3.1 power point

Upload: krothroc

Post on 08-Apr-2018

221 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/7/2019 3.1 Power Point

    1/17

    Section 3.1

    Scatterplots and Correlation

  • 8/7/2019 3.1 Power Point

    2/17

    Are Baseballs Juiced?

    Do these data provide convincing

    evidence that baseballs have begun

    to fly farther over time?

  • 8/7/2019 3.1 Power Point

    3/17

    What Do You Want?

    Sometimes explanatory variables are called

    independent variables, and response variables are

    called dependent variables.

    When examining relationships among variables, twoquestions become important. Do you want to simply

    explore the nature of the relationship, or do you

    think that some of the variables help explain or even

    cause changes in the others?

  • 8/7/2019 3.1 Power Point

    4/17

    Which Is Response, Which Is

    Explanatory? Neith

    er? Researchers give several different amounts

    of alcohol to mice, then measure the change

    in each mouses body temperature in the 15

    minutes after taking the alcohol.

    Jim wants to know the relationship between

    mean SAT Math and Verbal scores. He

    doesnt believe either score explains the

    other.

    Julie wonders if she can predict a states

    mean SAT Math score if she knows its

    mean SAT Verbal score.

  • 8/7/2019 3.1 Power Point

    5/17

    Whats the Relationship?

    If there is an explanatory variable, always plot it on

    the horizontal axis (x-axis)

    A scatterplot shows the relationshipbetween two quantitative variables

    measured on the same individuals.

  • 8/7/2019 3.1 Power Point

    6/17

    Interpreting a ScatterplotAs in any graph of data, look for the overall pattern

    and striking deviations from that pattern:

  • 8/7/2019 3.1 Power Point

    7/17

    Describing a Scatterplot Trend: positive, negative, or none

    Unusual Features: clusters,outliers, or influential points

    Shape: linear, curved, or neither

    Strength: strong, moderate, orweak; constant or varying

    strength

  • 8/7/2019 3.1 Power Point

    8/17

    Not just quantitative

    (Southern states in blue)

    Categorical variables can becommunicated by colors and symbols

  • 8/7/2019 3.1 Power Point

    9/17

  • 8/7/2019 3.1 Power Point

    10/17

    Related by Association

    Two variables can have a positiveassociation

    High values associated

    with high values

  • 8/7/2019 3.1 Power Point

    11/17

    High values associated

    with low values

    a negative association

  • 8/7/2019 3.1 Power Point

    12/17

    or no clear association or pattern.

  • 8/7/2019 3.1 Power Point

    13/17

    Which Correlation is Stronger?

    Set 1

    Neither!

    or

    Set 2

    Our eyes are not good judges of how

    strong a linear relationship is.

    They are the same set of data, plotted ondifferently sized fields.

  • 8/7/2019 3.1 Power Point

    14/17

    Do My Eyes Deceive Me?

    Data analysis of a scatterplot needs to

    be supplemented by a numerical

    measure:

  • 8/7/2019 3.1 Power Point

    15/17

    Correlation Coefficient: -1 r 1

  • 8/7/2019 3.1 Power Point

    16/17

    Cautions!

    Only measures thestrength oflinearrelationshipsnot

    curves, no matterhow strong they are

    Correlation is notresistant

    *CORRELATION CAUSATION*

    Correlation alone is not a numericalsummarymeans and standard

    deviations ofboth x and y are needed

    Bothvariablesmust be

    quantitative

  • 8/7/2019 3.1 Power Point

    17/17

    You Try It!

    Freshmen at the Webb Schools go on a backpackingtrip at the start of each school year. Students are

    divided into hiking groups of size 8 by selection of

    names from a hat. Prior to departure, each students

    body weight and backpack weight are measured (both in

    pounds). Here are data from one hiking group in arecent year:

    1. Enter values into L1 and L2

    2. 2nd StatPlot; 1st graph; Xlist:L1; Ylist:L2; Zoom; 9

    3. Correlation Coefficient (r): Stat; Calc;

    LinReg(a+bx); L1, L2

    4. Ifrand r2 dont show, go to 2nd; Catalog; DiagnosticOn

    Body weight (lb): 120 187 109 103 131 165 158 116

    Backpack (lb): 26 30 26 24 29 35 31 28