GED 111(CDS111) Statistics in Modern Society

advertisement
CLD9003
Statistics in Modern Society
Alan Lam
Dept of CDS
Course Outline
• A General Education course
• Sectional Approach
– Class discussion (Major Emphasis)
• Course text
– Bennett, J.O., Briggs, W.L. and Triola, Mario F.
2014 Statistical Reasoning for Everyday Life,
4/E, Addison-Wesley.
CLD9003 Statistics in Modern
Society
2
Course Outline
• Tentative Schedule
• Software
– Excel/SPSS
• Assessment
– Attendance
– Group Project
– Participation
– Examination
5%
20%
15%
60%
CLD9003 Statistics in Modern
Society
3
What Is/Are Statistics
• Two Definitions
– Statistics is the science of collecting,
organising, and interpreting data.
– Statistics are the data that describe or
summarize something.
GED111/CDS111 Statistics in
Modern Society
4
How does it work?
• Advertising
– Claim of 111.3 million watched a TV program
– Survey of 5000 homes
• Goal – population
– The population in a statistical study is the
complete set of people or things being studied.
– Population parameters
• Specific characteristics of the population
• Mean, median, mode, standard deviation, size, etc.
GED111/CDS111 Statistics in
Modern Society
5
What actually gets studied?
• Sample
– A subset of the population from which data
are actually obtained
– Sample statistics
• Characteristics of the sample found by
consolidating or summarizing the raw data
• Mean, median, mode, standard deviation, size, etc.
– Raw data
• The actual measurements or observations
collected
CLD9003 Statistics in Modern
Society
6
How do sample statistics relate to
population parameters?
• Estimation / Inferential statistics
– Margin of error
• The sample represents the population fairly
• The margin of error is generally smaller for a larger
sample
–
–
–
–
10% for sample size of 100
5% for sample size of 400
3% for sample size of 1,000
1 % for sample 10,000
GED111/CDS111 Statistics in
Modern Society
7
Margin of Error
• Used to describe the range of values likely
to contain the population parameter.
• The range is found by sample ± margin of
error
• From (sample statistic – margin of error) to
(sample statistic + margin of error)
GED111/CDS111 Statistics in
Modern Society
8
The Process of a Statistical Study
Start
1 Identify goals
2 Draw from population
Population
Sample
3 Collect raw data and summarize
5 Draw conclusions
Population Parameters
Sample Statistics
4 Make inferences about population(estimation)
GED111/CDS111 Statistics in
Modern Society
9
Decisions for an Uncertain World
• Decision making under uncertainty
– Ad hoc case with no historical record
– One off decision making
GED111/CDS111 Statistics in
Modern Society
10
Exercises
•
•
•
•
11. Stem Cell Research p.9
13. Galactic Distances p.9
21. Do People Lie About Voting? P.10
Exercise
– 24, 25, 26 (p.10)
GED111/CDS111 Statistics in
Modern Society
11
Sampling
• Census
– The collection of data from every member of a
population
• Sampling
– Collection of data from part of the population
• Representative Sample
– The relevant characteristics of the sample members
are generally the same as the characteristics of the
population
• Bias
– If a statistical study’s design or conduct tends to
favour certain results
GED111/CDS111 Statistics in
Modern Society
12
Sampling Methods
• Simple Random Samples
– Assign number to every possible outcome and each
outcome is equally likely to be selected
• Systematic Sampling
– Selection made every nth possible outcomes
• Convenience Samples
– Based on the convenience of selection
– More prone to bias than most other forms of sampling
GED111/CDS111 Statistics in
Modern Society
13
Sampling Methods
• Cluster Samples
– Selection of all members in randomly selected
groups
• Stratified Samples
– Population is divided into strata (subgroup)
– Selection is made from each stratum and
combined to form the sample
GED111/CDS111 Statistics in
Modern Society
14
Summary of Sampling Methods
• Regardless of how a sample is chosen,
the study can be successful only if the
sample is representative of the population
• A biased sample is very unlikely to be a
representative sample
• A well-chosen sample has a good chance
of being representative, but still may turn
out to be biased just because of bad luck
in the actual drawing of the sample
GED111/CDS111 Statistics in
Modern Society
15
Exercises
• Census or not
– 7. You want to determine the mean GPA of
the 50 students in a particular high school
Maths class
– 8. You want to determine the mean GPA of all
high school senior in USA
– 9. You want to determine the mean annual
energy costs of all homes in Missouri
– 10. You want to determine the mean
temperature of coffee server at Starbucks
GED111/CDS111 Statistics in
Modern Society
16
Exercises
• Identifying the Sampling Methods p.19
– 23-32
• Choosing a Sampling Methods p.20
– 39-42
GED111/CDS111 Statistics in
Modern Society
17
Types of Statistical Studies
• Observational Study
– Observe or measure characteristics of the subjects,
but do not attempt to influence or modify these
characteristics
• Experiment
– Apply some treatment and observe its effects on the
subject of the experiment
• Meta-analysis
– Study a topic that has been the subject of many
previous studies
GED111/CDS111 Statistics in
Modern Society
18
Observational Study :
Retrospective Studies (Case-Control
Study)
• A retrospective study is an observational study
that resembles an experiment because the
sample naturally divides into two (or more)
groups. The participants who engage in the
behaviour under study form the cases, like a
treatment group in an experiment. The
participants who do not engage in the behaviour
are the controls, like a control group in an
experiment.
• Usually uses data from the past, such as official
records or past interviews.
GED111/CDS111 Statistics in
Modern Society
19
Observational Study :
Prospective Studies (Longitudinal
study)
• Designed to collect observations in the
future from groups that share common
factors
CLD9003 Statistics in Modern
Society
20
Experiment :
Treatment & Control Groups
• Treatment group
– The group of subjects who receive the
treatment being tested
• Control group
– The group of subjects who do not received
the treatment being tested.
• To assign treatment & control groups
– Randomisation
– Large number of subjects
GED111/CDS111 Statistics in
Modern Society
21
Confounding Factor
• The effect of different factors are mixed so
that we cannot determine the effects of the
specific factors we are studying.
• The factors that lead to the confusion are
called confounding factors.
GED111/CDS111 Statistics in
Modern Society
22
Placebo Effect
• It refers to the situation in which patients
improve simply because they are receiving
a useful treatment
• A placebo lacks the active ingredients of a
treatment being tested in a study, but is
identical in appearance to the treatment.
Thus, study participants cannot distinguish
the placebo from the real treatment.
GED111/CDS111 Statistics in
Modern Society
23
Experimenter Effects
• Occurs when a research or experimenter
somehow influences subjects through
such factors as facial expression, tone of
voice, attitude.
• Blinding
– Single-blind
– Double-blind
GED111/CDS111 Statistics in
Modern Society
24
Exercises
• Observational – case-control study?
• Experiment – treatment/control groups?
• Meta-analysis
– Ex 10, 12, 13, 14 (p.32)
• Anything wrong?
– Ex 21, 22, 24, 25, 27 (p.32)
GED111/CDS111 Statistics in
Modern Society
25
Should you believe a Statistical
Study?
• Eight Guidelines for Critically Evaluating a
Statistical Study
GED111/CDS111 Statistics in
Modern Society
26
The guidelines
• 1. Get a big picture view of the study
– What was the study designed to determine?
– The population?
– Observation, experiment, meta-analysis?
GED111/CDS111 Statistics in
Modern Society
27
The guidelines (cont.)
• 2. Consider the source, particularly with
regard to whether the researchers may be
biased
– Peer review
• Several experts in a field evaluate a research
report before the report is published
• Published journal
GED111/CDS111 Statistics in
Modern Society
28
The guidelines (cont.)
• 3. Look for Bias in the sample
– Selection bias
– Participation bias
• Self-selected surveys/voluntary response surveys
GED111/CDS111 Statistics in
Modern Society
29
The guidelines (cont.)
• 4. Look for problems in defining or
measuring the variables of interest, which
can make it difficult to interpret any
reported results.
– Variable is any item or quantity that can vary
or take on different values
– Variables of interest in a statistical study are
the items or quantities that the study seeks to
measure
GED111/CDS111 Statistics in
Modern Society
30
The guidelines (cont.)
• 5. Watch out for confounding variables
that can invalidate the conclusions of a
study
GED111/CDS111 Statistics in
Modern Society
31
The guidelines (cont.)
• 6. Consider the setting and wording of any
survey, looking for anything that might
tend to produce inaccurate or dishonest
responses.
– Badly worded survey
– Sensitive information
GED111/CDS111 Statistics in
Modern Society
32
The guidelines (cont.)
• 7. Check that results are fairly represented
in graphics and concluding statements,
because both researchers and media
often create misleading graphics or jump
to conclusions that the results do not
support.
– Personal bias
– Unwarranted conclusions
GED111/CDS111 Statistics in
Modern Society
33
The guidelines (cont.)
• 8. Stand back and consider the
conclusions.
– Did the study achieve its goal?
– Do the conclusions make sense?
– Can you rule out alternative explanations for
the results?
– If the conclusions make sense, do they have
any practical significance?
GED111/CDS111 Statistics in
Modern Society
34
Focus on Sociology
• Does Daycare Breed Bullies pp45-48
CLD9003 Statistics in Modern
Society
35
Download