Mt139C6 - Experimentation Allows us to study the effects of the specific treatments we are interested in while holding constant as many lurking

Info iconThis preview shows pages 1–5. Sign up to view the full content.

View Full Document Right Arrow Icon
MT139C6.PPT 1 Experimentation Allows us to study the effects of the specific treatments we are interested in while holding constant as many lurking variables as possible. Unlike an Observational Study, a well designed Experiment can provided evidence of causation. Experiment Deliberately and systematically imposes some treatment on individuals in order to observe the effect of the treatment on the response. Example: 5,000 individuals are randomly selected, then randomly assigned to one of two groups. One group is given aspirin each day for 5 years. The other group is given sugar pills. Following the study, symptoms of heart disease in the two groups are compared.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
MT139C6.PPT 2 Experimental Units (Subjects) The individuals on which the experiment is done. Experimental Units are called Subjects when the individuals are people. Example: The Experimental Units or Subjects in the aspirin study are the people randomly assigned to receive either aspirin or the placebo. Treatment The experimental condition imposed on the individuals of a study. A Treatment is often called a Factor and is made up of one or more Levels. A study may include more than one Factor. Example: Whether or not an individual receives aspirin in the above example is the one Factor of this study. There are two Levels of the Factor - aspirin and placebo.
Background image of page 2
MT139C6.PPT 3 Placebo Effect Many subjects respond in some way (often favorably) to any Treatment, even when a dummy Treatment is given. The response to a dummy Treatment is called the Placebo Effect. Example: Patients may report better health after treatment, even if assigned to the control, or no treatment, group of a study. Comparative Experiment Rather than observe the effect of one level of a single Treatment, an Experiment should compare two or more levels of a Treatment. In many studies, one level of the Treatment is a control condition. Example: The aspirin study compares the effect of aspirin versus a sugar pill placebo on self-reported symptoms of heart disease. If only the aspirin condition had been studied, the confounding placebo effect may have biased the results.
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
MT139C6.PPT 4 Experimental Design Describes the response variable(s), the Treatments or Factors (treated as explanatory variables in the analysis), and how Experimental Units are assigned to the Levels of one or more Treatments. Randomized Comparative Experiment Comparison of the effects of several levels of a Treatment is valid only when groups receiving the Treatments are similar prior to Treatment. To ensure similarity among our experimental groups, chance is used to assign experimental units to the various Treatments. Example: Subjects in the aspirin study were randomly assigned to
Background image of page 4
Image of page 5
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 06/06/2011 for the course MTH 139 taught by Professor Mikekadar during the Winter '00 term at Moraine Valley Community College.

Page1 / 20

Mt139C6 - Experimentation Allows us to study the effects of the specific treatments we are interested in while holding constant as many lurking

This preview shows document pages 1 - 5. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online