Download as pdf or txt
Download as pdf or txt
You are on page 1of 108

Quantitative

Research
By
Example
Version 1.0.0 {β}

Ioan Gelu Ionas


Ph.D, MBA

My
Research
MRL
BETA

Lab
V ERSION : 1.0.0 {β}
version : 1.0.0 {β}
version : 1.0.0 {β}
version : 1.0.0 {β}

β
version : 1.0.0 {β}
version : 1.0.0 {β}
version : 1.0.0 {β}
version : 1.0.0 {β}
version : 1.0.0 {β}
ω2

ω2

ω2
|means1 − means2 |
d=
SD
version : 1.0.0 {β}

Effect Power
Size
!ω 2 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90
ω2
!α = 0.05

0.01 21 53 83 113 144 179 219 271 354

0.06 5 10 14 19 24 30 36 44 57

0.15 3 5 6 8 10 12 14 17 22

!α = 0.01

0.01 70 116 156 194 232 274 323 385 478

0.06 13 20 26 32 38 45 53 62 77

0.15 6 8 11 13 15 18 20 24 29
α

α = 0.6
ω2
version : 1.0.0 {β}

ω2
α
α
z 2 · p · (1 − p)
n0 =
e2

1.962 · 0.5 · (1 − 0.5)


n0 = = 384.16 ≈ 385
0.052
version : 1.0.0 {β}

n0
n=
n0 − 1
1+
N

385
n= = 234.76 ≈ 235
385 − 1
1+
600

N
n=
1 + N · e2
600
n= = 240
1 + 600 · 0.052
Density

0.0 0.1 0.2 0.3 0.4

−4
−2
0
2
4
Frequency

0 5 10 20

0.0
0.2
0.4
0.6
Frequency

0 5 10 20

0.2 0.4 0.6 0.8


Frequency

0 5 10 20

0.2
0.6
1.0
version : 1.0.0 {β}
Left−skewed distribution

25
0.6
Sample Quantiles

20
Frequency
0.4

15
10
0.2

5
0
0.0

−2 −1 0 1 2 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

Theoretical Quantiles

Not skewed distribution


0.8

25
Sample Quantiles

20
0.6

Frequency

15
10
0.4

5
0.2

−2 −1 0 1 2 0.2 0.4 0.6 0.8

Theoretical Quantiles

Right−skewed distribution
1.0

25
Sample Quantiles

0.8

20
Frequency

15
0.6

10
0.4

5
0.2

−2 −1 0 1 2 0.2 0.4 0.6 0.8 1.0

Theoretical Quantiles
version : 1.0.0 {β}
No outliers Outliers in dataset qq−plot

2.0
50

Sample Quantiles

1.5
20
Frequency

Frequency

30

1.0
5 10

0.5
0 10
0

0.2 0.4 0.6 0.8 0.0 0.5 1.0 1.5 2.0 −2 0 1 2

Theoretical Quantiles
version : 1.0.0 {β}

Scatterplot Box−plot
2.0

2.0
1.5

1.5
my.ns.o

1.0

1.0
0.5

0.5
0 20 40 60 80 100

Index

No outliers One outlier


2.0

2.0
1.5

1.5
1.0

1.0
y

y
0.5

0.5
0.0

0.0

0 20 40 60 80 100 0 20 40 60 80 100 140

x x
y = a×x+b
µ σ
25

20
20

15
Frequency

Frequency
15

10
10

5
5
0

0.2 0.4 0.6 0.8 −3 −2 −1 0 1 2

Original data z−scores


version : 1.0.0 {β}

−π/2 π/2
Histogram Normal Q−Q Plot
300

120
Sample Quantiles
Frequency

200

80
100

40
0

0
0 20 40 60 80 100 140 −3 −2 −1 0 1 2 3

Data Theoretical Quantiles

Histogram Normal Q−Q Plot


0.4

5
Sample Quantiles

4
0.3
Density

3
0.2

2
0.1

1
0.0

−2 0 2 4 −3 −2 −1 0 1 2 3

Log transformed data Theoretical Quantiles


& τ
version : 1.0.0 {β}

30
Miles/Gallon
20 10

2 3 4 5
Weight
χ2 (df = 1) = 573.48, p < 0.001
version : 1.0.0 {β}
0 1

F
Gender

Died in hospital

χ2 (df = 2) = 283.43, p < 0.001


version : 1.0.0 {β}

F M
121
Gender

122
123

Code
version : 1.0.0 {β}

Female Heights Male Heights

0.15
0.15

0.10
0.10
Density

Density

0.05
0.05
0.00

0.00

60 62 64 66 68 70 72 65 70 75 80

Height Height

Females Males
72

78
70

76
68
Sample Quantiles

Sample Quantiles

74
66

72
70
64

68
62

66
60

−2 −1 0 1 2 −2 −1 0 1 2

Theoretical Quantiles Theoretical Quantiles


version : 1.0.0 {β}
M Di2
chisqp

1.0
48 123 99 42
6

3
149 48
99 74
81
97
11
102
464
81 97 Cumulative probability 103
165886
0.8
34 46 80 143
459
36 54 5
4

91 24
20
192330
31
32 55 74
96
41
7
50
80
90
92
33 59 69
70
72 86 102 66
22
34
6
3 14 20
21 91
36
105
68 8
0.6
72
2

94 83
31
30
9
8292 78
14
12 5760
49
87 103
83 69
55
101
54
14 5 7 9 26
25
37 505663 104
32
33
23
18
38 516164
656773 84 95
93
29
0

71 104 10
12
0.4

100 100
76
79
35
4247 89 98
75 85
88 38

97.5% Quantile
10 1517 27 39 52 53
94
56
98
28 40 6266 77 95 105 15
26
25
84
101 89
−2

24 78 88
21
44
11 18 2935 68 79 93 82
85
59
17
0.2

43
44 76 77
27
51
70
22 53 90 96 3968
87
28
4145 64
60
40
57
−4

13 37
75
62
16 73
52
71
47
2 65
67
0.0

58 61
63

−30 −20 −10 0 10 20 0 5 10 15


Ordered squared robust distance

Outliers based on 97.5% quantile Outliers based on adjusted quantile


48 48
6

99 99
34 46 81 97 34 46 81 97
36 54 80 36 54 80
4

91 91
192331
30
32 55 74 192330
31
32 55 74
33 59 69
70
72 86 102 33 59 69
70
72 86 102
3 14 20
21 3 14 20
21
68 68
2

94
8292 94
8292
12 5760
49
87 103
83 12 49 60
5763 87 103
83
14 5 7 9 26
25
37 505663 14 5 7 9 26
25
37 5056
38 516164
656773 84 38 516164
656773 84
0

71 104
100 71 104
100
4247 89 98
75 85
88 4247 89 98
75 85
88
10 1517 27 39 52 95 105 10 1517 27 39 52 95 105
28 40 6266 77 101 28 40 6266 77 101
−2

−2

24
11 18 2935 78 24
11 18 2935 78
43
44 68 79 93 43
44 68 79 93
76 76
22 53 90 96 22 53 90 96
4145 4145
−4

−4

13 13
2 16 2 16
58 58

−30 −20 −10 0 10 20 −30 −20 −10 0 10 20


version : 1.0.0 {β}
version : 1.0.0 {β}

Punish by Ethnicity
African−American Hispanic Caucasian
Sample Quantiles

80

60

40

20
−2 −1 0 1 2 −2 −1 0 1 2 −2 −1 0 1 2
Theoretical Quantiles

Repeats by Ethnicity
African−American Hispanic Caucasian
Sample Quantiles

28

24

20

−2 −1 0 1 2 −2 −1 0 1 2 −2 −1 0 1 2
Theoretical Quantiles

Dispos by Ethnicity
African−American Hispanic Caucasian
32
Sample Quantiles

28

24

20

−2 −1 0 1 2 −2 −1 0 1 2 −2 −1 0 1 2
Theoretical Quantiles
Punish by Ethnicity
African−American Hispanic Caucasian Count
12
0.04

9
0.03
density

0.02 6

0.01
3
0.00
40 60 80 40 60 80 40 60 80
0
punish

Repeats by Ethnicity
African−American Hispanic Caucasian Count
0.3 12.5

10.0
0.2
density

7.5

0.1
5.0

2.5
0.0
21 24 27 21 24 27 21 24 27
0.0
repeats

Dispos by Ethnicity
African−American Hispanic Caucasian Count
10.0

0.2 7.5
density

5.0
0.1

2.5
0.0
21 24 27 21 24 27 21 24 27
0.0
dispos
version : 1.0.0 {β}
η2 η2

η 2 = .145(> .14)
version : 1.0.0 {β}

η2

η 2 = .096
Y = a + bX

Y X
a b y X =0
Y

Y = a + b1 X + b2 X + ... + bn X

bi

Y = a + (b1 + b2 + ... + bn )X

b = b1 + b2 + ... + bn
version : 1.0.0 {β}
version : 1.0.0 {β}
BuyRaw! = b0 + b1 · P rice + b2 · Quality + b3 · T aste
+b4 · LocGrown + b5 · P repEase + b6 · N utrition

±3
version : 1.0.0 {β}

2
1
Standardized residuals

0
−1
−2
−3

0 50 100 150

Case number
80
60
Frequency

40
20
0

0 10 20 30 40 50

Distances

5 110
Cook's distance

26
0.06
0.00

0 50 100 150

Obs. number
version : 1.0.0 {β}

Deleted residuals

1
−1
−3

−8 −6 −4 −2 0 2 4 6

Raw residuals
Standardized Residuals

No of observations
1 2

40
20
−1
−3

−3 −1 1 2 3 −3 −1 1 2

Normal Scores Residuals


5
Residuals

0
−5

10823
5

4 6 8 10 12

Fitted values

BuyRaw! = b0 + b1 · P rice + b2 · Quality + b3 · T aste


+b4 · LocGrown + b5 · P repEase + b6 · N utrition
version : 1.0.0 {β}

4 8 12 4 8 12 4 8 12

0.5 0.48 0.55 0.47

4 10
Price 0.34

Quality 0.91 0.59 0.62 0.7


4 10

0.59 0.65 0.7

4 10
Taste

0.63 0.67
4 10

LocGrown

PrepEase 0.77

4 10
4 10

Nutrition

4 8 12 4 8 12 4 8 12
version : 1.0.0 {β}

BuyRaw! = 3.9396 − 0.0607 · P rice + 0.4736 · Quality − 0.0404 · T aste


+0.0909 · LocGrown − 0.1725 · P repEase + 0.1629 · N utrition

β
version : 1.0.0 {β}
logit(P ) = a + b · X

logit(p) = β0 + β1 · X1 + β2 · X2 + ... + βn · Xn

Xi βi

P
logit(p) = log(odds) = ln( )
1−P
P
ln( 1−P = a + bX)
P a+bX
1−P = e
ea+bX
P = 1+ea+bX
version : 1.0.0 {β}
version : 1.0.0 {β}

logit(p) = β0 + β1 · group + β2 · year + β3 · group × year

group × year
version : 1.0.0 {β}

logit(p) = β0 + β1 · group + β2 · year + β3 · group × year

logit(p) = −1.264 + 0.6084 · group + 0.9836 · year


β

group odds ratio = e0.6084 = 1.84

year odds ratio = e0.9836 = 2.67


version : 1.0.0 {β}

logit(p) = β0 + β1 · group + β2 · pk + β3 · group × pk + β4 · year

group × pk

logit(p) = β0 + β2 · pk + β4 · year
logit(p) = β0 + β1 + β2 · pk + β3 · pk + β4 · year

logit(p) = (β0 + β1 ) + (β2 + β3 ) · pk + β4 · year


version : 1.0.0 {β}

You might also like