HARAMAYA UNIVERSITY

COLLEGE OF VETERINARY MEDICINE

Individual assignment for the course Applied biostatistics to post-graduate students (60%)
Note: (weight consideration, 30% document and 30% presentation)

Please try to do the following events accordingly and prepare a good report to your work. If it is
possible, use EndNote/Mendlay or Zotero reference management software for your references
depending on your preference.

I. Part one
 Consider the lungworm data set given (sheet 2) and

1. Summarize for each variable by using appropriate graphs or diagrams accordingly.

Describe the data by:

 Different appropriate DATA presentation methods (tabular and diagrammatic) for each
variable.

 Frequency Table

Sex

sex Frequency

male 197

female 181

Total 378
age

age Frequency

young 180

adult 198

Total 378

body condition

Body condition Frequency

good 87

medium 102

poor 189

Total 378

study site

Study site Frequency

adele 96

haramaya 164

awoday 118

Total 378

resuts

result Frequency

negative 248

positive 130

Total 378
Bar Chart
Pie Chart
  Different appropriate numerical measures (proportion) with interval estimates for each
variable.

 sex * resuts

resuts

negative positive Total

sex male Count 151 46 197

% within sex 76.6% 23.4% 100.0%

female Count 97 84 181

% within sex 53.6% 46.4% 100.0%

Total Count 248 130 378

% within sex 65.6% 34.4% 100.0%

 age * resuts

resuts

negative positive Total

age young Count 102 78 180

% within age 56.7% 43.3% 100.0%

adult Count 146 52 198

% within age 73.7% 26.3% 100.0%

Total Count 248 130 378

% within age 65.6% 34.4% 100.0%

 body condition * resuts

 resuts

negative positive Total

body condition good Count 70 17 87

% within body condition 80.5% 19.5% 100.0%

medium Count 76 26 102

% within body condition 74.5% 25.5% 100.0%

poor Count 102 87 189

% within body condition 54.0% 46.0% 100.0%

Total Count 248 130 378

% within body condition 65.6% 34.4% 100.0%

 study site * resuts

resuts

negative positive Total

study site adele Count 58 38 96

% within study site 60.4% 39.6% 100.0%

haramaya Count 108 56 164

% within study site 65.9% 34.1% 100.0%

awoday Count 82 36 118

% within study site 69.5% 30.5% 100.0%

Total Count 248 130 378

% within study site 65.6% 34.4% 100.0%

Is there any significance difference for the presence of lungworm in goat between different sex
and body condition? NO, because P<0.05 (0.000) for all and it is statistically significant

 Develop logistic regression model and check for its fitness lungworm data set.

 Case Processing Summary

Marginal
N Percentage

resuts negative 248 65.6%

positive 130 34.4%

sex male 197 52.1%

female 181 47.9%

age young 180 47.6%

adult 198 52.4%

body condition good 87 23.0%

medium 102 27.0%

poor 189 50.0%

study site adele 96 25.4%

haramaya 164 43.4%

awoday 118 31.2%

Valid 378 100.0%

Missing 0

Total 378

Subpopulation 36a
a. The dependent variable has only one value observed in 6 (16.7%)
subpopulations.

Model Fitting Information

Model Fitting
Criteria Likelihood Ratio Tests

Model -2 Log Likelihood Chi-Square df Sig.

Intercept Only 175.740

Final 118.745 56.995 6 .000

Goodness-of-Fit

Chi-Square df Sig.

Pearson 38.857 29 .104

Deviance 42.840 29 .047

Pseudo R-Square

Cox and Snell .140

Nagelkerke .193

McFadden .117

Likelihood Ratio Tests

 Model Fitting
Criteria Likelihood Ratio Tests

-2 Log Likelihood
Effect of Reduced Model Chi-Square df Sig.

Intercept 118.745a .000 0 .

sex 136.324 17.579 1 .000

age 132.017 13.272 1 .000

body condition 140.366 21.621 2 .000

study site 120.238 1.493 2 .474

The chi-square statistic is the difference in -2 log-likelihoods between the final

model and a reduced model. The reduced model is formed by omitting an
effect from the final model. The null hypothesis is that all parameters of that
effect are 0.

a. This reduced model is equivalent to the final model because omitting the
effect does not increase the degrees of freedom.

Parameter Estimates
95% Confidence Interval for Exp(B)
a
resuts B Std. Error Wald df Sig. Exp(B) Lower Bound Upper Bound
negative Intercept .22 .287 .590 1 .443
1
[sex=1] .97 .237 17.033 1 .000 2.657 1.670 4.225
7
[sex=2] 0b . . 0 . . . .
[age=1] -.85 .237 12.886 1 .000 .427 .268 .679
2
[age=2] 0b . . 0 . . . .
[body condition =1] 1.1 .321 13.753 1 .000 3.293 1.754 6.181
92
[body condition =2] .99 .285 12.201 1 .000 2.707 1.548 4.733
6
[body condition =3] 0b . . 0 . . . .
 [study site=1] -.34 .312 1.202 1 .273 .710 .385 1.309
2
[study site=2] -.03 .281 .016 1 .901 .966 .557 1.673
5
[study site=3] 0b . . 0 . . . .
a. The reference category is: positive.
b. This parameter is set to zero because it is redundant.

Consider the production data set (sheet 1) collected from Maya city and

Describe and interpret the data by;

 Different appropriate DATA presentation methods (tabular and diagrammatic/graphic

methods for each variable in the data set) and interpret.

Frequency Table

sample code(Haramaya=1,awaday=2,adele=3)

Sample code Frequency

haramaya 50

awaday 34

adele 36

Total 120
 BCS(good=3,medium=2,poor=1)

BCS Frequency

poor 14

medium 61

good 45

Total 120

season of birth(automn=1,summer=2,winter=3,spring=4)

Season of birth Frequency

automn 54

summer 44

winter 9

spring 13

Total 120

number of parity(first parity=1,second parity=2,third parity=3 fourth parity=4

fifthyparity=5,above6=6)

number of parity Frequency

first parity 15

second parity 25

third parity 27

fourth parity 31

fifth parity 20
 above 6 2

Total 120

age in month- N.B- this table do by continuous frequency table

age in month Frequency

30 2

31 1

32 2

33 1

34 4

35 1

36 7

38 1

42 3

44 7

45 3

46 1

47 2

48 8

49 2

50 2

51 1

52 3

54 2

55 2
 56 1

58 2

60 14

62 1

64 4

66 2

68 3

70 3

72 9

74 1

76 2

78 2

80 1

84 6

88 1

90 2

92 1

94 2

96 6

120 2

Total 120

daily milk yeild in litter- N.B- this table also do by continuous

frequency distribution table by class limit, class boundry and so on
daily milk yeild
in litter Frequency
9.0 2

9.2 1

9.5 1

10.0 5

10.2 2

10.6 2

10.8 1

11.0 4

11.3 1

11.5 1

12.0 9

12.4 2

12.5 3

12.8 1

13.0 4

13.2 3

13.5 3

14.0 5

14.5 7

14.8 1

15.0 3

15.5 4

15.6 1

15.8 3

16.0 7

16.2 1
 16.5 4

16.8 1

17.0 6

17.4 1

17.5 2

18.0 8

18.5 2

19.0 5

19.5 4

20.0 3

20.5 1

21.0 2

22.0 3

23.0 1

Total 120

calving interval in month-

N.B- this table also do by continuous frequency distribution table by

class limit, class boundry and so on
calving interval
in month Frequency

Valid 12 6

13 25

14 20

15 30

16 13
 17 2

18 17

20 7

Total 120

Bar Chart
Pie Chart
Histogram
  Different appropriate numerical measures (measure of central tendency and dispersions)
with interval estimates for each variable the data set.

Statistics
number of
parity(first
parity=1,sec
sample ond
code(H parity=2,third
aramay BCS(go parity=3
a=1,aw od=3,m season of fourth daily
aday=2 edium= birth(automn=1 parity=4 milk
,adele= 2,poor= ,summer=2,win fifthyparity=5 age in yeild in calving interval in
3) 1) ter=3,spring=4) ,above6=6) month litter month
N Valid 120 120 120 120 120 120 120
Missing 0 0 0 0 0 0 0
Mean 1.88 2.26 1.84 3.18 60.16 15.069 15.13
Std. Error of Mean .077 .060 .089 .121 1.810 .3043 .191
Median 2.00 2.00 2.00 3.00 60.00 15.000 15.00
Mode 1 2 1 4 60 12.0 15
Std. Deviation .842 .655 .970 1.328 19.832 3.3331 2.093
Variance .709 .429 .941 1.764 393.294 11.110 4.379
Range 2 2 3 5 90 14.0 8
Minimum 1 1 1 1 30 9.0 12
Maximum 3 3 4 6 120 23.0 20
Sum 226 271 221 382 7219 1808.3 1815

 Compare the average daily milk yield in cattle between Adele and Haramaya sub city?

 T-Test
 Group Statistics
sample
code(Haramaya=1,awaday=
2,adele=3) N Mean Std. Deviation Std. Error Mean
daily milk yeild in litter haramaya 50 14.538 2.7982 .3957
adele 36 12.914 2.4398 .4066

Independent Samples Test

Levene's Test
for Equality of
Variances t-test for Equality of Means
Mean 95% Confidence Interval of the
Sig. (2- Differen Std. Error Difference
F Sig. t df tailed) ce Difference Lower Upper
daily milk Equal variances 1.269 .263 2.79 84 .006 1.6241 .5803 .4701 2.7781
yeild in litter assumed 9
Equal variances 2.86 80 .005 1.6241 .5674 .4951 2.7531
not assumed 2 .8
74

Summary Data

N Mean Std. Deviation Std. Error Mean

Sample 1 50.000 14.538 2.798 .396

Sample 2 36.000 12.914 2.440 .407

Independent Samples Test

Std. Error
Mean Difference Difference t df Sig. (2-tailed)

Equal variances assumed 1.624 .580 2.799 84.000 .006

Equal variances not assumed 1.624 .567 2.862 80.874 .005

Hartley test for equal variance: F = 1.315, Sig. = 0.1957

95.0% Confidence Intervals for Difference

Lower Limit Upper Limit

Asymptotic (equal variance) .487 2.761

Asymptotic (unequal variance) .512 2.736

Exact (equal variance) .470 2.778

Exact (unequal variance) .495 2.753

 Compare the average daily milk yield between Adele, Awoday and Haramaya with
appropriate post-hoc test?

 daily milk yeild in litter

95% Confidence Interval for

Mean
Minim
N Mean Std. Deviation Std. Error Lower Bound Upper Bound um Maximum

haramaya 50 14.538 2.7982 .3957 13.743 15.333 9.0 19.5

awaday 34 18.132 2.6380 .4524 17.212 19.053 13.5 23.0

adele 36 12.914 2.4398 .4066 12.088 13.739 9.0 17.5

Total 120 15.069 3.3331 .3043 14.467 15.672 9.0 23.0

ANOVA

daily milk yeild in litter

 Sum of Squares df Mean Square F Sig.

Between Groups 500.361 2 250.180 35.624 .000

Within Groups 821.675 117 7.023

Total 1322.036 119

Post Hoc Tests

Multiple Comparisons

Dependent Variable: daily milk yeild in litter

Bonferroni

(I) sample (J) sample 95% Confidence Interval

code(Hara code(Haram
maya=1,aw aya=1,awad Mean
aday=2,ade ay=2,adele= Differen Upper
le=3) 3) ce (I-J) Std. Error Sig. Lower Bound Bound

haramaya awaday -3.5944* .5891 .000 -5.025 -2.164

adele 1.6241* .5793 .018 .217 3.031

awaday haramaya 3.5944* .5891 .000 2.164 5.025

adele 5.2185* .6337 .000 3.679 6.758

adele haramaya -1.6241* .5793 .018 -3.031 -.217

awaday -5.2185* .6337 .000 -6.758 -3.679

*. The mean difference is significant at the 0.05 level.

  Compare and interpret the average daily milk yield of cattle in Maya city if the average
milk yield in Ethiopia is 15 litter per day.

 Is there any difference in daily milk yield between various study sites? Do also post hoc
tests (multiple comparisons).

 daily milk yeild in litter

95% Confidence
Interval for Mean

U
p
p
e
r
Betwee
B n-
o Compon
Std. u ent
Deviat n Maximu Varianc
N Mean ion Std. Error Lower Bound d Minimum m e

haramaya 50 14.538 2.798 .3957 13.743 1 9.0 19.5

2 5
.
3
3
3

awaday 34 18.132 2.638 .4524 17.212 1 13.5 23.0

0 9
.
0
5
3
adele 36 12.914 2.439 .4066 12.088 1 9.0 17.5
8 3
.
7
3
9

Total 120 15.069 3.333 .3043 14.467 1 9.0 23.0

1 5
.
6
7
2

Model Fixed 2.650 .2419 14.590 1

Effects 1 5
.
5
4
8

Rando 1.4774 8.713 2 6.1767

m 1
Effects .
4
2
6

Test of Homogeneity of Variances

Levene Statistic df1 df2 Sig.

daily milk yeild in litter Based on Mean .637 2 117 .531

Based on Median .644 2 117 .527

Based on Median and with .644 2 116.710 .527

adjusted df

Based on trimmed mean .642 2 117 .528

 ANOVA

daily milk yeild in litter

Sum of Squares df Mean Square F Sig.

Between Groups 500.361 2 250.180 35.624 .000

Within Groups 821.675 117 7.023

Total 1322.036 119

Post Hoc Tests

Multiple Comparisons

Dependent Variable: daily milk yeild in litter

95% Confidence
Interval
(I) sample
code(Haramaya=1 (J) sample Mean Upper
,awaday=2,adele= code(Haramaya=1,awad Difference (I- Boun
3) ay=2,adele=3) J) Std. Error Sig. Lower Bound d

Tukey HSD haramaya awaday -3.5944* .5891 .000 -4.993 -2.196

adele 1.6241* .5793 .016 .249 2.999

awaday haramaya 3.5944* .5891 .000 2.196 4.993

adele 5.2185* .6337 .000 3.714 6.723

 adele haramaya -1.6241* .5793 .016 -2.999 -.249

awaday -5.2185* .6337 .000 -6.723 -3.714

LSD haramaya awaday -3.5944* .5891 .000 -4.761 -2.428

adele 1.6241* .5793 .006 .477 2.771

awaday haramaya 3.5944* .5891 .000 2.428 4.761

adele 5.2185* .6337 .000 3.963 6.474

adele haramaya -1.6241* .5793 .006 -2.771 -.477

awaday -5.2185* .6337 .000 -6.474 -3.963

Bonferroni haramaya awaday -3.5944* .5891 .000 -5.025 -2.164

adele 1.6241* .5793 .018 .217 3.031

awaday haramaya 3.5944* .5891 .000 2.164 5.025

adele 5.2185* .6337 .000 3.679 6.758

adele haramaya -1.6241* .5793 .018 -3.031 -.217

awaday -5.2185* .6337 .000 -6.758 -3.679

*. The mean difference is significant at the 0.05 level.

Homogeneous Subsets

daily milk yeild in litter

sample Subset for alpha = 0.05

code(Haramaya=1,awaday=2,ad
ele=3) N 1 2 3
 Tukey HSDa,b adele 36 12.914

haramaya 50 14.538

awaday 34 18.132

Sig. 1.000 1.000 1.000

Means for groups in homogeneous subsets are displayed.

a. Uses Harmonic Mean Sample Size = 38.865.

b. The group sizes are unequal. The harmonic mean of the group sizes is used. Type I error levels are not
guaranteed.

 Is there any difference in calving interval between various study sites? Do also post hoc
tests (multiple comparisons).

 Descriptives

calving interval in month

95% Confidence Interval for

Between-
Mean
Compone
Upper Maximu nt
N Mean Std. Deviation Std. Error Lower Bound Bound Minimum m Variance

haramaya 50 14.64 1.770 .250 14.14 15.14 13 20

awaday 34 14.21 1.919 .329 13.54 14.88 12 20

adele 36 16.67 1.852 .309 16.04 17.29 14 20

Total 120 15.13 2.093 .191 14.75 15.50 12 20

Model Fixed 1.838 .168 14.79 15.46

Effects

Random .741 11.94 18.31 1.515

Effects
 Test of Homogeneity of Variances

Levene Statistic df1 df2 Sig.

calving interval in month Based on Mean .583 2 117 .560

Based on Median .496 2 117 .610

Based on Median and with .496 2 115.217 .610

adjusted df

Based on trimmed mean .616 2 117 .542

ANOVA

calving interval in month

Sum of Squares df Mean Square F Sig.

Between Groups 126.046 2 63.023 18.664 .000

Within Groups 395.079 117 3.377

Total 521.125 119

Post Hoc Tests

Multiple Comparisons

Dependent Variable: calving interval in month

(I) sample (J) sample Mean Std. Error Sig. 95% Confidence Interval
 code(Haramaya=1,awa code(Haramaya=1,aw Difference (I- Lower
day=2,adele=3) aday=2,adele=3) J) Bound Upper Bound

Tukey HSD haramaya awaday .434 .408 .539 -.54 1.40

adele -2.027* .402 .000 -2.98 -1.07

awaday haramaya -.434 .408 .539 -1.40 .54

adele -2.461* .439 .000 -3.50 -1.42

adele haramaya 2.027* .402 .000 1.07 2.98

awaday 2.461* .439 .000 1.42 3.50

LSD haramaya awaday .434 .408 .290 -.37 1.24

adele -2.027* .402 .000 -2.82 -1.23

awaday haramaya -.434 .408 .290 -1.24 .37

adele -2.461* .439 .000 -3.33 -1.59

adele haramaya 2.027* .402 .000 1.23 2.82

awaday 2.461* .439 .000 1.59 3.33

Bonferroni haramaya awaday .434 .408 .870 -.56 1.43

adele -2.027* .402 .000 -3.00 -1.05

awaday haramaya -.434 .408 .870 -1.43 .56

adele -2.461* .439 .000 -3.53 -1.39

adele haramaya 2.027* .402 .000 1.05 3.00

awaday 2.461* .439 .000 1.39 3.53

*. The mean difference is significant at the 0.05 level.

Homogeneous Subsets
 calving interval in month

sample Subset for alpha = 0.05

code(Haramaya=1,awaday=2,ad
ele=3) N 1 2

Tukey HSDa,b awaday 34 14.21

haramaya 50 14.64

adele 36 16.67

Sig. .552 1.000

Means for groups in homogeneous subsets are displayed.

a. Uses Harmonic Mean Sample Size = 38.865.

b. The group sizes are unequal. The harmonic mean of the group sizes is used. Type I error levels
are not guaranteed.

 Try to develop full linear regression model and check for all assumptions and model
fitness by considering calving interval as your outcome variable for production data set.

 Regression


Descriptive Statistics
Mean Std. Deviation N
calving interval in month 15.13 2.093 120
 sample 1.88 .842 120
code(Haramaya=1,awaday=2,a
dele=3)
BCS(good=3,medium=2,poor=1 2.26 .655 120
)
season of 1.84 .970 120
birth(automn=1,summer=2,wint
er=3,spring=4)
number of parity(first 3.18 1.328 120
parity=1,second parity=2,third
parity=3 fourth parity=4
fifthyparity=5,above6=6)
age in month 60.16 19.832 120
daily milk yeild in litter 15.069 3.3331 120

Model Summary
Change Statistics
Mod Adjusted R Std. Error of R Square Sig. F
el R R Square Square the Estimate Change F Change df1 df2 Change
a
1 .390 .152 .107 1.977 .152 3.378 6 113 .004
a. Predictors: (Constant), daily milk yeild in litter, number of parity(first parity=1,second parity=2,third parity=3 fourth parity=4
fifthyparity=5,above6=6), BCS(good=3,medium=2,poor=1), sample code(Haramaya=1,awaday=2,adele=3), season of
birth(automn=1,summer=2,winter=3,spring=4), age in month



ANOVAa
Model Sum of Squares df Mean Square F Sig.
1 Regression 79.253 6 13.209 3.378 .004b
Residual 441.872 113 3.910
Total 521.125 119
a. Dependent Variable: calving interval in month
b. Predictors: (Constant), daily milk yeild in litter, number of parity(first parity=1,second parity=2,third
parity=3 fourth parity=4 fifthyparity=5,above6=6), BCS(good=3,medium=2,poor=1), sample
code(Haramaya=1,awaday=2,adele=3), season of birth(automn=1,summer=2,winter=3,spring=4), age in
month



Coefficientsa
 Standardize
Unstandardized d 95.0% Confidence Interval
Coefficients Coefficients t Sig. for B Collinearity Statistics
Lower Toleranc
Model B Std. Error Beta Bound Upper Bound e VIF
1 (Constant) 13.821 1.720 8.035 .000 10.413 17.228
sample .893 .233 .359 3.825 .000 .430 1.355 .851 1.175
code(Haramaya
=1,awaday=2,a
dele=3)
BCS(good=3,m .169 .316 .053 .535 .594 -.457 .795 .768 1.302
edium=2,poor=1
)
season of .043 .208 .020 .205 .838 -.369 .455 .808 1.238
birth(automn=1,
summer=2,wint
er=3,spring=4)
number of -.001 .197 .000 -.004 .997 -.390 .389 .482 2.076
parity(first
parity=1,second
parity=2,third
parity=3 fourth
parity=4
fifthyparity=5,ab
ove6=6)
age in month -.005 .014 -.045 -.332 .741 -.033 .024 .399 2.506
daily milk yeild -.036 .057 -.058 -.632 .528 -.150 .077 .900 1.112
in litter
a. Dependent Variable: calving interval in month



Collinearity Diagnosticsa
Mod Dimen Eigen Condition Variance Proportions
 number of
parity(first
parity=1,seco
nd
parity=2,third
sample season of parity=3
code(Hara birth(automn fourth
maya=1,aw BCS(good= =1,summer= parity=4
aday=2,ade 3,medium= 2,winter=3,s fifthyparity=5, daily milk yeild
el sion value Index (Constant) le=3) 2,poor=1) pring=4) above6=6) age in month in litter
1 1 6.348 1.000 .00 .00 .00 .00 .00 .00 .00
2 .261 4.930 .00 .14 .04 .31 .01 .01 .00
3 .164 6.225 .00 .09 .00 .37 .16 .04 .00
4 .131 6.966 .00 .52 .09 .00 .04 .00 .06
5 .051 11.174 .00 .02 .52 .10 .01 .00 .40
6 .037 13.061 .01 .01 .00 .11 .66 .54 .06
7 .008 27.618 .98 .22 .35 .10 .11 .40 .47
a. Dependent Variable: calving interval in month

 Try to develop full Poisson regression model and check for all assumptions and model
fitness by considering number of parity as your outcome variable for production data
set.

Case Processing Summary

Marginal
N Percentage
number of parity(first first parity 15 12.5%
parity=1,second parity=2,third second parity 25 20.8%
parity=3 fourth parity=4 third parity 27 22.5%
fifthyparity=5,above6=6) fourth parity 31 25.8%
fifth parity 20 16.7%
above 6 2 1.7%
sample haramaya 50 41.7%
code(Haramaya=1,awaday=2,a awaday 34 28.3%
dele=3) adele 36 30.0%
BCS(good=3,medium=2,poor=1 poor 14 11.7%
) medium 61 50.8%
good 45 37.5%
season of automn 54 45.0%
summer 44 36.7%
winter 9 7.5%
spring 13 10.8%
Valid 120 100.0%
Missing 0
Total 120
Subpopulation 23a
a. The dependent variable has only one value observed in 6 (26.1%) subpopulations.



Model Fitting Information
Model Fitting
Criteria Likelihood Ratio Tests
Model -2 Log Likelihood Chi-Square df Sig.
Intercept Only 194.002
Final 152.995 41.007 35 .224



Goodness-of-Fit
Chi-Square df Sig.
Pearson 53.028 75 .974
Deviance 64.501 75 .801



Pseudo R-Square
Cox and Snell .289
Nagelkerke .301
McFadden .104



Likelihood Ratio Tests
Effect Model Fitting Likelihood Ratio Tests
Criteria
 -2 Log Likelihood
of Reduced Model Chi-Square df Sig.
a
Intercept 152.995 .000 0 .
sample 159.209 6.214 10 .797
code(Haramaya=1,awaday=2,a
dele=3)
BCS(good=3,medium=2,poor=1 168.511 15.517 10 .114
)
season of 172.611 19.617 15 .187
birth(automn=1,summer=2,wint
er=3,spring=4)
The chi-square statistic is the difference in -2 log-likelihoods between the final model and a
reduced model. The reduced model is formed by omitting an effect from the final model.
The null hypothesis is that all parameters of that effect are 0.
a. This reduced model is equivalent to the final model because omitting the effect does not
increase the degrees of freedom.



Parameter Estimates
number of parity(first 95% Confidence Interval for
parity=1,second Exp(B)
parity=2,third parity=3 fourth
parity=4
fifthyparity=5,above6=6)a B Std. Error Wald df Sig. Exp(B) Lower Bound Upper Bound
first parity Intercept 12.354 1085.062 .000 1 .991
[sample 13.511 576.339 .001 1 .981 737785.76 .000 .b
code(Hara 0
maya=1,aw
aday=2,ade
le=3)=1]
[sample 34.339 1349.990 .001 1 .980 81892711 .000 .b
code(Hara 5117670.0
maya=1,aw 00
aday=2,ade
le=3)=2]
[sample 0c . . 0 . . . .
code(Hara
maya=1,aw
aday=2,ade
le=3)=3]
[BCS(good -11.274 .000 . 1 . 1.270E-5 1.270E-5 1.270E-5
=3,medium
=2,poor=1)
=1]
[BCS(good 33.510 1099.789 .001 1 .976 35743102 .000 .b
=3,medium 5047285.9
=2,poor=1) 40
=2]
[BCS(good 0c . . 0 . . . .
=3,medium
=2,poor=1)
=3]
[season of -12.447 1085.063 .000 1 .991 3.929E-6 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=1]
[season of -.102 1480.299 .000 1 1.000 .903 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=2]
[season of -33.129 1340.620 .001 1 .980 4.096E-15 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=3]
 [season of 0c . . 0 . . . .
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=4]
second parity Intercept 12.857 1085.062 .000 1 .991
[sample 12.416 576.339 .000 1 .983 246833.98 .000 .b
code(Hara 1
maya=1,aw
aday=2,ade
le=3)=1]
[sample 34.100 1349.990 .001 1 .980 64508809 .000 .b
code(Hara 4564532.8
maya=1,aw 00
aday=2,ade
le=3)=2]
[sample 0c . . 0 . . . .
code(Hara
maya=1,aw
aday=2,ade
le=3)=3]
[BCS(good 11.117 536.419 .000 1 .983 67314.698 .000 .b
=3,medium
=2,poor=1)
=1]
[BCS(good 34.448 1099.789 .001 1 .975 91362962 .000 .b
=3,medium 4783102.5
=2,poor=1) 00
=2]
[BCS(good 0c . . 0 . . . .
=3,medium
=2,poor=1)
=3]
 [season of -12.752 1085.062 .000 1 .991 2.896E-6 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=1]
[season of .286 1480.298 .000 1 1.000 1.331 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=2]
[season of -36.546 1340.619 .001 1 .978 1.344E-16 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=3]
[season of 0c . . 0 . . . .
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=4]
third parity Intercept 12.652 1085.062 .000 1 .991
[sample 12.682 576.339 .000 1 .982 321778.85 .000 .b
code(Hara 1
maya=1,aw
aday=2,ade
le=3)=1]
[sample 33.931 1349.990 .001 1 .980 54452956 .000 .b
code(Hara 0911663.7
maya=1,aw 50
aday=2,ade
le=3)=2]
[sample 0c . . 0 . . . .
code(Hara
maya=1,aw
aday=2,ade
le=3)=3]
[BCS(good 9.066 536.420 .000 1 .987 8651.838 .000 .b
=3,medium
=2,poor=1)
=1]
[BCS(good 33.599 1099.789 .001 1 .976 39074081 .000 .b
=3,medium 4511782.6
=2,poor=1) 00
=2]
[BCS(good 0c . . 0 . . . .
=3,medium
=2,poor=1)
=3]
[season of -11.974 1085.063 .000 1 .991 6.306E-6 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=1]
[season of 1.310 1480.298 .000 1 .999 3.705 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=2]
[season of -34.832 1340.620 .001 1 .979 7.460E-16 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=3]
 [season of 0c . . 0 . . . .
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=4]
fourth parity Intercept 12.766 1085.062 .000 1 .991
[sample 12.491 576.339 .000 1 .983 265954.77 .000 .b
code(Hara 4
maya=1,aw
aday=2,ade
le=3)=1]
[sample 33.762 1349.990 .001 1 .980 46000636 .000 .b
code(Hara 0265681.3
maya=1,aw 00
aday=2,ade
le=3)=2]
[sample 0c . . 0 . . . .
code(Hara
maya=1,aw
aday=2,ade
le=3)=3]
[BCS(good 10.832 536.418 .000 1 .984 50625.222 .000 .b
=3,medium
=2,poor=1)
=1]
[BCS(good 33.779 1099.789 .001 1 .975 46774796 .000 .b
=3,medium 9167927.1
=2,poor=1) 00
=2]
[BCS(good 0c . . 0 . . . .
=3,medium
=2,poor=1)
=3]
 [season of -11.713 1085.062 .000 1 .991 8.185E-6 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=1]
[season of 1.027 1480.298 .000 1 .999 2.793 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=2]
[season of -36.199 1340.619 .001 1 .978 1.900E-16 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=3]
[season of 0c . . 0 . . . .
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=4]
fifth parity Intercept 13.889 1085.062 .000 1 .990
[sample 12.527 576.339 .000 1 .983 275815.70 .000 .b
code(Hara 1
maya=1,aw
aday=2,ade
le=3)=1]
[sample 34.655 1349.990 .001 1 .980 11230465 .000 .b
code(Hara 91290448.
maya=1,aw 800
aday=2,ade
le=3)=2]
[sample 0c . . 0 . . . .
code(Hara
maya=1,aw
aday=2,ade
le=3)=3]
[BCS(good 10.718 536.419 .000 1 .984 45182.253 .000 .b
=3,medium
=2,poor=1)
=1]
[BCS(good 33.592 1099.789 .001 1 .976 38813952 .000 .b
=3,medium 6843194.3
=2,poor=1) 00
=2]
[BCS(good 0c . . 0 . . . .
=3,medium
=2,poor=1)
=3]
[season of -13.757 1085.062 .000 1 .990 1.060E-6 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=1]
[season of -1.755 1480.298 .000 1 .999 .173 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=2]
[season of -36.180 1340.619 .001 1 .978 1.937E-16 .000 .b
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=3]
 [season of 0c . . 0 . . . .
birth(autom
n=1,summe
r=2,winter=
3,spring=4)
=4]
a. The reference category is: above 6.
b. Floating point overflow occurred while computing this statistic. Its value is therefore set to system missing.
c. This parameter is set to zero because it is redundant.

Do the appropriate regression analysis for the above data, develop model equation by appropriate
variable screening methods and write a full document by incorporating all components of
research report writing protocols including (introduction, statement of the problem, rational of
the study, objectives, literature review, conceptual framework, materials and methods(study area,
study design, study period, source and study population, sample size determination, sampling
method and techniques, operational definitions, dependent and independent variables, data
quality assurance and a data management and analysis, participant information sheet and consent
forms for animal owners), result, discussion, conclusion, recommendations based on your
findings and references) with appropriate interpretation for all the above statistical graphs and
tests result.

Note: There will be individual presentation, so you will be ready for that accordingly
(presentation accounts 15% and document 15% loads).

II. Part two

Non parametric statistical test
 Single-sample sign test
 Paired-sample sign test
 Wilcoxon rank sum test
 Wilcoxon Sign rank test
 Kruskal Wallis test
 Spearman rank correlation coefficient
Describe briefly about the above listed tests concerning on the formula, its counterpart
parametric tests, conditions to use those test, strength and weakness of those tests and their
practical application on different statistical data and practices in different available software's.

Part three:

1. Plasma urea and creatinine are routinely measured to evaluate the renal function. In
healthy cats the mean urea value in a given pathology laboratory is 7.5 mmol/l. Plasma
urea value in a random sample of 60 healthy cats in January were measured to verify the
assay. The data were approximately normally distributed with a mean urea content of
9.7mmol/l and an estimated standard error of 0.22 mmol/l. Is there any evidence at 99%
confidence level to indicate that the assay performance in this laboratory changed in
January?
2. In a given district animals were reared in a grazing area where by phytotoxic plants dominates.
Animals that graze on such grazing land and animals that feed in stall barn were compared for the
occurrence of photosensitization. In a ten days collected data, out of a total 400 cattle reared in
grazed land 60 of them had developed photosensitization. However, only 5 animals on the stall
barn were developed the condition from 100 cattle feed on a stall barn.
A. Draw 2x2 contingency table?
B. Is there a difference in the occurrence of photosensitization in the two animal groups?
C. How did you declare the association between grazing on abundant phytotoxic plant and
occurrence of photosensitization?
D. State your hypothesis and test the association at 90% confidence level?
E. Interpret the result?
3. A veterinary laboratory technologist wants to predict the interval estimate for the average age of
cattle visiting a certain veterinary clinic. He assumes that the mean age of cattle are
approximately normally distributed with population standard deviation of 3 years. A sample of 20
cattle that participate in the study yielded a mean age of 4 years. Estimate and interpret the 90%,
95% and 99% CI for the population mean?
4. A veterinary research team wants to estimate the prevalence of Brucellosis at Harar intensive
dairy farms. A random sample of 400 cattle was obtained and it was found that 5% of them were
positive for Brucellosis upon CFT test. Compute and interpret the 90%, 95% and 99% CI of
Brucellosis in Harar intensive dairy farms.
5. A laboratory technologist at Haramaya veterinary clinic was interested to compare the mean PCV
value of sheep and goat of the same age he enrolled. Ten animals from each species and their
PCV measured values was shown from the table below. Assume that the two population are
normally distributed with equal variance for the population. Does this data provide sufficient
evidence between the mean PCV levels among the species at 90%, 95% and 99% Confidence
level? Interpret the result?

ID Sheep PCV Goat PCV

1 29 30
2 30 32
3 40 43
4 21 45
5 22 47
6 30 32
7 20 34
8 19 29
9 18 30
10 20 40

6. A study was conducted to evaluate the efficacy of Trypanocidal drugs. The efficacy was assessed
using a change in PCV value. Three types of tripanocidal drugs were used and 24 trypanosome
infected animals were randomly assigned to one of the treatment groups. The PCV value of
animals after the experiment was measured and presented below in the table.

Treatment I PCV I Treatment II PCV II Treatment III PCV III Treatment IV PCV IV
Control 24 DA 35 Trypamidium 31 Homidium.B 24
Control 25 DA 40 Trypamidium 21 Homidium.B 25
Control 24 DA 23 Trypamidium 27 Homidium.B 25
Control 23 DA 30 Trypamidium 30 Homidium.B 24
Control 18 DA 25 Trypamidium 27 Homidium.B 23
Control 20 DA 40 Trypamidium 42 Homidium.B 22

A. Which test statistics is appropriate and why?

B. Test the hypothesis?
C. Did the use of trypanocidal drug improve the PCV value?
D. Present the result by ANOVA table and interpret it?
7. A researcher wants to determine the effect of artificial insemination (AI) on the incidence of
dystocia. He randomly assigns 100 cows to treatment (AI) group and 100 cows to control (natural
mating) group. During follow-up, he got dystocia in 20 cows who took AI and 10 cows who took
natural mating. By this scenario, answer the following questions?
 A. What type of study design and control group was the researcher applied?
B. Create two-by-two table?
C. Calculate the appropriate measure of association and interpret the result?
D. Test the hypothesis and declare the effect of AI on the incidence of dystocia at α = 0.05?
8. Assume that the data given below in the table will be the results of ten VLT students at
HU with their respective duration of study in hour at daily basis for the course
introduction to statistics and research methodology. Based on the data:
A. Calculate and interpret the Karl Pearson’s correlation coefficient for the study hour of
students in years and exam score.
B. Develop a regression equation for student’s exam score by study hours and interpret
it?
C. Estimate the exam score of students when study hour is 55 years.

Study hour Exam score

3 80
5 90
2 75
6 80
7 90
1 50
2 65
7 85
1 40
5 90

9. A physician claims that joggers maximal volume oxygen uptake is greater than the
average of all adults. A sample of 15 joggers has a mean of 40.6 milliliters per kilogram
(ml/kg) and a standard deviation of 6 ml/kg. If the average of all adults is 36.7 ml/kg, is
there enough evidence to support the physician’s claim at α=0.05?
10. Compare and interpret the mean difference between aflatoxin level of feed stuffs on a
sample of 25 groundnut and 20 soybean feed samples at α=0.05 and the data is given here
under in the table.

Feed stuffs Number of SD Mean of aflatoxin level

samples
Groundnut 25 10 30
Soya been 20 20 25
 11. A study was made to compare the plasma lactate concentration (mmol/l) in Dutch warm
blooded horses cantering (galloping) at constant speed either on a track or an inclined
treadmill. The speed was chosen as a horses own comfortable speed on the track.
Samples were taken after 5 minutes cantering on the track and treadmill, the order of
which was randomized for ten horses and we show the plasma lactate concentrations as
follows: Test whether exercise exerted by the horse can be considered to be of similar
metabolic demand in both situations at 95% confidence level?

Plasma lactate concentration (mmol/l)

Horse 1 2 3 4 5 6 7 8 9 10
Track 2.0 7.7 4.7 4.7 2.9 2.5 5.3 4.8 3.1 3.9
Treadmil 3.5 7.2 4.6 5.7 5.5 4.4 5.6 4.6 3.5 4.9
l

Submission and presentation date must be before final exam.

You can submit your final document via this email: getahun2470@gmail.com