1.

Residents of three villages with three different types of water supply were asked to participate
in a survey to identify cholera careers and the proportion of carriers in each village was
compared. The type of study design used is:
Answer: Cross-sectional study
since in cross-sectional study data on both disease and determinants are collected for a specified
population at the same point in time.

2. Give true or false about observational studies

a. Events are observed as they occur, with no active intervention by the researcher
b. Comparison groups may differ with respect to factors related to the outcome
c. Subjects may be followed forward from exposure to outcome or backward from outcome
to exposure
Answers:
a. True: because in observational studies the researcher observes the object of research but does
not intervene.
b. True:
c. True: since under observational studies there are basically two types of cohort studies;
prospective and retrospective (historical). The difference between the two lies in where the
starting point of the study is deemed to begin. In a prospective cohort study the starting point of
observation (time zero) is ‘now’, and the population is followed into the future. The exposure of
interest may or may not have occurred when the study was initiated, but the outcome has not.
In a retrospective the ‘starting point’, that is the point of initial exposure occurred sometime in
the past and the experience of the population is followed up to the present time. At the time the
study is initiated, both the exposure and outcome have occurred.

3. Give true or false about retrospective cohort study

a. Allows to investigate many outcomes
b. Incidence rates can be calculated
c. Selection bias is eliminated
d. Recall bias is minimum compared to case-control study
e. Usually requires less time than prospective cohort
Answers:
a. True: a retrospective cohort study is an observational research design which begins when a
group of people (a cohort) initially free of disease (outcome of interest), are classified according
to a given exposure, and then followed up over time. The researcher compares whether the
subsequent development of any new cases of a disease (or other outcome of interest) differs
between the exposed and non-exposed groups.

b. True: measure of disease frequency in cohort study is incidence.

c. True: in cohort study no chance of bias being introduced due to awareness of being sick as in
encountered in case-control studies.

1
d. True: in cohort study less chance for the problem of selective survival or selective recall.

e. True: Historically constructed cohorts share several advantages of the prospective cohort. If
all requirements are satisfied, a historical cohort may suffer less from the disadvantages of time
and expense.

4. Identify the type of study designs for each of the following

a. Study of past mortality trends to predict the future mortality
b. Study of the incidence of cancer in men who have quit smoking
c. Study of people who are not treated for a certain disease
d. Study of people who have a homogeneous characteristics for a certain disease
Answers:

a. Retrospective cohort study

A historical cohort study depends upon the availability of data or records that allow
reconstruction of the exposure of cohorts to a suspected risk factor and follow-up of their
mortality or morbidity over time. In other words, although the investigator was not present when
the exposure was first identified, he reconstructs exposed and unexposed populations from
records, and then proceeds as though he had been present throughout the study.

b. Prospective cohort study

Incidence can be estimated in prospective cohort study.

c. Cross-sectional study
It provides a prevalence rate at a particular point in time (point prevalence) or over a period of
time (period prevalence). The study population at risk is the denominator for these prevalence
rates. Included in this type of descriptive study are surveys in which the distribution of a disease
is assessed.

d. Cohort study

5., You are evaluating the impact of a certain drug on patients aged 75 and above admitted to
Tikur Anbessa Hospital.
a. What study design is most appropriate? Why?
b. What is the target population?
c. What is the possible source of population?
d. What possible sampling frame do you use?
e. How do you select your sample? Why?
f. If no previous study is available, what sample size is needed to estimate the impact in the
target population at a 95% CI and marginal of error

2
Answers:
a. Intervention study
In an intervention study, we are interested in the consequences of some treatment on some
outcome.

b. Target population: All patients aged 75 and above in Tikur Anbessa Hospital
Target population: A collection of items that have something in common for which we wish to
draw conclusions at a particular time.

c. Source population: All patients aged 75 and above admitted to Tikur Anbessa Hospital

d. Sampling frame: Lists of all patients aged 75 and above admitted to Tikur Anbessa
Hospital

e. Simple random sampling

This is the most common and the simplest of the sampling methods. In this method, the subjects
are chosen from the population with equal probability of selection. One may use a random
number table or use techniques such as putting the names of people into a hat and selecting the
appropriate number of names blindly.
The simple random sample has the advantages that it is easy to administer, is representative of
the population in the long run, and the analysis of data using such a sampling scheme is
straightforward.

f. Sample size:
2
Z P (1−P)
n=
d2
Where, n= sample size

P=0.5 and 1-p=0.5

d= margin of sample error=0.05

Z= confidence interval (95%) = 1.96

n= (1.96)2(0.5)(0.5)= 384.16= 384

(0.05)2

6. A study in a hospital showed that 30% of those who came for emergency reasons were aged
15 years or less.
a. What is the variable of interest of the study?
b. What is the type of the variable?
c. What is the scale of measurement for the variable?
d. Which descriptive measure is most appropriate for the data?

Answers:

3
 a. Age
b. Qualitative/ categorical
c. Ordinal: classifies age based on a qualitative assessment.
Numeric age but there is no implication of more or less.
d. proportion/ percentile

7. The table below shows results (out of 50) of pre-test and post-test scores for students attending
biostatistics course using mean , median, standard deviation and inter-quartile range.
Exam type Score
Mean Median SD IQR
Pre-test 10 22 12 35
Post-test 30 35 10 20
a. What type of variable is pre-test?
b. What is the scale of measurement for the variable?
c. The mean of the difference between pre and post tests is 20 (True, False)
d. The median of the difference between pre and post tests is 13 (True, False)
e. The SD of the difference between pre and post tests is -2 (True, False)
f. The IQR of the difference between pre and post is 15 (True, False)
g. Which graphical display is most appropriate to compare pre and post tests
Answers:

a. Quantitative/ continuous

b. Ratio

c. True

d. True

e. False

f. True

g. Frequency polygon/ Box plot: It is the most appropriate for comparing two or
more sets of data.

8. What is the most appropriate graphical method to display for the following data?
a. The distribution of diarrhea in an outbreak investigation
b. The weight of newborns in a health center
c. The marital status of pregnant women attending ANC
d. Treatment failure among TB patients
Answers:

a. Frequency polygon/ Line graph

4
 b. Histogram

c. Bar chart/ Pie chart

d. Bar chart/ Pie chart

9. The following table shows frequency of diastolic blood pressure (DBP) of men aged 30-69
with mean DBP of 84mmHg

DBP in mmHg Freq. Relative freq. Cumulative

freq
Below 65 60
65-74 270
75-84 540
85-94 420
95-104 150
105-115 45
Above 115 15
a. Fill the relative and cumulative relative frequency columns in the table
b. People with 95mmHg and above are considered hypertensive. What is the percentile
hypertensive?
c. The frequency for 800 women is almost similar with mean 79mmHG and same SD as
that for men. Say True or False for the following and justify your answers
(i) The median DBB will be the same for both sexes
(ii) The proportion of hypertensive is the same for both sexes
(iii) The variability of DBP is lower for women
Answers:

a.

DBP in mmHg Freq. Relative freq. Cumulative

rel.freq
Below 65 60 4 4
65-74 270 18 22
75-84 540 36 58
85-94 420 28 86
95-104 150 10 96
105-115 45 3 99
Above 115 15 1 100

b. 86% and above

c. (i) False
Although they have the same SD, the median not be the same for both
sexes.
 since median is insensitive to extreme values (outliers).

5
  but SD is affected by extreme values (outliers).

(ii) False
♦ Men proportion of hypertensive: 210/1010 = 0.21
♦ Women proportion of hypertensive: 800/1010 = 0.79

(iii) False

S
CV = x 100 %
x̄

The standard formulation of the coefficient of variation (CV) is the ratio of the standard
deviation to the mean of a give data.

Since men and women has same SD. And women mean is 79 mmHG which is lower than
men 84 mmHG. Therefore women CV is higher than men.

10. The following table shows crude death rates (CDR) in four different villages

Village No of CDR
population
A 50,000 10
B 100,000 8
C 250,000 8
D 250,000 4
a. Which of the villages has the largest number of deaths?
b. Can we say village D has a better survival compared to other villages? Justify your
answer.

Answers:
a. Village C

Total number of death per year

CDR = x 1000
Mid year population

For village A: Total number of death per year = CDR*No. of population

1000
= 10*50,000= 500
1000
For village B: Total number of death per year = CDR*No. of population
1000
= 8*100,000= 800
6
 1000
For village C: Total number of death per year = CDR*No. of population
1000
= 8*250,000= 2000
1000
For village D: Total number of death per year = CDR*No. of population
1000
= 4*250,000= 1000
1000

b. False
village D has the largest number of death than villages A and B. Therefore it has n a better
survival compared to other villages

11. An instructor gave a quiz with 3 questions each worth one point. 30% of the class scored 3
points, 50% scored 2 points, 10% scored one point and the rest scored zero.
a. Calculate the mean, median and mode of the scores if there were 30 students in the
class
b. Is it possible to find out the three measures without being told the number of
students? Justify your answer

Answers:
a. 30% of 30 = 30*30/100 = 9 students score 3 points
50% 0f 30 = 50*30/100 = 15 students score 2 points
10% of 30 = 10*30/100 = 3 students score 1 point
10% of 30 = 10*30/100 = 3 students score 0 point

 order the observations in ascending order

Observation Score Observation Observation

1 0 16 2
2 0 17 2
3 0 18 2
4 1 19 2
5 1 20 2
6 1 21 2
7 2 22 3
8 2 23 3
9 2 24 3
10 2 25 3
11 2 26 3
12 2 27 3
13 2 28 3
14 2 29 3
15 2 30 3

7
Mean =

1+1+1+2+2+2+2+2+2+2+2+2+2+2+2+2+2+2+3+3+3+3+3+3+3+3+3 = 60/30
30
Mean = 2

= 30+1/2 =15.5
15th+16th/2 = 2+2/2 = 2
Median = 2
Mode = 2 since has frequency 15

b. Yes! by taking any number of students we can get the three measures
median, mean and mode.

12. The following are life expectancies of males in six European countries: 74, 77, 73, 75, 77 and
78. If the observed value of 74 is mistakenly recorded 740. What is the effect on each of the
following? . Mean, Median, SD, IQR variance

Answers:
Before mistakenly recorded:
 order the given observations in ascending order

Observation Age
1 73
2 74
3 75
4 77
5 77
6 78

Mean =

73+74+75+77+77+78 = 75.6
5
Mean = 75.6

8
Median=76

variance = (73-75.4)2 + (74-75.4)2 +(75-75.4)2 +(77-75.4)2 + (77-75.4)+(78-75.4)2

=19.76/5=3.9
Standard deviations of sample:- is the positive square root of the variance.
S (SD) = √S2 = √ 4.3 = 1.9
IQR
(Q 1) ={0.25 (n+1)}th observation = {0.25 (6+1)}th observation= 1.75th observation which lies
b/n the 1st and 2nd observation
Q 1 = 1st + 0.75 (2nd -1st) = 73 + 0.75 (74-73) =73.75
About ¼ of the data falls on or below 73.75

(Q 3) = ={0.75 (n+1)}th observation = {0.75 (6+1)th observation = 5.25th observation which

lies b/n the 5th and 6th observation
= 4th + 0.25 (6th - 5th) = 77 + 0.25 (78-77) =77.25
About ¾ of the data falls on or below 77.25
IQR= Q3-Q1= 77.25-73.75= 3.5

After mistakenly recorded:

 order the given observations in ascending order

Observation Age
1 73
2 75
3 77
4 77
5 78
6 740

Mean = 73+75+77+77+78+740 = 208.6

6
Mean =186.7

9
Median=77

variance = (73-186.7)2 + (75-186.7)2 +(77-186.7)2 +(77-186.7)2 (78-186.7)2+ (740-186.7)2

=367,429.3/5=73,485.86
S (SD) = 271.1
IQR = 169
 Effect: Only median is insensitive to extreme values (outliers).
 Others Mean, SD, IQR and Variance are affected by extreme values (outliers).

13. A physical examination and audiometric tests were given to 500 individuals with suspected
ear problem. The results are shown in the following tables.

Hearing Physical Examination Hearing Audiometric test

problem Present Absent problem Present Absent
Yes 240 40 Yes 270 60
No 60 160 No 30 140

Compared to physical examination calculate the sensitivity and specificity of audiometric test

Answers:
Sensitivity and Specificity:

Sensitivity: is the proportion of diseased people who would be correctly classified

Specificity: is the proportion of non diseased people who would be correctly classified

Physical examination sensitivity Audiometric test sensitivity

sens = 240 = 80% sens = 270 = 90%
240+60 270=30

Physical examination Specificity Audiometric test Specificity

spec = 160 = 80% spec = 140 = 70%
160+40 140+60

Comparison:
 Audiometric test is more sensitive than physical examination.
 Physical examination is more specific than audiometric test.

10