(b) For a one inch gain in height, we would expect a
4.356 pound gain in weight. Someone who is
0 inches tall would be expected to weigh –149.934
pounds. The unreasonable answer reflects curvi-
linearity of the relationship at the extremes. The
correlation is .904, and both the slope and the cor-
relation are significant.9.33 As a 5 8 male, my predicted weight is 5
4.356(Height) 149.934 5 4.356 68 149.934 5
146.27 pounds.
(a) I weigh 146 pounds. (Well, I did a few years ago.)
Therefore the residual in the prediction is
146 2 146.27 52 0.27.
(b) If the students on which this equation is based
under- or over-estimated their own height or weight,
the prediction for my weight will be based on in-
valid data and will be systematically in error.
9.35 The male would be predicted to weigh 137.562 pounds,
while the female would be predicted to weigh 125.354
pounds. The predicted difference between them would
be 12.712 pounds.
9.37 Although the regression line has a slight positive slope,
the slope is not significantly different from zero. The
equation for the regression line is 5 0.429X 1
221.843.
Chapter 10
10.1 (b) rpb 52 .540; t 52 2.72
(c) Performance in the morning is significantly related
to people’s perception of their peak periods.
10.3 It looks as though morning people vary their perform-
ance across time, but evening people are uniformly poor
performers.
10.5 t 5 2.725. This is equal to the ttest on rpb.
10.7 0.202X 1 0.093; when 5
10.9 (b). (c) t 51 .27, not significant.
10.11 (a). (b) 5 12.62, p ,.05.
10.13 (a). (b) z 5 4.60,p ,.05.
10.15
10.17 An would correspond to The
closest you can come to this result is if the subjects
were split 61 39 in the first condition and 39 61 in the
second (rounding to integers).
10.19 (a).
(c) This approach would be preferred over the ap-
proach used in Chapter 7 if you had reason to be-
lieve that differences in depression scores below
x^2 =2.815 [p=.245]; fC=.087> >r^2 =.0512 x^2 =10.24.t=.733.t=.886f=.628 x^2f=.2560.608=YY N= X=X=2.903, YNYNY 2 YN=- 3 -
¿ – YNthe clinical cutoff were of no importance and
should be ignored.
10.21 (b) If a statistic is not significant, that means that we
have no reason to believe that it is reliably differ-
ent from 0 (or whatever the parameter is under
H 0 ). Here we have no reason to believe that there
is a relationship between the variables. Therefore,
it cannot be important.
(c) With the exception of issues of power, sample size
will not make an effect more important than it is.
It will simply increase the level of significance.Chapter 11
11.1 Source df SS MS F
Group 2 2100.00 1050.000 40.127*
Error 15 392.50 26.167
Total 17 2492.50
*p,.05 [F.05 (2,15) 5 3.68]
11.3 (a)
Source df SS MS F
Group 3 1059.80 353.267 53.301*
Error 36 238.60 6.628
Total 39 1298.40
*p,.05 [F.05 (3,36) 5 2.89]
(b)
Source df SS MS F
Group 1 792.10 792.10 59.451*
Error 38 506.30 13.324
Total 39 1298.40
*p,.05 [F.05 (1,38) 5 4.10]
(c) The results are difficult to interpret because the er-
ror term now includes variance between younger
and older participants, Moreover, we don’t know
if the levels of processing effect applies to both
age groups.
11.5 (a)
Source df SS MS F
Group 1 224.583 224.583 18.8*
Error 20 238.917 11.946
Total 21 463.500
*p,.05 [F.05 (1,20) 5 4.35]
(b) The twithout pooled variance 5 4.27; t^25 18.2.
(c) twith pooled variance 5 4.34; t^25 18.8.Answers 743