intoduction of biostatistics2006b

Introduction of Biostatistics

HERTANTO WAHYU SUBAGYO

Biostat-1 Hertanto

1

Statistics
• The science of :
- collecting
- summarizing
- presenting
- interpreting data,
g them to test hypotheses.
yp
and of using

Biostat-1 Hertanto

2

Biostatistics
S i i in
Statistics
i the
h area off biological
bi l i l and
d
health sciences

Biostat-1 Hertanto

3

The purpose of a biostatistics
. to provide the numbers / tables / graphics
that contain information about a certain
situation
. to present them in such a way that valid
interpretations are possible

Biostat-1 Hertanto

4

Increasing role of biostatistics
• Biostatistics provides a way of organizing
information on a wider and more formal basis than
relying on the exchange of anecdotes and personal
experience
y in
• More thinggs are now measured qquantitatively
medicine
• There is a great deal ofintrinsic variation in most
bi l i l processes
biological

Biostat-1 Hertanto

5

To answer these q
questions,, we rely
y on
the methods of biostatistics
• Is the new drug effective + safe?
• Does ht e use off seat bbelt
l reduce
d
the
h chance
h
of death in motor vehicle accident?
• Where
h should
h ld government invest
i
its
i
resources if it wishes to reduce infant

mortality?
• etc
Biostat-1 Hertanto

6

Population
p
and samples
p
Except when a full cencus is taken,
taken the data are for a
sample from a larger group called the population.
The sample
Th
l is
i off interest
i
not in
i its

i own right,
i h but
b for
f
what it tells the investigators about the populations.
Because of chance, difference samples give different
results and this must be taken into account when
using
i a sample
l to
t make
k inferences
i f
about
b t the
th
populations. This phenomenon, called sampling
variation
variation.
Biostat-1 Hertanto

7

Sampling and representativity
Target Population

Sampling
Population

Sample

Target Population Î (Sampling Population) Î Sample
Biostat-1 Hertanto

8

Sampling
the process of selecting units from a population of interest

Sa m plin g M ode l

Biostat-1 Hertanto

9

Langkah-langkah penelitian
Fakta

Teori
Masalah
Tinjauan Pustaka
Generalisasi

- Identifikasi variabel
- Kerangka Teori
- Kerangka Konsep

Hi t i
Hipotesis
Verifikasi
- Desain

- Sampel
- Instrumen
- Pengumpulan Data
- Analisis

Biostat-1 Hertanto

Simpulan

10

Analysis
• By the time you get to the analysis of your
data, most of the really difficult works have
been done.
• It's much more difficult to :
define the research problem; develop and
i l
implement
a sampling

li plan;
l
conceptualize,
li
operationalize and test your measurements;
and develop
p a structure design.
g
• If you have done these works well, the
analysis of the data is usually a fairly
straightforward affair.
affair
Biostat-1 Hertanto

11

Data analysis involves three
major steps
• Cleaning and organizing the data for
analysis

l sis (Data
(D t Preparation)
P
ti )
• Describing the data (Descriptive
St tisti s)
Statistics)
• Testing Hypotheses and Models
(Inf
(Inferential
nti l St
Statistics)
tisti s)

Biostat-1 Hertanto

12

Data Preparation
•

•
•
•
•

logging the data
checking the data for accuracy
developing a database structure
entering
t i th
the data
d t into
i t the
th computer
t
transforming the data

Biostat-1 Hertanto

13

Logging the Data
In any research project data come from
a number of different sources at
d ff
different
times:
• mail surveys returns
• coded
d d interview
• laboratory
• etc

Biostat-1 Hertanto

14

Checking the Data For Accuracy
• As soon as data are received you should screen it
for accuracy.
accuracy
• In some circumstances, doing this right away
allows you to go back to the sample to clarify any
problems or errors.
• There are several questions you should ask as part
of this initial data screening:
9
9
9
9

Are the responses readable ?
Are all important questions answered?
A th
Are
the responses complete?
l t ?
Is all relevant contextual information included
(e.g.
g data, time, p
place, researcher)?
Biostat-1 Hertanto

15

Developing a database structure

• Defining variables
• Entering the data into the computer
• Data transformations

Biostat-1 Hertanto

16

Defining variables
•
•
•
•
•
•
•

variable name
variable description/label
value
l labels
l b l
missing values
variable
i bl ttype ((numeric,
i string,
t i
d
date
t etc)
t )
column format (width, alignment)
measurement level (N O I R).
R)

Biostat-1 Hertanto

17

Entering the Data into the
Computer
• There are a wide variety of ways to enter the
data into the computer for analysis.
• In order to assure a high level of data
accuracy, the analyst should use a procedure
y
called double entry.

Biostat-1 Hertanto

18

Data Transformations
•
•
•
•

Recode
Compute
p
Select cases
Rank cases
etc.

Biostat-1 Hertanto

19

Modification of data files
•Opening an existing data file
•Defining
D fi i new variables
i bl
•Entering new data
•Inserting and deleting cases and
variables
•Saving data files
Biostat-1 Hertanto

20

A l i
Analysis
• Descriptive
• Inferensial

Biostat-1 Hertanto

21

Descriptive Statistics
• are used to describe the basic features of the data
in a study.
• provide
id simple
i l summaries
i about
b
the
h sample
l andd
the measures.
• together with simple graphics analysis,
analysis they form
the basis of virtually every quantitative analysis of
data.
• with descriptive statistics you are simply
describing what is, what the data shows.
Biostat-1 Hertanto

22

Inferential Statistics
• investigate questions, models and
hypotheses.
f
statistics to try
y to
• we use inferential
infer from the sample data what the
population
p
p
thinks.
• Thus, we use inferential statistics to
make inferences from our data to more
general conditions.
Biostat-1 Hertanto

23

•
•
•
•
•

STATA
BMDP
EPIINFO
PEPI
SPSS
etc

Biostat-1 Hertanto

Softwares

24