SP 225 Lecture 11 - PowerPoint PPT Presentation

1 / 18
About This Presentation
Title:

SP 225 Lecture 11

Description:

SP 225 Lecture 11 Introduction to Hypothesis Testing The Most Liberal Senator Article by the National Journal rated Senator Kerry as the most liberal member of the ... – PowerPoint PPT presentation

Number of Views:82
Avg rating:3.0/5.0
Slides: 19
Provided by: Megha50
Category:

less

Transcript and Presenter's Notes

Title: SP 225 Lecture 11


1
SP 225Lecture 11
  • Introduction to Hypothesis Testing

2
The Most Liberal Senator
  • Article by the National Journal rated Senator
    Kerry as the most liberal member of the Senate
    during the 2004 Presidential Election
  • Aid decision-making by simplifying large amounts
    of data
  • Can be misapplied
  • Can be an essential tool for interest groups

3
Questions
  • Explain the National Journal's methodology for
    determining its scores. What method do the
    authors use to reexamine the roll call data used
    by the National Journal?
  • From a statistical point of view, what is the
    importance of the absentee data presented in
    Table 1 for the analysis of legislators voting
    records?
  • What factors influence a legislators voting
    decisions? How do these issues complicate the
    interpretation of voting records?
  • Explain the format and the purpose of the
    graphing technique employed in figures 1-4.
  • According to Figure 3, where does President Bush
    stand in comparison with the members of the
    Senate for those 62 votes? What makes the
    authors uncertain of this rank order? Why do the
    authors maintain it is necessary to analyze
    President Bushs stated issue positions in order
    to interpret Senator Kerrys voting record
    properly?
  • Using the results shown in figure 4, explain how
    selecting a larger set of voting decisions over a
    longer period of time results in a different rank
    ordering.
  • What was your overall assessment of the validity
    of the National Journals liberal-to-conservative
    scorecard? Was the public well-served by the
    magazines publication of its analysis?

4
Research Questions
  • Are the elderly more likely to vote than the
    population as a whole?
  • Do college graduates really make more money?

5
Investigating Questions
  • Compare group statistics to population statistics
  • All data obtained through EPSEM sampling

6
Evaluating Claims
  • Reasons for differences between sample and
    population statistics
  • Sampling variability
  • Genuine differences

7
What CAN We Say?
  • We can give the probability there are genuine
    differences

8
Making Probability Statements
  • We are 95 confident there is a difference in
    income between men and women
  • Benadryl makes patients more drowsy than Claritin
    (95 confident)

9
Example of Mean Test
  • Nationally, the population as a whole watches 6.2
    hours of TV per day. A random sample of 1,017
    senior citizens reports watching an average of
    5.9 hours per day with a standard deviation of
    .7. Is the there a difference between the amount
    of television senior citizens watch as compared
    to the population as a whole?

10
Review of Scientific Method
The Wheel of Science
11
Hypothesis Testing
  • 5 Step Process
  • Making assumptions about meeting test
    requirements
  • Stating the null and alternative hypothesis
  • Selecting the sampling distribution and
    establishing the critical region
  • Making a decision and interpreting the test
    results

12
Test Requirements
  • Randomly collected data
  • Interval-ratio data
  • Normally distributed or large sample size

13
Null Hypothesis
  • Always a statement of no difference
  • The researchers belief is the alternative
    hypothesis
  • Null hypothesis can be disproven but never proven

14
Alternative Hypothesis
  • Statement which is true when the null hypothesis
    is false

15
Selecting a Distribution
  • Currently, only use the normal distribution
  • Size of the critical region determines the level
    of certainty desired
  • Rejection region is the range of z-scores where
    the null hypothesis is rejected
  • Alpha is the level of uncertainty

16
Decision Making Criteria
  • Probability of our observed statistic given the
    null hypothesis is true.
  • Exact value is called a p-value

17
SPSS Test
  • Using SPSS and GSS test the following claims
  • Extreme liberals sampled in the GSS share the
    opinion of the general population when asked if
    homosexuals should be allowed to become teachers
    (colhomo).
  • Extreme conservatives sampled in the GSS share
    the opinion of the general population when asked
    if homosexuals should be allowed to become
    teachers (colhomo).

18
Essay
  • In the GSS, respondents are asked to rank their
    ideological beliefs (polviews)on a scale from 1
    to 7, where 1 represents extremely liberal and 7
    represents extremely conservative. Choose one
    ideology, either extremely liberal or extremely
    conservative, and write an essay describing the
    difference in beliefs between people sampled in
    GSS rating themselves as extremely ideological
    and the general population. Use any three
    measures that show a significant difference.
    Complete a full hypothesis test for each variable
    including the hypotheses, result and conclusion.
    Staple all SPSS output and calculations to your
    final essay. (1-2 pages. Due in 1 week)
Write a Comment
User Comments (0)
About PowerShow.com