chap. 5 sec. 2a notes.pptx

Upload: jhaselhorst

Post on 04-Jun-2018

225 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/14/2019 Chap. 5 Sec. 2a Notes.pptx

    1/12

    Designing ExperimentsChapter 5 Section 1a

  • 8/14/2019 Chap. 5 Sec. 2a Notes.pptx

    2/12

    Variables

    Response The variable we are studying

    by the outcome of an

    experiment Also called dependent variable

    Explanatory Helps explain or influences

    changes in the response variable

    Also called independent variable

  • 8/14/2019 Chap. 5 Sec. 2a Notes.pptx

    3/12

    Basic Vocab of Experiments Experimental Units

    The individuals on which the experiment is done Called SUBJECTS if the units are humans

    Factors The explanatory variables of an experimentTreatment

    The specific experimental condition applied to the units

  • 8/14/2019 Chap. 5 Sec. 2a Notes.pptx

    4/12

    #5.34 Identify the experimental units, factors,treatments, and the response variables You can use your computer to make long-

    distance telephone calls over the Internet. How will the cost affect the use of this service? Auniversity plans an experiment to find out. It will offer the service to all 350 students in one ofits dormitories. Some students will pay a low

    flat rate. Others will pay higher rates at peakperiods and very low rates off-peak. Theuniversity is interested in the amount and timeof use and in the effect on the congestion of the

    network.

  • 8/14/2019 Chap. 5 Sec. 2a Notes.pptx

    5/12

    More Vocab (Ex. 5.15 pg. 356) Basic design of a single treatment experiment

    Treatment Observe Response Placebo Effect

    When a dummy treatment is as effective as theactual treatment

    This is caused by outside factors

    Control Group A group of experimental units that receive aplacebo or no treatment, but experience thesame conditions as the treatment group

  • 8/14/2019 Chap. 5 Sec. 2a Notes.pptx

    6/12

    Basic Principles of Statistical Design ofExperiments1. Control2. Replication3. Randomization

  • 8/14/2019 Chap. 5 Sec. 2a Notes.pptx

    7/12

    Control The overall effort to minimize the effects of

    lurking variables on the response (Dont confuse w/ Control Group)

    Comparison is the simplest form of control. Compare two or more treatments (Control Group

    & Treatment Group) in order to preventconfounding the effect of a treatment with otherinfluences, such as lurking variables.

    Think Gastric Freezing Example (pg. 358) Uncontrolled experiments usually show a much

    higher success rate than proper comparative

    experiments

  • 8/14/2019 Chap. 5 Sec. 2a Notes.pptx

    8/12

    Replication Use enough subjects to reduce (it cannot

    eliminate) chance variation & bias This DOES NOT mean repeat the same

    experiment a 2 nd time in this instance. Would you expect to get the exact same results if

    you did?

  • 8/14/2019 Chap. 5 Sec. 2a Notes.pptx

    9/12

    Randomization Use impersonal chance (probability) to assign

    experimental units to treatments. Systematic differences among groups of

    experimental units in a comparative experimentcause biaschance (probability) eliminates this!

    Randomization allows us to assume that thetreatment groups are essentially similar.

  • 8/14/2019 Chap. 5 Sec. 2a Notes.pptx

    10/12

    Diagramming Experiments

    #5.44 on page 366What to show: randomization, size of groups, treatments,response variable

  • 8/14/2019 Chap. 5 Sec. 2a Notes.pptx

    11/12

    Statistical Significance An observed effect

    so large that it would rarely occur by chance.

    Think back to

    Gastric FreezingExample (pg. 356)

  • 8/14/2019 Chap. 5 Sec. 2a Notes.pptx

    12/12

    Pg. 357 5.33-5.37 oddPg. 364 5.39-5.43 oddHomework Assignment

    Read pages 358-371