SOC/STAT/CSSS 221: Statistical
Concepts and Methods for the Social Sciences
Homework Assignment
8
Please
use this form to record your answers and include any work that you would like
to show the grader (you can paste in photos or scans of your hand-written work
if you like). Save this document in Word,
rtf, or pdf format and submitONE document for grading via the course
website.
Even in the modern era, countries differ
dramatically in terms of the health of their populations. For example, average life expectancy is in
the 80’s in some countries but remains in the low 40’s in other countries. Some researchers claim that the most
effective way to improve the health of the worst-off populations is to provide
more education for people in these countries. To assess this argument, use the
following analysis of data from a random sample of countries
1.
Interpretation of Model 1.
a.
Write
out the regression equation implied in the results for Model 1.
b.
Predict
the life expectancy for a country in which 75% of the population can read and
write.
c.
Interpret
the meaning of the constant in Model 1.
d.
Interpret
the slope coefficient for the effect of % Literate in Model 1.
e.
Is the
association between literacy and life expectancy statistically
significant? Explain how you reached
your conclusion.
f.
Interpret
the coefficient of determination in Model 1.
g.
Do
these results support the idea that population health (as measured by life
expectancy) is affected by education levels (as measured by % literate)? Explain your answer.
2.
A critic of the education argument suggests
that any association between literacy and life expectancy is a spurious
byproduct of urbanization; in countries in which the population is concentrated
in cities people have access to both schooling and health care, creating an
association between literacy between literacy and life expectancy that is not
causal. To test this argument, interpret
Model 2 of the table above.
a.
Write
out the regression equation implied in the results for Model 2.
b.
Predict
the life expectancy for a country in which 75% of the population can read and
write and 50% of the population lives in cities.
c.
Interpret
the meaning of the constant in Model 2.
d.
Interpret
the slope coefficient for the effect of % Literate in Model 2.
e.
Interpret
the slope coefficient for the effect of % Urban in Model 2.
f.
Interpret
the coefficient of multiple determination in Model 2
g.
Do
these results support the argument that the association between life expectancy
and % literate is spurious? Explain your
answer.
