4.33 Seriousness of the fire is a lurking (common—response) variable

Chapter 4
P1:Ll..L
a
___—
a
fires require
4.33 Seriousness of the fire is a lurking (common—response) variable: more serious
fire causes”
large
a
that
iriore attention and do more damage. It would he more accurate to say
more firefighters to be sent, rather than vice versa.
E)anage
done
Seriousness
of fire
direction:
4.34 A reasonable explanation is that the cause-and-effect relationship goes in the other
Doing well makes students feel good about themselves, rather than vice versa.
Amount
of selfesteem
Quality
of
work
h
0.925 —more than 90% of the variation in one SAT score can be explained throug
2
4.35 (a) r
indi
since
r,
smalle
much
a linear relationship with the other score. (b) The correlation would be
greater verbal
vidual students have much more variation between their scores. Some may have
some will be
skills and low scores in math (or vice versa); some will be strong in both areas, and
of large
e
scores
of—th
median
the
weak in both areas. By averaging—or, as in this case, taking
groups of students, we muffle the effects of these individual variations.
ies—shoe
4.36 Age is the lurking (common—response) variable here: we would expect both quantit
size and reading level—to increase as a child ages.
4
More on TwoVariabIe Data
81
4.37 The explanatory variable is whether or not a student has taken at least two years of foi-eign
language, and the score Ofl the test is the response. The lurking variable is the students’ English
ski1ls before taking (or not taking) the foreign language: students who have a good command of
English early in their high school career are more likely to choose (or be advised to choose) to take
a foreign language and to do well on an English test in ay case.
Amount
foreign
language
taken
English
score
1.
English
skills
1’
i
i.
4.38 In this case, there may be a causative effect, but in the direction opposite to the one sug
gested: People who are overweight are more hkely to be on diets, and so choose artificial sweet
eners over sugar. (Also, heavier people are at a higher risk to develop diabetes; if they do, they are
likely to switch to artificial sweeteners.)
I
Use of
sweeteners
\Veight
gain
4.39 Time standing up is a confounding variable in this case. The diagram below illustrates the
confounding between exposure to chemicals and standing up.
Chapter 4
ts (who are
uudmg) variable: Stronger studen
nfo
(eo
ing
lurk
a
be
may
,
e
enc
t, It) \ si udent’s intellig
ose to take these iiiath courses
iw niore likely to cho
)
ege
coll
to
get
they
for
nilce
iations on this idea;
Ic likely to succeed
er possible answers might he var
Oth
.
them
id
avo
nay
ts
den
stu
wi ile s eaker
s selfconfidc’nce, we might sup
ege depends on a student’
coll
in
s
ces
suc
t
tha
eve
c\uIple, if we beli
.
inure likely to choose math courses
) sc that coufu lent students are
College
suecess (C PA)
taken
in I lS.
i\1t)
IC) or sclf-onfdenue
elligence watch
be that children with lower int
ld
cou
It
se.
pon
res
households
4.41 This is a case of
ldren from lower Socioeconomic
chi
Or
l.
wel
as
des
gra
er
low
ir children
many hours of TV and get
ing and be unable to help the
view
on
visi
tele
it
lim
to
ly
watch more TV and get
where the parent(s) are less like
themselves lack education may
s
ent
par
the
e
aus
bec
k
wor
with their school
lower grades as a result.
common
Grades
Number of hours
spent watching
TV
(CPA)
j
[0 or
SOCIOCCQflOTflIC
status
l” relationship.
p explain this supposed “causa
hel
ld
cou
es
iabl
var
ing
lurk
are hope
4.42 Several possible
English and mathematics classes
in
e
enc
eri
exp
and
ce
gen
elli
Both music experience and
Factors such as innate int
music experience on test scores.
of
cts
effe
the
h
wit
d
nde
fou
dent. Sturlents from
lessly con
socioeconomic status of the stu
the
to
g
din
pon
res
he
also
may also do bet
test scores could
extensive exposure to music and
ive
rece
to
ly
like
re
mo
n.
be
affluent families may
other costly means of preparatio
and
rses
cou
ing
-tak
test
rd
affo
ter Ofl tests because they can
I