Chapter 8
How to Test for Trends
‘The first statistical problem we posed in this book, in connection with
Fig. 1-24, dealt with a drug that was thought to be a diuretic, but that
experiment cannot be analyzed using our existing procedures. In it, we
selected different people and gave them different doses of the diuretic
then we measured their urine output. The people who received larger
doses produced more urine, The statistical question is whether the re-
sulting pattern of points relating urine production to drug dose provided
sulffcient evidence to conclude thatthe rug inereased urine production
in proportion to drug dose. This chapter develops the tools for analy2-
ing such experiments. We will estimate how much one variable in-
ereases (or decreases) on the average as another variable changes with a
regression line and quantifies the strength of the association with a cor-
relation coefcient.”
"Simple linear regression sa special case ofthe more general method of mudple
‘regression in which ease there are multiple independent variables. Fora discussion of
multiple regression and related procedures writen in the same style as this book, see
S.A, Glantz and BK, Slinker, Prinerof Applied Regression and Analysis of Variance,
(@nd ea), McGraw-Hill, New York, 200
20HOW To TEST FOR TRENDS. 2
MORE ABOUT THE MARTIANS
‘As inal other statistical procedures, we want to use a sample drawn at
random from a population to make statements about the population
Chapters 3 and 4 discussed populations whose members are normally
distributed with mean and standard deviation rand used estimates of
these parameters to design test statistics (like F and 1) that permitted us
to examine whether or not some discrete treatment was likely to have
affected the mean value of a variable of interest. Now, we add another
parametric procedure, linear regression, to analyze experiments in
Which the samples were drawn from populations characterized by a
mean response varying continuously with the size ofthe treatment. To
understand the nature of this population and the associated random
samples, we retum again to Mars, where we can examine the entire
population of 200 Martians.
Figure 2-1 showed that the heights of Martians are normally dis-
tributed with a mean of 40 cm and a standard deviation of $ em. In ad-
dition to measuring the heights of each Martian, lt us also weigh each
one, Figure 8-1 shows a plot in which each point represents the height x
and weight y of one Martian, Since we have observed the entire popula-
tion, there is no question that tall Martians tend to be heavier than short
Martians.
‘There are a number of things we can conclude about the heights
and weights of Martians as well as the relationship between these two
variables. As noted in Chapter 2, the heights are normally distributed
with mean p= 40 em and standard deviation = 5 cm. The weights
are also normally distributed with mean j.= 12 g and standard devia-
tion ¢ = 2.5 g. The most striking feature of Fig. 8-1, however, is that
the mean weight of Martians at each height increases as height in-
creases.
For example, the Martians who are 32 em tall weigh 7.1, 79, 83,
and 8.8 g, so the mean weight of Martians who are 32 em tall is 8 g.
The 8 Martians who are 46 cm tal weigh 13.7, 14.5, 148, 15.0, 15.1,
15.2, 15.3, and 15.8 g, so the mean weight of Martians who are 46 em
tall is 15 g. Figure 8-2 shows that the mean weight of Martins at each
height inereases nearly as height increases.
This Tine does not make it possible, however, to predict the
weight of an individual Martian if you know his or her height. Why
not? There is variability in weights among Martians at each height
Figure 8-1 reveals that standard deviation of weights of Martians232 CHAPTER @
200 Marans oe
;
2
Oa a
% & = & & &
Height em
Figure 8-1 Tho relationship between height and weight in the population of
200 Martians, with each Martian represented by a cele. The weights st any
Given height follow a normal distribution. In addition, the mean weight of Mit-
tiers at any given height increases linearly with height, and the variailty 9
‘weight at any given height is the same regardloss of height A population must
have these cheracterstics to be sutable for near regression or correlation
analysis.