How to Collect Data

Ram Kumar
Assistant Professor
DPOST, MRSPTU
 DAT
• A data as facts or figures
A dictionary defines
from which conclusions may be drawn.

• A collective recording of observations

either numerical or otherwise is called data
Statistics is a tool for converting data
into
Statistics

informati
Data on Information

•But where then does data come from ?

•How is it gathered ?
•How do we ensure its accurate ?
•Is the data reliable?
•Is it representative of the population from which
it was drawn?
5.3
 Methods of Collecting Data

• Direct Observation
• Experiments, and
• Surveys.
 Methods
Mailing paper questionnaires to respondents,
who fill them out and mail them back
Having interviewers call to respondents on the
telephone and ask them the question in a
telephone interview
• Sending the interviewers to the
respondent’s home or office to administer
the questions in face-to-face interviews
Depending upon the of
the variable data
nature is classified
broad categories
into 2

• Qualitative Data

• Quantitative Data:-
1. discrete

2. continuous
• Qualitative data :-
(characterized by words)

when the data is collected on

the basis of attributes or qualities like
sex, malocclusions, cavity etc.
• Quantitative data :- (characterized by
numbers)
when the data is collected through
measurement, like arch length, fluoride
concentration etc.

• Discrete data :-
when the variable under
observation takes only fixed values like whole
numbers.

• Continuous data :-
if the variable can take any value in
a given range, decimal or fractional.
Quantitative: numbers breadth generalizability
Qualitative: words depth specific

Remember, "Not everything that counts can be counted."

 Quantitative methods –
Qualitative methods
Quantitative Qualitative

Surveys Focus groups

Questionnaires
Tests Unstructured
interviews
Existing databases Unstructured

observations
Common data collection methods
• Survey • Testimonials
• Case study • Tests
• Interview • Photographs,
• Observation videotapes, slides
• Group assessment • Diaries, journals,
logs
• Expert or peer
reviews • Document review
and analysis
• Portfolio
reviews
 Are the data reliable and valid?
• Validity:
Are you measuring what you think you are
measuring?
• Reliability:
if something was measured again using the
same instrument, would it produce the same
(or nearly the same) results?
• The main sources of data are :-
1) surveys
2) experiments
3) records in OPD

• Data can be collected through either :-

1) primary source
2) secondary source
• Primary Source :-
Here the data is obtained by the investigator
himself. This is a first hand information.
• Advantages :-
Precise information and reliable.
• Disadvantages :-
Time consuming, expensive.
• Primary data can be obtained using :-
1) Direct personal
2) Oral health examination
3) Questionnaire method
• Secondary Source :-
The data already recorded is utilized
to serve the purpose of the objectives of
study.
Ex:- The records of opd of the dental
clinics.
 Data Gathering Techniques
• Observation

• Questioning
 Data domains
Cognitive -- paper and pencil
Affective -- interview
Psychomotor -- observation

Data Presentation
The objective of classification of data is to make the data
simple, concise, meaningful and interesting and helpful in
further analysis.
There are 2 methods of presenting data:-

1) Tabulation
2) Diagrams
• Tabulation :-
Is the first step before the data is used for
analysis or interpretation.

• In the process of tabulation the following

type of classification are encountered.

1) Geographical i.e area wise

2) Chronological i.e on the basis of time
3) Qualitative i.e. according to attribute
4) Quantitative i.e. in terms of magnitude
 TABLE FORMAT :
Number of Primary Health Centres in India at 5-year Intervals during 1970-1990

YEAR No.of PHCs

1970 5015

1975 5293

1980 5484

1985 7284

1990 18,981
• Classification by Space :-

1) Data are classified by location of occurrence

2) Arrangement of set of categories in

alphabetical order of the terms defining these
categories,

3) In the order of their geographical location may be

found to be suitable in many cases.
 Table 2
Number of Literates in Rural and Urban Areas in India (state wise list).

No.of literates ( in thousands)

STATES
RURAL URBAN
Andhra Pradesh 14,821 10,119
Assam 7921 1709
Bihar 20,358 6485
Maharastra 22,164 20,774
Gujarat 12,096 9179
Kerala 16,443 6228
Madhya Pradesh 14,464 9027
Tamil Nadu 17,424 12,958
Karnataka 12,267 8812
Orissa 10,303 2608
Punjab 6253 3699
Rajastan 8189 5428
Uttar Pradesh 33,079 13,968
West Bengal 20,337 12,382
TOTAL 216,119 123,376
• Chronological i.e. On the basis of time :-

1)In this case data are classified by time of

occurrence of the observations.

2)Arrangement of categories is almost

always in chronological order.
Table 3

Number of Primary Health Centres in India at 5-year Intervals during 1970-1990

YEAR No.of PHCs

1970 5015

1975 5293

1980 5484

1985 7284

1990 18,981
• Classification by attribute :-
1)When the data represent observations made
on a qualitative characteristic the classification in
such a case is made according to this qualities.

2)Alphabetical arrangement of categories may

be suitable for general purpose table

3)In the case of special purpose table

arrangement may be made in the order of
importance of these categories
Table 4

Patients who attended Leprosy Clinics during 1993

Type of leprosy No. of patients

Tuberculoid 604

Lepromatous 272

Indeterminate 72

Borderline 48

TOTAL 996
Classification by the size of observations :-
1.when the data represent observations of
some characteristic on a numerical scale,
classification is made on the basis of the individual
observations.

2.The range of observations is suitable divided

into smaller divisions called class intervals.

3.The numerical scale adopted may be either

discrete or continuous.
Table 5
Distribution of Single Live Born with respect to Birth Weight

Birth Weight (grams) No.of children (n)

< 1500 24
1500 - 1999 60
2000 - 2499 422
2500 and above 3202
TOTAL 3708
Principles (in tabulation of data) :-
1. Every table should contain a title, should be
concise and meaningful.
2. The tables should be numbered.
3. The heading of columns or rows should be clear
and concise .eg: ht in cm, age in years , wt in kg
etc.
4. The number of class intervals should be sufficient
to condense the data bringing out their significant
features.
5. The class intervals should be at equal width.
6. Uniform size class intervals are preferable.
7. Sometimes open end class intervals are used.
8. The class intervals should be clearly defined
to avoid ambiguity .eg- 0-4, 5-9, 10-14 etc.
9. Units of measurements should be specified.
10.If the data is not original , the source of the
data should be mentioned at the bottom of
the table.
11.Groups should be tabulated in ascending or
descending order.
12. If certain data is omitted or excluded deliberately, the
reason for same should be given.

13. For many arithmetic calculations the midpoint of each

class interval will be used as a representative of each item
in that interval.

14. The interval should be so chosen that the mid point of

each interval is approximately the average of the items in
that interval.

15. It will be convenient if mid points and limits are whole

numbers.
16. The class intervals should be same through out
table except in case of age:
• 0-<1 (infant up to 1yr)
• 1-4 (toddlers 1 to 5yrs but not completed
5 yrs)
• 5-14 ( School children < 15yrs)
Presentation by Graphs and Diagrams:-
Diagrams and graphs are extremely useful because

They are attractive to the eyes

1. Give a birds eye view of the entire

data

2. Have a lasting impression on the mind

of the layman

3. Facilitate comparison of data.

Basic rules in the construction
of diagrams and graphs :-
1. Every diagram must be given a
title
2. It should be simple
3. The vertical axis is always labeled
as the ‘y’ axis. It is also
“ordinate”.
4. The horizontal axis is always
labeled as ‘x’ axis . It is also called
“abscissa”.
5. The x axis and y axis meet at
right angles at a point called
origin (o)
6. The values of variables are
presented on the x axis and the
frequency on y axis
7. The number of lines drawn in any graph should
not be many so that the diagram does not look
clumpsy.
8. The scale of presentation for the x axis and y axis
should be mentioned at the right hand corner of
the graph
9. The scale of division of two axis should be
proportional and the division should be mark
along with the details of the variables and
frequencies presented on the axis.
Presentation of quantitative data is
through graphs, the common
graphs in use are:-

1.
Histogram
2. Frequency polygon
3. Frequency curve
4. Line graph
5. Scatter or dot diagram
Presentation of qualitative data is
through diagrams, the common
diagrams in use are:-

1. Bar
diagram
2. Pie/sector diagram
3. Pictogram or picture diagram
4. Map diagram or spot map
 Line diagram:
• This diagram is useful to study changes
of values in the variable overtime.
• Simplest type of diagram.
• On the X axis the time such as hours,
days, weeks, months or years are represented.
• The value of any quantity pertaining to
this is represented along the Y axis.
OP

500
450 467
400
380 400 402 380
350 346 367 354
NO OF OUT-PATIENT

300 300 289

250 250 265
200
150
100
50
0
Feb-06

Apr-06

Aug-06

Nov-06

Dec-06
Sep-06
Jun-06

Jul-06
May-06

Oct-06
Mar-06
Jan-06
Frequency polygon:

1. The most commonly

used graphic device to
illustrate statistical
distribution.
2. Used to represen
frequency t
distributio
of quantitative data.
n
3. Useful to compare 2 or
more frequency
distributions.
• A frequency polygon is a variation of a histogram, in
which the bars are replaced by lines connecting the
midpoints of the tops of the bars.
• Advocates of the frequency polygon argue that the
purpose of a histogram is to show the shape of the
data distribution and removing the bars makes the
shape clearer and smoother.
Construction of frequency polygon:
1. To draw a frequency polygon, A point is marked in the
midpoint of the class interval, corresponding to the
frequency.
2. Then these points are connected by straight lines.
3. The first point and last point are joined to the
midpoint of previous and next class respectively.
4. Rather than leaving the graph suspended in space ,
we assume that there is another interval above and
below which is having frequency of 0 .
5. so the midpoint of group are assume
frequencyhaving
of 0.the group is now allowed to meet x-
axis on both the ends.
 Advantages of frequency polygon:
• It is very easy to construct and very easy
to interpret.

• It is useful in portraying than

more distributionson thesame
graph colors. So it is very useful to comparetwo
different 2 or
more than 2 distributions. paper with
 Frequency curve:-
When the number of observations are very large and groups
are more (ie; small class intervals) the frequency polygon
tends to loose its angulation and it forms a smooth curve
known as frequency curve.
 Scatter diagram or dot diagram:
1. It is a graphic presentation of data.
2. It is used to show the nature of
co- relation between 2 variables
 Bar diagram
1. This diagram is used to represent
qualitative data.
2. It represent only one variable.
3. The width of the bar remains the same
and only the length varies according to the
frequency in each category.
4. There are 3 types of bars:
a) simple bar
b) multiple bar
c) component bar diagram
• Simple bar:
The limitation of this method is that they can
represent only on the classification and hence
cannot be used for comparison

500 467 OP

450
400 402
400 380 367 380
346 354
350
300 289
NO OF OUT-PATIENT

300 265
250
250
200
150
100
50
0

Oct-06

Nov-06
Jul-06
Feb-06

Jun-06
Jan-06

Aug-06

Dec-06
Sep-06
Apr-06

May-06
Mar-06
• Multiple bar:
1. This diagram is used to compare qualitative data with
respect to a single variable.
2. This diagram is similar to the bar diagram except that
for each category of the variable we have a set of bars of
the same width corresponding to the different section
without any gap in between the width and the length
corresponds to the frequency.
OP
IP

46
500

7
450
40

40
38

38
0

2
36
400
0

0
35
7
34

4
6
350
30
NO OF PATIENT

28
0

9
300
26
25

5
0

250
200

15
4
12

150 10
0

90
86

78
70
67

100
60

59

48
45

50
0

Nov-
Feb-

Aug-
Apr-

Dec-
May-

Sep-

Oct-
Jun-
Jan-

Jul-
Mar-

06

06
06

06

06
06

06
06

06
06

06
06
Component bar diagram:
• This diagram is used to represent qualitative
data.
• It is desired to represent both the no of
cases in major groups as well as the subgroups
simultaneously
 Histogram:
this diagram is used to
depict
quantitative data

1. It is bar
a diagram
bars.
without gap
between
2. If we draw frequencies of
each group or class
intervals in the form of
columns or rectangles
such a diagram is called
histogram.
3. It represents a frequency
distribution.
The histogram is constructed as follows:
1. On the X axis the size of the observation is
marked.
2. Starting from 0 the limit of each class interval is
marked, the width corresponding to the width of
the class interval in the frequency distribution.
3. On the Y axis the frequencies are marked
4. A rectangle is drawn above each class interval with
height proportional to the frequency of that
interval.
Advantages of Histogram:

Easy to understand

Disadvantages of Histogram:

• Only 1 histogram can be placed at a time.

• More time consuming to construct than a
frequency polygon.
 Pie diagram:
1. These are popularly used

to show percentage break downs for qualitative

data.
2. It is so called because the entire graph looks
like a pie and its components represent slices
cut from a pie.
3. A circle is divided into different sectors
corresponding to the frequencies ofthe
distribution.

4. Some knowledge of circles and

degrees is necessary
5. The total angle at the center of the circle is 360 degrees and it
represents the total frequency.

6. After the calculation of angle, segments are drawn in the circle and the
segments are shaded with different shades or colors and an index is
provided for the shaded colors

7. Cannot be used to represent 2 or more data sets

 Pictogram
1. Display of data through pictograms was
initiated by Dr Otto Neurath in 1923.
2. Data are displayed by the pictures of the
items to which the data pertain.
3. A single picture represents a fixed no.
4. They are the least satisfactory type of
diagrams.
5. They are inaccurate too.
Map diagram or spot map or cartograms:

1. These maps are used to show

geographical distribution of frequencies of a
characteristics.
 References

1. Kirkwood BR, Sterne JAC. Essentials in Medical

Statistics. 2nd edition. Oxford: Blackwell; (2010).
2. Jay S. Kim And Ronald J. Dailey. Biostatistics
For Oral Healthcare. Blackwell Publishing
Company. 2008
3. C.R Kothari. Research Methodology methods and
technologies. 2nd edition. New age international
private ltd publishers; 2004. reprint 2007
THANK
S