| Hand in:
Two-way tables Today, SPSS. pp. 612ff. SPSS Intro Handout, p. 6. Re-create the results shown on the handout. Also create and print two-way tables with the row, column, and total percents. Use both the raw data and the pre-tallied data and observe that the results are identical. IPS give no raw data sets to practice on; all the data are
pre-tallied.
9.26 Web ref's (SPSS) Do everything they ask
for except
for the "significance test."
9.24 a, b mutations (SPSS)Fill in the blank
row,
then type the data into SPSS in the appropriate form, and do part b.
|
Read, discuss
- - - - - -
|
Optional
|
Categorical data with SPSS: (p. 6,
Intro
handout)
Pre-tallied? Data> Weight Cases>Count to Frequency
box.
Analyze>Descriptive Statistics>Crosstabs. Cells
button.
(3-way? Third to Layer box)
Graph>Interactive> Bar: 100% box for stacked
percents,
one variable to horiz. axis, other to legend box, stacked or clustered
. Third to panel.
Start here Wed.
Simpson's paradox: An association or
comparison that holds
for all or several subgroups can reverse direction when
the
data are combined into a single group.
Example from text. p. 588
example 9.10
SPSS output
Parallel Continuous situation:
Cars.sav , like econ graduates problem (Ch.2). (X=weight,
Y=time to accelerate to 60. Heavier car should be slower? Oops.
Panel
with #of cylinders, or color with horsepower.)
- - - - - - - - - - - - - - - - - - - - - - - -
Chapters 1 and 2 have covered analyzing
data
that was given to us--what it said about itself.
"Exploratory Data Analysis"
Informally, use to develop
guesses, suspicions, hypotheses about the world the data came
from.
"Anecdotal evidence"--haphazard information, often noticed
because
striking. Often unrepresentative of anything.
Ch.
3: Producing Data:
Aim:
create data sets that will allow us to make inferences to a
larger
world than just the data we have. "Statistical
Inference" = "Confirmatory Analysis"
Design how to get data...
Observational
Study: Observes individuals, measures variables, does not
influence the responses. (3.3)
Take Sample from a population, examine it,
hope it's representative so we can infer population is like
sample.
(Not very useful for cause-and-effect--see sec. 2.5)
(Census--whole population)
Experiment:
Imposes
treatment
on individuals, to see how the treatment
influences the response.
(3.2)
Best for cause-and-effect.
Confounding: Two variables
(explanatory
or lurking) are confounded when you can't sort out their
effects
on a response variable.
--Used to be: coffee drinking and smoking--most
people did both, or neither...
--2 years ago:: women who ate at least one
serving/day
of whole grain (cereal, bread) much less likely to have heart attack.
(Who eats whole grains? Were
those variables taken into account? ?)
| Sievers home | Math251-Fall05/Dayps14.htm | 11:45pm | 9/25/05 |