8.processing

Processing
&
Analysis of data
D.A. Asir John Samuel, MPT (Neuro Paed),
Lecturer, Alva’s college of Physiotherapy,
Moodbidri

Dr.Asir John Samuel (PT), Lecturer, ACP

Processing operations

• Editing

• Coding

• Classification

• Tabulation


Editing

• Process of examining the collected raw data

• Editing is done to assure that data are
accurate, consistent with other facts gathered,
uniformly entered, as complete as possible

• Field editing

• Central editing

Field editing

• Review of reporting forms by the investigator
for completing, translating or rewriting

• Individual writing styles

• On the very next day or on the next day

• Not correct errors of omission by simply
guessing

Central editing
• Take place when all forms or schedules have
been completed and returned to fitness

• Correct errors such as an entry in wrong place,
wrong month, and the like

• Respondent can be contacted for clarification

• No bias


Coding

• Process of assigning numerals or other
symbols to answers
• Should be appropriate to research problem
under consideration
• Necessary for effective analysis
• Extraction of data

Classification

• Large volume of raw data is reduced into
homogeneous group
• Arranging data in groups or classes on basis of
common characteristics
• Classification according to attributes
• Classification according to class-intervals

Tabulation

• Arranging in concise and logical order

• Summarising raw data and displaying in
compact form

• Orderly arrangement of data in columns and
rows


Tabulation is essential because of

• Conserves space and reduces explanatory and
descriptive statement to a minimum

• Facilitates process of comparison

• Facilitates summation of items and detection
of errors and omissions

• Basis for various statistical computations

Problems in processing

• Problem concerning “Don’t Know” responses

• Use of percentages


Problem concerning “Don’t Know” responses

• When DK group is small, it is of little significance

• In big group, it becomes mater of concern

• Actually may not know the answer or

• Researcher may fail in obtaining appropriate
information (failure of questioning process)

• Keep as a separate category in tabulation

Use of percentages
• 2/more percentages must not be averaged
unless each is weighted by group size

• Too large percentages should be avoided
because difficult to understand and confuse

• Hide base value

• Real differences may not be correctly read

• Can never exceed 100 percent and for decrease

Statistics in Medical Research

• Documentation of medical history of disease,
their progression, variability b/w patient,
association with age, gender, etc.

• Efficacy of various types of therapy

• Definition of normal range

• Epidemiological studies

Statistics in Medical Research
• Study the effect of environment, socio-
economic and seasonal factors

• Provide assessment of state health in
common, met and unmet needs

• Success/failure of specific health programme

• Promote health legislation

• Evaluate total health programme of action

Statistics in Medical Research - Limitation

• Does not deal with individual fact

• Conclusion are not exact

• Can be misused

• Common men cannot handle properly


Normal distribution

• Represented by a family of infinite curves
defined uniquely by 2 parameter the mean
and the SD of the population
• The curve are always symmetrically bell
shaped. The width of the curve is defined by
population, SD


Normal distribution
• Mean, median and mode coincide

• It extends from - ∞ to + ∞

• Symmetrically about the mean

• Approx 68% of distribution is within 1SD of
mean (68.27%)

- 95% - 2SD (1.96 SD)

- 99% - 3SD (2.58 SD)

Normal distribution
• The total area under the curve is 1

• The value of measure of skewness is zero. It is
not skewed

• The curve is asymptotic. It approaches but
never touches baseline at extremes

• The curve extends on the both sides -3σ
distance on left to +3σ distance on the right

Normal distribution - Uses

• Construct confidence interval

• Many statistical techniques makes an
underlying assumption of normality

• Distribution of sample means is normal

• Normality is important in statistical inference


Skewness
• Measure of lack of symmetry in a distribution

• Positive skewed

- Right tail is longer

- Mass of distribution is concentrated on left
side

- Distribution is said to be right skewed

Negative skewed

• Left tail is longer

• Mass of distribution concentration on right
side

• Distribution is said to be left skewed

• Value of skewness is 0 for normal distribution


Kurtosis

• Measure of degree of peakness in distribution

• For normal distribution, value of kurtosis is 3

• Leptokurtic – High peakness

• Mesokurtic – normal

• Platykurtic – Low peakness


Descriptive statistics

• Measures of location

- Central tendency

- Mean, median and mode

• Measures of variation

- Dispersion

- Range, quartile, IQR, variance and SD

Mean

• Sum of all observation divided by total no. of
observation


Mean - merits

• Well understood by most people

• Computation of mean is easy

• More stable

• All items in a series are taken into account

• Used in further statistical calculation

• Good basis for comparison

Mean - Demerits

• Affected by extreme values

• Cannot be computed by mere observation

• Not suitable for skewed distribution

• May not be an actual item

• Not in qualitative data


Median

• Middle most observation when data is
arranged in ascending/descending order of
magnitude

• Divides number into 2 halves such that no.of
items below it is same as no.of items above


Median

Odd = n+1/2

Even = n/2 + (n+1)/2
2


Median - Merits

• Widely used measures of CD

• Not influenced by extreme values

• Can be determined if extremes are not known

• Not a typical representation of series

• Useful for skewed distribution


Median - Demerits

• When no. of items are small, median may not
be representative

• It is effected by frequency of neighboring
items

• Not a typical representation of series


Mode

• Most frequently occurring observation in data

• If all values are different then no mode


Mode - Merits

• Can be computed by mere observation

• Simple

• Precise

• Less time consuming

• Less strain


Mode - Demerits

• Not an amenable to further algebraic
treatment

• Not rigidly defined

• Affected by no. of frequency of items


Measures of Dispersion (variation)

• Range

• Interquartile range

• Variance

• Standard Deviation


Range

• Difference between largest and smallest value

Range = Largest no. – Smallest no.


Quartile

• Value that divide data into 4 equal parts when
data is arranged in ascending order

Q1 = (n+1/4)th ordered observation

Q1 = [2(n+1)/4]th ordered observation

Q3 = [3(n+1)/4]th ordered observation


Interquartile range

• Provides range which covers middlemost 50%
of observation

• Good measures of dispersion if there are
extreme values

IQR = Q3 – Q1


Variance

• Sum of squares of difference of each
observation from mean, divided by n-1

𝜀 𝑥−𝑥 2
Variance =
𝑛−1


Variance - Merits

• Easy to calculate

• Indicate the variability clearly

• Most informative


Variance - Demerits
• Units of expression of variance is not the same


Standard Deviation (SD)

• Square root of variance

𝜀 𝑥−𝑥 2
SD = √
𝑛−1


Standard Deviation - Merits

• Most widely used

• Used in calculating standard error


Standard Deviation -Demerits

• Lengthy process

• Gives weightage to only extreme valves


8.processing

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to 8.processing (20)

More from Asir John Samuel (20)

Recently uploaded (20)

8.processing