Practical 3: Biostatistics

Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

Depart. Of Community Med. Biostatistics Pract.

2
Al-Kindy College of Med
ِِ
Practical 3
Descriptive statistics; Tables and Graphs
Descriptive statistics serve as device for organization, summarization of data.
{{Measurements that have not been organized summarized or otherwise manipulated are
called ''raw data''}}.
Organization and summarization and the displaying of data ‫عرض البيانارن‬can be done by the
following methods.
1- Tabular (Tables):
2- Graphs (figures)

Tables and graphs are often displayed the distributions, which are useful:
(a) To detect any trends in the data
(b) To detect “unusual” scores in a set of data
(c) To quickly present/summarize a great deal of info

3- Numbers (mathematics)
Single number to summarize the whole data

Note: The purpose of tables and graphs is to present information in a concise way so
that readers can comprehend ‫ افهم‬and remember it more easily.
Tables and Graphs should stand alone in a report.
 This means that the reader should not have to refer to the text in order
to understand and interpret the information in them.
 In practice this means, they require descriptive titles and clear,
meaningful labels.

1- Tabular (Tables):
Table consists of row(S) & column(S), could be 2x2, 2x3….etc.
List is the simplest form of table, consists of two columns only, the first giving
an identification of the observation unit and the second giving the value of the
variable of that unit.
Table must be:
a- As simple as possible (it is better to have 2-3 simple tables than one
complicated).
b- Understandable & self explanatory without references to the text. This is
done by:
 The title should be clear (placed above the table), and answer
the questions of: What? Where? And When?
 Each row and column should be labeled clearly and concisely.
 Specific unit of the measure for the data should be defined.

1
Depart. Of Community Med. Biostatistics Pract.2
Al-Kindy College of Med
ِِ
 Total should be placed.
 Illustrate symbols, code, and abbreviation by putting a footnote
below the table.
c- Source of the table (if not original).
d- Avoid too much over ruling.

Tables can display two types of data


1- Categorical
Ex: 12 subjects were tested for blood type: A, B, O, A, AB, B, O, B, A, A, A, AB

2- Numerical
Ex: The ages of 15 subjects were: 14, 15,16 20,22, 27, 28, 31, 33, 37, 48, 50, 59, 60, 60

Data must be grouped in classes

2
Depart. Of Community Med. Biostatistics Pract.2
Al-Kindy College of Med
ِِ
2- Diagrammatic "Graphical","Figuer".
Graphs represent pictorial representations of the distribution of data. Graphs should
present a clear and accurate picture about the data and convey an easy-to-understand
message. Graphs include labels with units of measurement, a title that accurately
describes the contents of the graph, and a scale on each axes that neither distorts nor
exaggerates the data. Sample size should be included on the graph or legend.
Graphs versus Tables
• Advantages : Simplicity, clarity
Easily remembered visual image
Picture of complex relationships
Emphasis
Popular
• Disadvantages : Lack of precision
Lack of flexibility
Provide for distortion

 The form of the diagram varies according to the nature of the data:
a- For categorical data, we have "chart", this can be:
 Bar chart; This is a graphical representation of the (relative) frequencies or
magnitudes by rectangles of constant width drawn with length proportional to
the (relative) frequencies or magnitudes concerned, as the EX. Below:

15

10
No.

5 12
6 5
0
2
Medical Surgical Cardiac Other

Types
Fig: the distribution ofaccording
of patients ICU patints.
to ICUs admission.
We have: 1- Clustered bar chart 2- Segmented bar chart

3
Depart. Of Community Med. Biostatistics Pract.2
Al-Kindy College of Med
ِِ
Useful for displaying the distribution of more than one categorical variable

10

Male
5
Female

0
A B AB O

Male 4 7 6 3
Female 5 2 4 8

15

10 Female
5 Male

0
A B AB O

Female 5 2 4 8
Male 4 7 6 3

Bar chart can be plotted horizontally instead of vertically


Conditions for Which Patients were referred for treatment

Back and Neck


Arthritis
Anxiety
Skin
Digestive
Condition

Headache
Gynecologic
Respiratory
Circulatory
General
Blood
Endocrine

0 20 40 60 80 100 120
Number of Patients

4
Depart. Of Community Med. Biostatistics Pract.2
Al-Kindy College of Med
ِِ
 Pie chart; This is a graphical representation of the (relative) frequencies or magnitudes by
a circle whose area represent the total frequency and which is divided into segments
which represent the proportional composition of total frequency, as the EX. Below:

8%
20%
Medical
48%
Surgical
Cardiac
24%
Other
Types of ICU patients

 Picto-chart ((Picto-gram)). This is a graphical representation of the (relative) frequencies


by using symbols (drawing or picture) relevant to the subject matter. Symbols of
different size should not be used. A unit value of the data should be represented by
standard symbol which may repeat to represent magnitude. As in the Ex.

Parity 0 ♀♀♀♀♀

1-3 ♀♀♀♀♀♀♀
≥4 ♀
Fig. Pictogram, No. of mothers (all ages) in 100, 000.

 Flow chart. Flow-chart. The sequence of series of events is often illustrated by flow chart,
as in the Ex.

Data

Quantitative Qualitative

Discrete Continuous Nominal Ordinal

5
Depart. Of Community Med. Biostatistics Pract.2
Al-Kindy College of Med
ِِ

b-For numerical data


 One dimensional dot diagram. This is a diagrammatic representation of the distribution of
a variable in which every observation is marked as a dot corresponding to its value on a
line (usually horizontal line) calibrated within the range of values of the variable. As in the
Ex.

 Arithmetic Scale Line Graph

Secular Trends in never smoking among adolescent boys and girls


aged 12 through 17 years: California, 1990-1999

75

70
Percentage Never Smokers

Female
Male

65

60

55
1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
Year

6
Depart. Of Community Med. Biostatistics Pract.2
Al-Kindy College of Med
ِِ
 Scattered diagram (dot-graph). Each observation is marked as a dote corresponding to its
value on each axis (X & Y). The pattern made by the dotes is an indication of relation
between the two axes, which may be linear if the follow a straight line or curved if not.

 Histogram: This is a graphical representation of frequency distribution in which rectangle


proportional in the area to the frequencies are erected on the horizontal axis. The base
lines are continuous (because we are dealing with continues variables). The width of the
rectangles should be equal. As in the Ex.

7
Depart. Of Community Med. Biostatistics Pract.2
Al-Kindy College of Med
ِِ
 Frequency polygon: If we join the tops of the rectangles in the histogram→Polygon
(total area of histogram= total area of polygon). It is only appropriate when the variables
on the horizontal axis are continues and there is only single value on the vertical axis. As
in the Ex.

*Advantage: can present more than one set of data.


100
90
80
70
60
50
40
30
20
10
0

8
Depart. Of Community Med. Biostatistics Pract.2
Al-Kindy College of Med
ِِ
Analyze _______________ Descriptive Statisrtics ________Frequency distribution
______ Chart ( we choose the suitible graph )_______________ok

You might also like