INTRODUCTION TO

BIOSTATISTICS.
COURSE CODE; ML 0112

ELISHA E. MGOJI , BMLS

COLLEGE OF HEALTH SCIENCE- UDOM
 What is statistics?
• Statistics as a ‘Despline’ is a field of study concerned in
broad terms with:
(i) Collecting, organizing and summarizing data is
systematic way.
(ii)Drawing inferences about a population on the basis of
only a part of the population targeted.
• The first part of the subject is usually referred as
DESCRIPTIVE STATISTICS, while the second part is called
INFERRENCIAL STATISTICS.
• In this course, we are mainly dealing with the second part
of statistics as a despline. This kind of Biostatistics will be
that of MEDICAL BIOSTATISTICS.
• i) Descriptive Statistics – Deals with Collection,
organization, summarization, and presentation of data.
• ii) Inferential Statistics-Deals with generalizing from
samples to populations using probabilities. Performing
hypothesis testing, determining relationships between
variables, and making predictions.
• Therefore, Biostatistics can be defined as the application
of statistics to biological problems.
• Note: In Agriculture and veterinary medicine, they prefer
the term BIOMETRY rather than Biostatistics meaning,
the measurement of life. ie. Application of statistics
methods to analyze biological data.
 WHY STUDY BIOSTATISTICS IN MEDICAL
SCHOOL?

• -Because of some element of variability in life, hence we

need standardized technique to couple with inevitable
biological variability. Eg Blood pressure, Hemoglobin in
blood etc. -But also in decision making.
WHERE DO WE APPLY BIOSTATISTICS METHODS?

Biostatistics methods have a role to play in

• Office health statistics( statements)
• Epidemiology e,g. Association of disease with some etiological
agents
• Clinical studies e,g. comparison of treatments in clinical trials
• Laboratory studies( Dose-response studies)
• Health service Administration e.g to priotize areas for intervention
• Human Biology e.g .Growth patterns.
 DESCRIPTIVE STATISTICS

Covers
Types of variables
Descriptive methods for qualitative data
Descriptive methods for quantitative data
 VARIABLES

• What is a variable?
• An observation or characteristic that can take different
values for different objects.
 Examples of Variables
• Height cm • 158,169,170,200
• Weight kg • 10.2.50.69.34
• Outcome of disease • Recovery, chronic illness,
• Marital status death
• Single, married, widowed,
separated
 Types of variables
• Qualitative variables: do not take numerical values.
• e.g. sex, type of health facility, marital status, etc.
• Quantitative variables: take numerical values
• e.g. age, no. of sexual partners, hb, parity, etc.
• Quantitative variables
• Discrete; Take only fixed values e.g age, counts
• Continuous; take any values within meaningful extremies eg
heigh, weight
Relationships between variables
• Two variables that show some connection with one
another are called associated (dependent)
• Association can be further described as positive or
negative
• If two variables are not associated, they are said to be
independent
 Descriptive methods for
qualitative data
• Tables
– frequency tables (one-way table)
– cross-tabulations (two-way table)
• Diagrams
– pie chart
– bar chart
– map
 TABLES
• Set of data arranged in rows and columns.
• Should be as simple as possible
• Should be self explanatory (stand alone)
 One-way table
• Distribution of types of latrines
 TWO-WAY TABLE
• Two variables presented simultaneously

• Both variables are categorical

• Two way tables allow comparisons

• Present the number and percentage in each category

 EXRICISE
• Suppose that in a study of the factors affecting the
utilization of antenatal clinics we find that 51 out of the 80
women who lived within 10 km of the clinic came for
antenatal care, while 35 out of 75 who lived more than 10
km away went for ANC care.

• Present the information in a two way table to show how the

proportion of women utilizing ANC varies with distance
UTILIZATION OF ANTENATAL CLINICS
BY DISTANCE FROM A CLINIC
UTILIZATION OF ANTENATAL CLINICS
BY DISTANCE FROM A CLINIC
UTILIZATION OF ANTENATAL CLINICS
BY DISTANCE FROM A CLINIC
• The percentages may be row or column percentages
depending on the purpose of the table.
• e.g In the above example the rationale is to investigate
whether distance from ANC has an influence on utilization
of ANC.
• aim is to see if utilization of ANC is higher in those who
live close to ANC than in those who live far from the clinic
• row percentages are more appropriate in this case.
 QUIZ
• The table below shows the results of a study investigating
whether bicycle safety helmets are effective in preventing
head injury. Calculate the appropriate percentages. And
make conclusion based on that findings
 DIAGRAM
• Presentation of information/data (e.g. report) for ease of
appreciation
• To serve as an aid for further statistical analysis
 Types of diagrams
• Pie charts
• Bar charts
• Maps
 PIE CHARTS
• Express the distribution of individual observations into
different categories
• Percent in categories represented by size of angle and
area of sector
• Sometimes difficult to show small categories
 Distribution of types of latrines
• 20% of
None/bush households had
20 %
no latrines
• 63% of
Flush latrine
17 %
Pit latrine households had
63 %
pit latrines
• 17% of
households had
flush lattrines
 Bar chart
• Can be used for single variable or several
• various categories of a variable are represented on one
axis
• frequency or relative frequency (percent) in each category
represented by length of bar
Current Use of Family Planning among Currently
Married Women

Any method 26

Any traditional method 6

Any modern method 20

Pill 6

Injectables 8

Male condom 2
 MAPS
• used to show the location of events or attributes
• useful for showing the geographic distribution of an event
THANK
YOU
 Descriptive methods for
quantitative data
♦ Tables
– Frequency distribution
♦ Diagrams
– Histograms
– Scatter diagrams
– Line diagrams
– Frequency polygon
– Cumulative frequency curve
Frequency Distribution
Cumulative
No of partners Frequency Frequency
1 2 2
2 3 5
3 5 10
4 9 19
5 1 20
2+3+5
Total 20 =10
 Frequency Distribution.
• Example: on mobile phone cost
• Sample of mobile Cost for 50 UDOM residents
91 78 93 57 75 52 99 80 97 62
71 69 72 89 66 75 79 75 72 76
104 74 62 68 97 105 77 65 80 109
85 97 88 68 83 68 71 69 67 74
62 82 98 101 79 105 79 69 62 73
Frequency Distributions
 SUMMARIZING DATA
Central Tendency (measures of center)
• Mean
• Median
• Mode
Variation (measures of spread)
• Range
• Interquartile Range
• Variance
• Standard Deviation
 Introduction: distributional
shapes
• Recall: the mid-points of the bars in a histogram
for a continuous variables can be joined smoothly
to produce a curve.
• In the process, a distributional shape of such a
variable can be visualized.
• In general the distribution can be UNIMODAL
(i.e. exhibiting one peak) or, less commonly,
BIMODAL
• In UNIMODAL distributions, the curves can be symmetric
(“normal”) or asymmetric
• Asymmetric distributions can be skewed positively
(skewed to the right: with a long tail on the right) or
negatively skewed (skewed to the left: with a long tail on
the left)
• In UNIMODAL distributions, the curves can be symmetric
(“normal”) or asymmetric
• Asymmetric distributions can be skewed positively
(skewed to the right: with a long tail on the right) or
negatively skewed (skewed to the left: with a long tail on
the left)
Symmetric (normal) distribution
Positively skewed distribution
Positively skewed distribution
Negatively skewed distribution
Negatively skewed distribution
Bimodal distribution
 Measures of central tendency
♦ Mean =  xi n
♦ Example. Consider the following data:
2, 3, 9, 5, 4, 0, 6, 3, 4
xi = 36; n = 9
å xi n = 36/9 = 4.0
♦ It is useful for symmetrical
♦ For skewed distributions, it can mislead
♦ For example, mean of 5, 5, 5, 7, 10, 20, 102 is 22
 Measures of central tendency
♦ Median is the central value
– e.g. in example1 data, rearranging the values, we obtain: 0, 2, 3, 3, 4,
4, 5, 6, 9
– middle value = (n+1)2th value
= 5th value = 4
♦ Consider following: 0, 2, 3, 3, 4, 5, 5, 6, 6, 9
Median = (10+1)th value
2
Median= 5.5th value
= (4+5)/2= 4.5
 Measures of central tendency
♦ Mode is the most commonly occurring value in a
set of values
♦ That is, mode is value with highest frequency
 What measure to use?

♦ For symmetric distributions, mean is preferred

♦ Why? - utilises all values

- characterised by mathematical niceties

♦ For asymmetric distributions, median preferred
♦ Why? – not affected by extreme values
 Measures of spread/variability
• Range
• This is the difference between the lowest and the highest
observation
• The wider the range, the more variation
• Affected by extreme values
• Exclude 25% bottom and top cases, becomes
INTERQUARTILE range
 Variance

♦ The average squared deviation from the mean

♦ Variance = s2 = (xi - x)2 (n-1)

♦ Equivalently, s2 =  xi2  (xi )2 n (n-1)

 Variance: example

♦ The average squared deviation from the mean

♦ n=9; xi=36; =4; xi2=196;  (xi - x)2=52

♦ Variance =  (xi - x)2 (n-1)

= 52/(9-1) = 6.50
OR Variance =  xi2  (xi )2 n (n-1)
= [196-144]/(9-1)
= 52/8=6.50
 The standard deviation (SD)

♦ The root mean square of deviations from average

♦ hence, we take the square root of the variance:
this is the standard deviation
– (i.e. standard deviation = {variance})
♦ It measures roughly how far o the entries are from their
average
 Relationship between mean, SD
and the frequency distribution
• For symmetric distributions, approximately, 95% of all
observations lie in the interval  2s
• The limits: - 2s and + 2s are referred to as the 95%
tolerance or spread or reference limits, and the values
contained in the interval are commonly termed as “THE
NORMAL VALUES”