Sie sind auf Seite 1von 25

Expected values,

covariance,
correlation and
expected values
Introduction to Bivariate Regression

Review
Mean
Mode
Median
Freq
Variance
Standard deviation

Is the perception that the


majority of Russians
believe the same way you
do related to how often
you discuss politics with
friends?

Is this a causal
relationship?
Majority of
Russians
believe the
same

Discussions of
politics with
friends

Discussions of
politics with
friends

Majority of
Russians
believe the
same

When it comes to politics, how close do you


think your opinions are to the opinions of the
majority of Russians? very close, rather close,
not very close, not close at all
maj rcl How close your opinions to the opinions of the maj ority of Russians about
politics

Frequency
Valid

1.00 not c lose at


all
2.00 not very close
3.00 rather close
4.00 very c lose
Total

Missing
Total

System

Percent

Valid Percent

Cumulative
Percent

.9

1.1

1.1

74

22.7

28.1

29.3

178

54.6

67.7

97.0

2.5

3.0

100.0

263

80.7

100.0

63

19.3

326

100.0

freq vars = majrcl / stats = mean stddev var.

How often do you do the following discuss political


questions with friends, neighbors, or coworkers
almost never, a few times a year, a few times a month,
a few times a week, or practically every day?
discfrnd How often do you discuss political questions w ith friends, neighbors

Valid

Frequency
78

Perc ent
23.9

Valid Percent
24.6

Cumulative
Perc ent
24.6

2.00 A few times a year

41

12.6

12.9

37.5

3.00 A few times a


month

97

29.8

30.6

68.1

4.00 A few times a week

77

23.6

24.3

92.4

5.00 Practically every


day

24

7.4

7.6

100.0

317

97.2

100.0

8.00 Refuse

.3

9.00 Unsure

2.5

Total

2.8

326

100.0

1.00 Almost never

Total
Missing

Total

freq vars = discfrnd /


stats = mean stddev var.

Review standard deviation and


variance
Variance: for each unit or observation,

it is the distance from the mean


squared and then divide by the number
of units
Standard deviation squareroot of
variance
since variance is in squared units, it
doesnt make any sense. The standard
deviation can be understood in terms of
the original measurement unit

Calculating variance and standard


deviations

Review: Units, mean, variance


and standard deviation
majrcl discfrnd
2.00
2.00
.
.
2.00
3.00
3.00
3.00
.
2.00
.
3.00
3.00
3.00
3.00
3.00

4.00
3.00
4.00
1.00
1.00
4.00
3.00
3.00
3.00
2.00
3.00
3.00
4.00
5.00
1.00
3.00

Descriptiv e Statistics
N

Mean

Std. Deviation

Variance

majrcl How close your


opinions to the opinions of the
majority of Russians about
politics

12

2.6667

.49237

.242

discfrnd How often do you


discuss political questions with
friends, neighbors

16

2.9375

1.18145

1.396

Valid N (listwise)

12

Expected value v. probability


If our population set of numbers is:

1,1,3,3,17, then the expected value is 5, even


though P(5) = 0.
Suppose we know that E(X) = 5 with the
equation y = 5 + 7x.
What is E(Y)?

Expected values
Statistics
majrcl How close your opinions to the opinions
of the majority of Russians about politics
N
Valid
263
Missing
Mean
Std. Deviation
Variance

63
2.7262

What is the expected value of majrcl?


What is the range?
Mode?
Why are there 63 missing?

.53249
.284

Statistics
discfrnd How often do you discuss political questions with friends, neighbors
N
Valid
317
Missing

What is the expected value of


discfrnd?
Why is the standard deviation and
variance so high?

Mean
Std. Deviation
Variance

9
2.7729
1.26996
1.613

Crosstab
maj rcl How close your opinions to the opinions of the maj ority of Russians about politics * discfrnd How often do you discuss political questions w ith friends, neighbors
Crosstabulation
discfrnd How often do you discuss political questions with friends, neighbors

majrcl How c lose your


opinions to the opinions
of the majority of
Russians about politic s

1.00 not c lose at


all

2.00 not very close

Count
% within discfrnd How often do
you discuss political questions
with friends, neighbors
Count
% within discfrnd How often do
you discuss political questions
with friends, neighbors

3.00 rather close

Count
% within discfrnd How often do
you discuss political questions
with friends, neighbors

4.00 very close

Count
% within discfrnd How often do
you discuss political questions
with friends, neighbors

Total

Count
% within discfrnd How often do
you discuss political questions
with friends, neighbors

1.00 Almost never


1

2.00 A few
times a year
0

3.00 A few
times a month
1

4.00 A few 5.00 Prac tically


times a week
every day
1
0

1.6%

.0%

1.3%

1.4%

.0%

1.2%

28

21

14

73

44.4%

34.8%

27.3%

19.2%

9.5%

28.4%

33

15

52

56

17

173

52.4%

65.2%

67.5%

76.7%

81.0%

67.3%

1.6%

.0%

3.9%

2.7%

9.5%

3.1%

63

23

77

73

21

257

100.0%

100.0%

100.0%

100.0%

100.0%

100.0%

Total
3

Causation
Time ordering
Covariation

Co-variation from
variation?
(xi - xmean)^2/n

average distance between the


mean of x and each x value,
squared
aka (xi - xmean) (xi - xmean)/n

Covariation?

(xi - xmean) * (yi - ymean) / n-1

Covariation
covariance can take any

value
negative infinity to
positive infinity

Intuitive explanation
(xi - xmean) * (yi - ymean) / n-1
When x and y are high at the same time

and x and y are low at the same time,


then the covariance is positive
They are both higher than their means
and so the products being added
together are positive

Plot showing positive


covariance
Mean urban %

Mean female literacy

Intuitive explanation
(xi - xmean) * (yi - ymean) / n-1
When x is low when y is high and vice

versa, then the covariance is negative


They are both higher than their means
and so the products being added
together are negative

Plot showing negative


covariance
Mean calorie intake

Mean infant mortality

Intuitive explanation
(xi - xmean) * (yi - ymean) / n
When sometimes:
x and y are high at the same time and x and y

are low at the same time


And about half of the other time
x is low when y is high and vice versa
Then the covariance is about 0
High positive numbers are added to high

negative numbers

Plot showing no covariance


Mean GDP

Mean crop
production

Covariance is a function
of
Variance (standard deviation) of x
Variance (standard deviation) of y
Relationship between x and y

How can you compare a


covariance of 132 and

134,847?
134, 847 could be high variance of x,
high variance of y, high variance of both
variables, or a high relationship
between x and y?
Not that helpful?

How can you change the covariance to a number


that tells you only the magnitude of the
relationship between x and y?
Divide by the standard deviation of x * the

standard deviation of y

Correlation = (x-xmean)*(y-ymean) /Sd(x) *

sd (y)

Pearson r ranges from -1 to +1


Weak correlation = .1
moderate correlation = .4
strong correlation = .7

Das könnte Ihnen auch gefallen