BIOSTATISTICS NOTES

HISTORY

The history of statistics dates back to the Biblical times. It was used by the ancient Romans, early
Egyptians and ancient China. The statistical record books of Charles the Great and William the conqueror
are examples of the Middle age use of statistics

MEANING

Statistics however, is from the Latin word ' status ' meaning ' state ' / ' condition '

Biostatistics is also called biomedical data or biometry.

DEFINITION

Biometry is defined as a scientific method for collation, collection, compilation, organisation,

presentation, analysis and interpretation of data with a view to offering inferences, and making
reasonable conclusions based on the analysis.

Parts of biometry

Biostatistics has two parts namely:

-Descriptive aspect

-Inferential aspect

1. Descriptive statistics deals with simple data presentation example in the form of graphs, tabular or
calculated basic statistical measures like mean, mode, %, ratios etc

2. Inferential statistics is concerned with the sampling of the population with a view to infer for the
entire population. Here, data are subjected to statistical hypothesis for accurate, appropriate inferences
to be made.

TERMINOLOGIES

1. Vital statistics - These are statistics or number of events in the life of a man or community in general.
It deals with events such as deaths, births, migration, marriages and deaths, divorces etc. These 3 vents
are registrable events with established data sources example Federal Office of Statistics, immigration
services, national Population commission,LGA etc
2. Hospital statistics -are statistics of events in the hospital setup like: personnel statistics, number of
beds, discharges, notification of selected diseases (morbidity statistics ) attendances, births and deaths
etc

3. Health statistics -deals with personnel data, attendances, number of health facilities, services delivery,
births n deaths, notification on selected diseases etc.

4. Epidemiological statistics -These helps to trace the sources of diseases as related to sex, age,
geographical location, cultural practices etc. It is defined as statistics or number of events 2hich are
collected as a result of occurrence of infectious or communicable disease such as TB, measles and
rabbies etc

5. Parameter - is a value calculated on the basis of a population study. It is represented by (mu)-

longtailed u. When a whole population is to be investigated or studied, while statistics gives value on the
basis of sampled calculated, parameters deals with the whole. So statistics is an estimate of the
parameter.

6 . Population - refers to aggregate of units from which observations are made. Population doe not
necessarily refer to living things but to non living as well example insects, houses. Population size is
usually represented by capital N (N).

7. Sample - is a part or subset of the population from which observations are made. Sample size is
usually represented by small n (n).

VARIABLES AND ATTRIBUTES

A variable is a quantity which changes in value from subject to subject or from time to time in the same
subject. A measurement variable therefore describes the characteristics of a numerical value. The value
varies from subject to subject or from time to time in the same subject example height (m), and weight
(kg )

A variable is divided into two parts

1. discrete variable -collectively called quantitative variables and a qualitative variables otherwise called
an attribute.

A continuous quantitative variable is one with potentially infinite number of possible values in any
interval. It can assume either an integral or fractional value and can be measured to different levels of
accuracy example heart beat, Blood pressure, weight, temperature measurements

A discrete quantitative variable on the other hand, is assigned finite number of values in any interval,
that is, it takes countable or whole numbers example parity, age last birth day in a family, WBC count etc

An attribute or qualitative variables is a type of variables which has no numerical value attached to it. It
is simply the characteristics to which a subject either belongs or doe not belong or a property or quality
that a subject either possesses or does not possess example type 9f blood, marital status, outcomes of
patient treatment etc.

COLLECTION AND ORGANISATION OF DATA AND SCALES OF MEASUREMENT

1. Regular system

2. Adhoc system of data collection

Regular or routine system of data collection refers to a registrable procedures for collecting data as they
become available example

-a regular system to collect births, deaths, marriages, divorces and migration (vital events).

-a reporting system for cancer cases

-a disease notification system to collect information on cholera, yellow fever, TB etc.

2. Adhoc collection of data - is usually in form of a survey to collect information that is not available on a
regular basis include special investigative studies and elaboration on aspects of data that have been
collected on a regular basis. Data collected through this system can be used for admission and research
purposes example :

*A survey to estimate the number of persons in the population with malnutrition

*A national health manpower survey

* survey to determine the number of persons in the community of Family planning meth9ds (procedure)

*An investigation of breastfeeding practice among women who registered a birth in the previous year

PROCEDURES FOR DATA COLLECTION

-A). Regular :

*Establish the rules and requirements that govern the system. These rules give legal backing to the
system especially those involving national activities. The rules and regulations must be enacted by the
relevant legislative or competent bodies.

*Establishment of offices, infrastructure facilities, personnel and denominators of appropriate

information (wide publicity ) on what you intend to do.

*Decisions on information to be collected

*Design of forms/registers for recording information

*Training of personnel on information recording

*Decisions on persons who supply information and how often information is to be supplied

*Designing the registration receipts for those who come to register

-B) PROCEDURE FOR DATA COLLECTION IN ADHOC SYSTEM

Simply implies a survey to investigate a group.

*Define or state the objective of the collection exercise indicating the type of information to be
collected and how it will be used

*Definition and identity of target population

*Decision on whether information is required from all the units of the target population or froma
selected group (sample size)n=1000?

*Decision on the sampling method to be used to collect the sample - simple random, systematic
sampling, stratified sampling, cluster sampling, multistage sampling - all collectively referred to as
probability sampling procedures

Non-probability sampling procedures - also called convenient sampling or purposive or judgemental

sampling procedures - usually not the best because of the elements of bias.

Sampling method determine the success of the overall research study and may make study irrelevant or
otherwise.

-Designing of instruments - Examples questionnaire forms or interview guides that is format

-Selection and training of personnel for recording information -very importantto eliminate interview
bias, ensures standardised record of data

-Real data collection by filling the forms by either face to face administration or otherwise

All the above are survey methods

INSTRUMENTS FOR DATA COLLECTION

There are three main instruments for measuring data namely :

1. Human

2. Apparatus

3. Combination of 1&2 above

-Human instruments -Here measurements are done by persons with little or no involvement of
apparatus Eg diagnosis, history taking etc

-Apparatus -Here measurement is done using mechanical devices example weighing scales,
thermometers, BP apparatus etc

-Combination of 1&2 example -Reading of X-ray films, blood film (interpretation )

RELIABILITY AND VALIDITY OF PROCEDURES

A reliable instrument is one which gives consistent results when applied more than once on the same
subject/unit under same conditions. The inherent performance of the instrument is assessed through its
reliability. However, the reliability of an instrument is usually affected by a number of
factors:fluctuations in the substance been measured; observer error; interobserver error. Consistent
result implies reliable instrument.

VALIDITY

An instrument has validity if it is indicative of the condition that it is supposed today measure exampl-
fever maybe a valid indicator of Malaria though insufficient in an malaria endemic areaarea

Components of Validity

-Sensitivity

-Specificity

The sensitivity of a test, a procedure or a measuring instrument is the quotient of the change in an
observed measure and the corresponding change in the value in the quantity or factor that is being
measured.

The larger the value of the quotient, the greater the sensitivity. In Epidemiology, sensitivity is defined as
the proportion of true positives correctly identified by a test which is expressed as follows:

Sen=a/a+c

Where a=number of true positives correctly identified by a test

c=number of false negatives given by the test

Expressed in % by multiplying by 100.

Specificity

Is defined as the extent today which a test or a procedure or a measuring instruments gives a response
today the presence of a given variable and is dormant (dead) Topic the presence of other variables. In
Epidemiology, specificity is defined as the proportion of true negatives correctly identified by a test and
is given by the relation

Spec=d/b+d

Where d=number of true negatives correctly identified

b=number of false negatives given by the test

DATA PRESENTATION

There are two main methods of presenting data:

1. Tabular method

2. Graphical method -Less ued

Tabular method involves the systematic arrangement of facts and figures in series of boxes made up of
rows and columns (stubs and cokumns). This makes a table easy to read and facilitate quick
comparisons.

There are three types of tables:

-Simple table

-Frequency distribution table

-Complex table

Parts of a table

a) The title

b) The caption or box head

c) The stub &

d) The source

The TITLE describes the subject matter of the table, usually placed above the table

The CAPTION also called the columnsubtitles the classification of items along the columns. Usually
placed vertically in arrangement

The STUB also called the row subtitles the classification of items along the ROWS
The SOURCE sites the place or explains the from where the original data or information contained in the
table were obtained from. Usually placed at the bottom or below the table. The source is usually stated
where the data is of secondary information. If it is a primary source of data, there may not be any need
for a source to be stated.

POINTS TO CONSIDER WHILE CONSTRUCTING STATISTICAL TABLES TO SERVE AS EFFECTIVE DEVICE FOR
DATA PRESENTATION

-The TITLE should be brief and self explanatory

-The number of columns should be kept as few as possible in order that the table could be taken in at a
glance

-The CAPTION and the STUB should be clear in meaning

-Figures to be compared should be placed as close as possible to each other

-Units of measurement used in the table must be clearly indicated preferably at the right hand top
corner of the table

-Percentages, averages and other computed ratios should be placed very close to the figures to which
they relate

-Source note must be shown at the bottom of the table. This makes it possibleto make references to the
original source of the data should the need arise.

Types of Tables

1. A simple table is one concerned with the presentation using a single variable. Consists merely of a list
of objects in one or two or more columns

Examples :Enrolment in Nigeria University in 1974/7 - university vs number of students enrolled.

Total =

Source :NUC

2. Frequency Distribution Table

Is another example of a simple table used importantly in statistical analysis

Must have a title, column stating the number of frequency. The arrangement maybe vertical or
horizontal. The source may be there depending on the type of data. A variable or variate refers to any
entity that varies. Measures like weight, height, income are variables because they vary from one
person to another
Yield is a variable associated with crops etc. A variable can assume different values. We use the term
variate - values to refer to the values taken by a variable. Tallies or strokes are used to indicate the
number of respondents to an enquiry in a population Example

Marital status :

Single #

Married #

Widowed#

Separated #

Divorced #

The total number of tallies or strokes must correspond with the total number of respondents.

CONSTRUCTION OF A FREQUENCY DISTRIBUTION TABLE FOR A QUANTITATIVE CONTINOUS

OBSERVATION

Example

Age (yrs) Fre (F)

0-4

5-9

10-14

15-19

20-24

25-29

29+

This an example of a grouped/classed frequency distribution because the cariate-variables are

presented in groups. It enables us to obtain by statistical computation various quantities such as the
mean, median, mode, variance, standard deviations etc. It has lower class limit LCL and Upper class limit
UCL. The class interval size I is the difference between the LCL and the UCL +1

I.e, I=UCL-LCL+1 Eg 9-5+1=5

Therefore, 5 is the class interval

Upper class boundary I=UCB-LCB I.e to get the class boundaries just minus 0.5 from the LCL and add 0.5
to the UCL. Then you subtract LCB from UCB Eg in above, 5-9, 9.5-4.5=5. Which means I is UCB-LCB+1

SAMPLING METHODS

There are two types of sampling methods

1. Probability - more useful sampling method

2. Non-probability sampling (Systematic sampling method ). Is a simple random sampling method