Dr Lizmy Ajith
Statistics is the field of science concerned with the collection, classification, summarising & interpretation of numerical data
Biostatistics: application of statistical methods in biological sciences to deal with living things
Medical Statistics
Application of statistics in medicine
- To compare efficacy of a drug
- The % cured, relieved or died in experiment
- To find an association B/W 2 attributes
- In epidemiological studies
Sources of data collection
- Experiments
- Surveys
- Records
Presentation of data
- Tabulation
- Diagrams
Sampling
- The process of selecting a representative part from the whole
- Sample: the representative part
- Population: the whole from which sample is drawn
Measures of central Tendency
- Mean
- Median
- Mode
Mean
Arithmetic average obtained by summing up all the observation & dividing the total by the No: of observations
X = ∑ X
n
Eg: E.S.R of 7 subjects – 7,5,3,4,6,4,5
Mean = 7+5+3+4+6+4+5 = 4.86
Median
The middle observation when all the observation are arranged in ascending or descending order
It implies the mid value of series.
Eg: E.S.R. of 7 subjects arranged in ascending order are, 3, 4, 4, 5, 5, 6,7
The 4th observation i.e. ‘5’ is the median
Mode
Most frequently occurring observation in a series
Eg: the size of induration in tuberculin test of 10 boys – 3, 5, 7, 7, 8, 8, 8, 10 ,11, 12
The mode here is ‘8’
Measures of variability
Variability: biological data collected by measurement or counting.
No 2 measurements in man are absolutely equal
Measures of variability of observations help to find how individual observations are dispersed around the mean
Measures of variability
- Range
- Mean deviation
- Standard deviation
- Standard error
Range
The normal limits of a biological characteristic
Eg: systolic B.P 100-140mm of Hg
Mean deviation
Mean of the absolute deviation from the central tendency
MD = ∑ X – X
n
Mean deviation about mean
= sum of the absolute deviation from the mean / No: of observation
Standard deviation
Square root of the average of the sum of the squares of deviations taken from the mean
√ ∑ (x – x )2
n
Indicates whether the variation of difference of an individual from the mean is by chance or real
Standard error
Difference between sample & population values
Is a measure of chance variation
SE = SD / √n
Correlation
The relationship or association between two quantitatively measured variables.
Eg: relation between cholesterol level & B.P
Regression
- Change in the measurements of a variable character
- Helps to predict the value of one character from the knowledge of the other character
- Eg: to estimate height when weight is known
Probability or chance
It is a ratio of occurrence of favorable chances out of the total possible outcomes.
Eg: chances of one drug being better than the other
Test of significance
- Mathematical methods by which the probability of an observed difference occurring by chance is found
- The common test in use are ‘Z’ test, ‘t’ test & ‘Chi square’ test.
Stages in performing a test of significance
- State null hypothesis(Ho) : statement of no differenceEg: vitamins A & D makes no difference in growth
- State alternate hypothesis(H1): vitamins A & D play a significant role in promoting growth
- Determine the probabiility(P) of occurrence of your estimate
- Draw conclusion on the basis of ‘P’ value i.e accept or reject the null hypothesis – decide whether the difference observed is due to chance or due to vitamins A & D
Paired ‘t’ test
- Applied to paired data of independent observations from one sample only when each individual gives a pair of observation
- To study the effect of a drug before & after
- To compare the effects of 2 drugs
t = x
SE
Dr Lizmy ajith, B.H.M.S, M.D(HOM)
Dept. of Practice of Medicine,
Govt. Homoeopathic Medical College,Kozhikode
Be the first to comment