1
00:00:00,000 --> 00:00:04,129
Hi. In this lecture we're going to talk
about a type of decision making known as

2
00:00:04,129 --> 00:00:07,948
multi-criterion decision making. And the
idea here is that there's lots of

3
00:00:07,948 --> 00:00:11,767
different dimensions and you've got a
couple choices, and you're trying to

4
00:00:11,767 --> 00:00:16,051
figure out which choice, you know, sort of
makes you happy, which choice you think is

5
00:00:16,051 --> 00:00:20,232
sort of the better of the two or three or
four, right? So how does this work? We're

6
00:00:20,232 --> 00:00:24,431
going to do this in two ways, we're going
to do first, do it qualitatively. And then

7
00:00:24,431 --> 00:00:28,264
we'll do it quantitatively. So, let's do a
qualitative assessment first. By that I

8
00:00:28,264 --> 00:00:31,905
just mean just sort of no numbers, right?
It just means sort of like, which is

9
00:00:31,905 --> 00:00:35,786
better? Lets suppose you're looking at a
house. [laugh] This is a very small house,

10
00:00:35,786 --> 00:00:39,427
but let's suppose you're looking at a
house, and you're trying to decide, you

11
00:00:39,427 --> 00:00:43,308
know, should I buy this house, or there's
some other house down the street and I'm

12
00:00:43,308 --> 00:00:46,901
also looking at that. Now, if you think
about making that sort of decision, a

13
00:00:46,901 --> 00:00:50,782
house has all sorts of dimensions to it,
right? So there's square feet, there's the

14
00:00:50,782 --> 00:00:54,435
number of bedrooms, there's number of
bathrooms, there's the lot size. There's

15
00:00:54,435 --> 00:00:58,306
the location. There's the condition of the
house. There's all these things. So, If

16
00:00:58,306 --> 00:01:02,121
you sit there and you go through one house
and you go through the other house and you

17
00:01:02,121 --> 00:01:05,825
try to keep all these things in your head,
that can be really, really complicated. So

18
00:01:05,825 --> 00:01:09,803
one way to do multi-criterion decision
making is just to say okay let's just

19
00:01:09,803 --> 00:01:13,834
create a chart like this. Let's create the
following model where these are the

20
00:01:13,834 --> 00:01:17,967
dimensions right here. These are all the
dimensions I'm gonna decide on. And then

21
00:01:17,967 --> 00:01:22,293
for each house, right, house one and house
two, I have a column. So now, what I can

22
00:01:22,293 --> 00:01:27,772
do is I can write down the square footage
of these houses. So I could say this house

23
00:01:27,772 --> 00:01:33,185
is maybe 2,000 square feet and this house
is 1,800 square feet. Right? And number of

24
00:01:33,185 --> 00:01:38,468
bedrooms. I could say maybe this house has
four bedrooms and this house has three

25
00:01:38,468 --> 00:01:43,490
bed, bedrooms. So, house one is sort of
bigger with more bedrooms, but house one

26
00:01:43,490 --> 00:01:48,577
may only have one one-half bathrooms where
as house three, house two has three

27
00:01:48,577 --> 00:01:53,600
bathrooms. Now look at the lot size. It
may be this one is only a quarter acre.

28
00:01:53,600 --> 00:01:58,157
And this one is a half acre. Right? And
then I look at the location, and the

29
00:01:58,157 --> 00:02:02,838
location, now this, maybe, maybe I put
miles from work. Maybe this is ten miles

30
00:02:02,838 --> 00:02:07,765
from work, and maybe this is fifteen miles
from work. And the last I can ask okay

31
00:02:07,765 --> 00:02:13,183
what's the condition of the house? And
maybe this one. Is excellent, and this one

32
00:02:13,183 --> 00:02:17,700
is only good. What I would like
[inaudible] these criteria and say, okay,

33
00:02:17,700 --> 00:02:21,912
which one wins, right? So, what I can do
is I can say, okay, well, on square

34
00:02:21,912 --> 00:02:26,599
footage this one wins, on bedrooms this
one wins, on bathrooms house two wins, on

35
00:02:26,599 --> 00:02:31,345
lot size house two wins, on location, cuz
this is distance to work, house one wins,

36
00:02:31,345 --> 00:02:36,210
and on condition of the house, house one
wins. So what I get is that house one gets

37
00:02:36,210 --> 00:02:40,778
a total of four, and house two gets a
total of two, [cough] so if I can deduce

38
00:02:40,778 --> 00:02:45,346
anything and say, well look, house one
seems like a better house, right? Because

39
00:02:45,346 --> 00:02:50,427
it, it wins on four of the six criteria.
And this is a way, you know, one way to

40
00:02:50,427 --> 00:02:55,404
make decisions. You can do this on other
issues as well. A few years ago the state

41
00:02:55,404 --> 00:02:59,908
of Michigan had a referendum called the
Michigan Civil Rights Initiative, and this

42
00:02:59,908 --> 00:03:04,411
essentially, was a referendum saying that
state employees, you know, state employers

43
00:03:04,411 --> 00:03:08,420
including universities could not give
preferential treatment to women or

44
00:03:08,420 --> 00:03:12,868
under-represented minorities. So what I
did with Liz Suhey, who was a grad student

45
00:03:12,868 --> 00:03:17,372
at the time at the University of Michigan,
was we put together a, a decision theory

46
00:03:17,372 --> 00:03:21,436
guide for people to help make this
decision. So what we did, is for each of

47
00:03:21,436 --> 00:03:25,720
the different dimensions of this issue,
how it was going to affect things like.

48
00:03:25,720 --> 00:03:30,922
Equality, the quality of education, help
or harm women, social cohesion, reward

49
00:03:30,922 --> 00:03:36,398
merit and whether it was an appropriate
limit or not. We wrote a guide that sort

50
00:03:36,398 --> 00:03:40,487
of laid out. The argument for mcri and
against the mcri. And then what a person

51
00:03:40,487 --> 00:03:44,448
could do is they could go down each of
these dimensions and say I'm in favor of

52
00:03:44,448 --> 00:03:48,459
the mcri or I'm against it, right, and so
on and then count up the columns and say

53
00:03:48,459 --> 00:03:52,370
okay there's four reasons why I should
support it, and two reasons why I should

54
00:03:52,370 --> 00:03:56,579
be against it, where alternatively someone
might read through these things and say

55
00:03:56,579 --> 00:04:00,590
well on every one of these dimensions I
think we should reject the mcri and vote

56
00:04:00,590 --> 00:04:04,402
against it. So what this was, is this was
a guide to help people recognize the

57
00:04:04,402 --> 00:04:08,165
dimensions, weight the alternatives along
those two dimensions, and then make

58
00:04:08,165 --> 00:04:12,533
hopefully more informed choices. So if you
think about how you vote or how you buy a

59
00:04:12,533 --> 00:04:16,547
house, this can be really a useful way,
because there are so many dimensions,

60
00:04:16,547 --> 00:04:20,299
right? What you do is just lay out each
dimension. Put columns for the

61
00:04:20,299 --> 00:04:24,590
alternatives, and then decide which
alternative is better on which dimension,

62
00:04:24,590 --> 00:04:29,107
and then add those up and make a choice.
So let's go back to our house case. So I

63
00:04:29,107 --> 00:04:33,624
do this, I put down all the criteria, and
I, you know, list house one and house two,

64
00:04:33,624 --> 00:04:38,141
and then it says, okay, you know, I should
buy, if you're doing this, right, remember

65
00:04:38,141 --> 00:04:42,997
we came down and we had four reasons why I
should buy house one and two reasons why I

66
00:04:42,997 --> 00:04:47,349
should buy house two, and I say, okay,
therefore I'm buying house one. Suppose I

67
00:04:47,349 --> 00:04:51,851
do that and I don't like house one
[laugh]. I've done different houses and I,

68
00:04:51,851 --> 00:04:56,238
in my gut, I say you know what, I don't
like house one. I really like house two.

69
00:04:56,238 --> 00:05:00,797
Well it's a couple things going on. One is
maybe I'm missing some criteria. Maybe

70
00:05:00,797 --> 00:05:05,412
house two is a Spanish style house and
house one is a Craftsman house and what I

71
00:05:05,412 --> 00:05:09,971
really like is Spanish style houses. Well,
if that's the case, right? What I should

72
00:05:09,971 --> 00:05:14,187
do is go back to my criteria and add
Spanish style house. But if I do that,

73
00:05:14,187 --> 00:05:18,595
that's still only going to make it four to
three. Well, what can I do then? Well, one

74
00:05:18,595 --> 00:05:22,562
thing I can do then is I could say, well,
do I really like house two? [laugh] Right,

75
00:05:22,562 --> 00:05:26,283
is it really the case I like two or is
there, is there something [inaudible]

76
00:05:26,283 --> 00:05:30,152
irrational in what I'm doing? Is this, is
it some sort of romantic attachment to

77
00:05:30,152 --> 00:05:34,020
house two, when in fact, when I look at it
somewhat objectively according to the

78
00:05:34,020 --> 00:05:37,888
criteria that I care about, house one is
better? Well, another way we can go, more

79
00:05:37,888 --> 00:05:41,218
sophisticated multi-criterion
decision-making model, is to take into

80
00:05:41,218 --> 00:05:44,646
account quantitative weights, right, to
not only just take into account

81
00:05:44,646 --> 00:05:48,220
qualitatively which is better, but
actually quantitatively, to measure the

82
00:05:48,220 --> 00:05:52,288
differences. So what do I mean? So now I
can say okay, here's all these things I

83
00:05:52,288 --> 00:05:56,618
care about, right? And let's suppose now
that I've added in here that I also care

84
00:05:56,618 --> 00:06:00,948
about the basement. So now, there's house
one and house two and maybe again house

85
00:06:00,948 --> 00:06:05,170
one wins on square feet, number of
bedrooms but it doesn't win on bathrooms

86
00:06:05,170 --> 00:06:09,717
or lot size and it wins on lot location
and condition, but suppose house two wins

87
00:06:09,717 --> 00:06:14,244
on basement, right? And before I had this
total of. Four to three. But it could be

88
00:06:14,244 --> 00:06:19,181
that you know, square footage doesn't
matter to me that much. Nor does number of

89
00:06:19,181 --> 00:06:24,180
bedrooms so go, give each of these one.
But number of bathrooms matters to me two.

90
00:06:24,180 --> 00:06:29,059
And lot size matters to me too. Location
and condition don't matter to me very

91
00:06:29,059 --> 00:06:35,227
much, and having a basement is incredibly
important. Well now. If I do this, house

92
00:06:35,227 --> 00:06:40,889
one gets a total of one, two, three, four.
Right? And house two gets a total of two,

93
00:06:40,889 --> 00:06:45,199
four, six. And so now, it's house two. And
so what I see is, okay, what really was

94
00:06:45,199 --> 00:06:49,509
going on here is that when I put in how
much I care about these dimensions, I

95
00:06:49,509 --> 00:06:54,043
realize that's the reason why in my gut, I
sort of felt that house two was better.

96
00:06:54,043 --> 00:06:58,465
Now again, let's think about this one.
Remember we talked about, we don't want to

97
00:06:58,465 --> 00:07:03,167
let the models tell us what to do. We want
the models to help us make better choices.

98
00:07:03,167 --> 00:07:07,478
So if I'm buying a house and I write down
all the criteria, and I compare two

99
00:07:07,478 --> 00:07:11,545
houses. And I see boy, [laugh], according
to my very, my own criteria, house one is

100
00:07:11,545 --> 00:07:15,628
better than house two, but in my heart of
hearts, I prefer house two. That doesn't

101
00:07:15,628 --> 00:07:19,558
mean we shouldn't buy house two. But it
does mean we should ask ourselves are

102
00:07:19,558 --> 00:07:23,590
there, are there criteria we're leaving
out, or other weights? So we can at least

103
00:07:23,590 --> 00:07:27,674
understand ourselves and know why we're
making the choice. And that, no we're not

104
00:07:27,674 --> 00:07:31,555
making it because you know, perhaps we
liked. The blue curtains in the living

105
00:07:31,555 --> 00:07:35,643
room or something, right? So, it helps us
make more rational choice. So, that's sort

106
00:07:35,643 --> 00:07:39,731
of normatively why these things are so
useful. Positive way too, because when you

107
00:07:39,731 --> 00:07:43,666
see somebody else buy a house. When you
see, boy house one was better on every

108
00:07:43,666 --> 00:07:48,566
dimension except for lot size. We can then
infer, well, this is someone who cares a

109
00:07:48,566 --> 00:07:52,830
lot about lot size. Okay. There we go. So
that's it. Multi-criterion decision

110
00:07:52,830 --> 00:07:57,036
making. What you do is you list all your
criteria, take your alternatives and you

111
00:07:57,036 --> 00:08:00,978
see which one wins in each, on each
criteria. Now you can also add weight to

112
00:08:00,978 --> 00:08:05,308
make it more quantitative so some criteria
matter more than others. Again, two ways

113
00:08:05,308 --> 00:08:09,511
we can use it. One is make better choices
ourselves, right? We just saw how when

114
00:08:09,511 --> 00:08:13,768
we're make, deciding like whether to buy a
house or how to vote this can be real

115
00:08:13,768 --> 00:08:18,206
useful, and second we can use them to
understand the choices of others. Thank