take repeated baseline measurements & average them Compensatory Equalization of Treatments Problem - individuals providing the interventions in the study purposefully/inadvertently supplement the activities of subjects in the control group to "make up for" what subjects in the experimental group are receiving Compensatory Equalization of Treatments Solution - mask the investigators clouds the true effects, if any, of the intervention Statistical Regression to the Mean Solution - eliminate outliers provide thorough, explicit instructions to subjects Statistical Regression to the Mean Problems - subjects enter a study with an extreme value for their baseline measure of the outcome of interest provide thorough, explicit instructions to subjects Diffusion/Imitation of Treatment Problem - subjects from different groups share aspects of the treatment in their group that prompts changes in behaviors by members in a different group Diffusion/Imitation of Treatment Solution - keep subjects in each group separated may alter behaviors of the control group Compensatory Rivalry - "we'll show them" attitude Resentful Demoralization - "we're getting the short end of the stick, so why bother" attitude Compensatory Rivalry/Resentful Demoralization Solutions - keep subjects in each group separated follow a protocol design to reduce problems related to timing Compensatory Rivalry/Resentful Demoralization Problems - communication among study participants is not controlled tightly -> subjects acquire knowledge about the different groups * longer time between measures increases possibility of maturation effects occurring Maturation Solution - random assignment to a treatment & control group document characteristics/reasons for withdrawal Maturation Problems - natural change over time ** may affect the power of the study to detect differences among the groups Attrition/Dropout/Mortality Solutions - replace lost subjects Assignment/Allocation Problems - process of placing subjects into groups results in differences in baseline characteristics between (or among) the groups at the onset of a study Assignment Solutions - random assignment Attrition/Dropout/Mortality Problems - loss of subjects during the course of a study
0 kommentar(er)
