Cogn Chris © Biostatistics * Definition * Derivation of Term © History of Biostatistics * Statistical Approach * Applications of Biostatistics + Types of Statisties * Limitations of Statistical Methods + Aims of Statistics * Role of Biostatistics in Modern Research ‘= Limitations of Statistical Methods © Statistical Terms © Notations Used in Statistics 1.1.1 Definition of Statistics © Achenwall ‘the Father of * Acconling to A.L. Bowley * © According to Croxton and Cowden * collection, classification, presentation, in 1749 defined statistics as the political se atistics is the science of estimates and probabilities c study of statistical methods an analysis and int pretation of numerical data, a from it 1.1.2 Derivation of Term Statistics tistics’ is derived from the Latin word ‘status’, the Italian word ‘statistica’, the French word ‘statistique’, istik’, AIL these words are used for political state or a government. quantitative information about population, ‘The term * or the German word Initially, statistics was used by Kings and emperors for coll land, income and taxes ete, Nowadays statistics is used to collect, document, analyse and interpret quantitative information in various fields, such as economics, finance culture, forestry, indus es, healthcare, pharmacology, human ‘or cattle population, education, and re production, ag ch in various fi 1.1.3 Statistical Approach Term statistics is used in two different senses: Singular form and Plural form. 1. Singular Form In singular form it means the seience of statistics or statistical methods. These statistical methods 3 tion, analysis, interpretation of quantitative data for reaching some conclusion ¢ collection, 3 Scanned with CamScannerDitferent steps of statistical wettinds ‘Step! on siops EL a Frain Sao Tari es] steps ee Fig. 1.1 Diferent steps or stages in singular form of statistics 2. Plural Form In plural form statistics refers to the numerical or quantitative information collected in a systematic manner. statistical data with some definite ai or objective in view. In other words, itis used as the plural of the noun ss which refers to any one of many computed or estimated statistical quantities, such as the mean. the stand. or the correlation coefficient, 1.1.4 Types of Statistics Statistics is divided into following five types: + Theoretical + Descriptive + Applied and + Predictive a 1, Theoretical Statist Theoretical statistics iy associated with the derivation an eerie and development of Various formulae used for compet ties 2. Descriptive Stat Descriptive sta : ‘al procedures that deal representation and processing OF summarisation of d, I with jes is the study of statist M informative and mie. comprehensive Scanned with CamScannerfor the purpo Wis basically a device foro conclusion He invalye drawn tone the The to focus its essential featur niu data for bringing: 1 with the implications or conclusion that can be “ eaphicol and tabular presentation, but 7 ooh data presented {Wo types oF descriptive statistic © Measures H tendency i mnean and median He. tanpe, variance and standard deviation Example: An example of are ich all the Indian residents J occupation. The data, thus feristics of the population at deseriptive 110 provide such inten obtained, ts ¢ s the decennial census in India, in w al status, education smut praphs to reflect the chiar a given place Hil Statisties al statistics invalves those 1 the condition dures which are used to draw an inference aby © population by studyiny attributes of some small samples drawn from that population The inference is considered ay a peneralisation about the Known ay sampling statisties. Besides drawings interence methods of sampling on the basis of natin on. This branch of statistics iy also arge popu tial statistics also helps in suggesting proper infer of population, Example: An example of inferential statisties in which physician will have only limited the eff tients to find the efficacy of the drug. y of a new hypertensive or a new cancer druj ies 4. Applied St The statistics used in collecting data about econ and poliucal matters ts known as applied Statistics, Records aind figures related with per eapil aypeets of human population fall in this category. ase in population, fertility and mortality and other 5. Predi The statistics used to make predhetions Statistical techniques like mean, standard devia biostatistics. ve Statistios sing the data available, 1s known as predictive jon, correlation and normal probability are used in predictive TABLE 1.1 Differences between Descriptive, Inferential and Predictive Statistics ial Statistiey Predictive Statistics 11 includes It includes + Collection of data + Collection of samples from a large + Analysis of data population + Organisation of da + Making inference based on data obtained + Making predictions after analysing the Organisation of d ie from samples + Making hypothesis as generalisation + Providing the future conclusion + Determining relationship. I is also called sampling statistics + Summarisation of data F data + Present 1.1.5 Aims of Statistics tistics is basically concerned with three-fold purpose: data through experimental investigation and sample surveys. data in suitable tables, diagrams, charts or graphs, ete collected, put forth definite interpretations or predict the future fe the statist e and represent the id inferences from the data «To gener ie lat © To draw outcomes from the Scanned with CamScanner| Methods all kinds of phenomeri xr all our doubts. They and cannot answel 1.1.6 Limitations of Statistical atistical methods cannot be applied 10 certain limitations. or chemical laws. They are based on the average ates and not as precise measur individual fact for every individual. Thus the statisticay ot exact laws Tike mathem fed as approximations OF te of Facts an stical Laws. . as est id not the 1 data ean be trea © Statist «Statist ; es and are not (rue «They are derived by taking a majority of eases and are inf sare uncertain jive aspects, like beauty, goo. : se data onl. It ignores quaitative aspeets TRE Beatty, s00dne, a numerical shape technique deals with the q y, ete, They cannot be studied unless nclusions. ¢. ation is useless’ is st ven the argument that ‘in a country, 15,000 vaccinated intelligence, hone atistically defective, since we are not told ight lead to fallacious € va + Statistical results persons died of small pox. Therefore, ff persons who were not aceinated and died. the pereer sciences. «Statistical technique is same for the social as well as for phys helps to iptions. It mere ng policy pres «Tis only a tool and not an end itself, It cannot help in form throw light on the phenomenon, a sound knowledge of statistical methods ean knowingly or unknowingly can draw faulty © The greatest limitation of statistics iy that only one who efficiently handle statistical data, Persons with poor expertise conclusions © Manipulations of figures can lead to wrong conclusions. © Statistics neither proves anything nor disapproves anything. REQ siostaristics 1.2.1 Definition of Biostatistics Biostatistics combines two fields of study: © Bio stands for biology, the study of living thing «Statistics is concemed with the accumulation/collection, tracking, organisation, summary, classification, analysis and evaluation of facts to reach some inference. Therefore: " oo ‘Biostatistics is the application of statistical methods to a wide variety of fields of biology or life sciences neluding human biology, medicine, public health, agriculture, veterinary microbiology and geneti Ys 2 genetic Biostatistics is also called biometry, literally mea ly meaning biological measu ; Su. al measurement. Biometry has Greek origin (b fe + metron = ough, the etry has Greek origin (bios Ii etron = measure), Although, the (Wo tenns are used interchangeably, biomet ne iologic nd agricultural applications and biostatistics for medical usage. It involves a ty is largely used for biological echniques to scientific research in health related fields including moi ff environmental study. bel ‘Modern statistics and biostatistics are not the met re devic i ‘ces for collecting numerical data but a means ‘es development and application of statistical icine, epidemiology, public health, and areas of: + developing sound techniques for handling and analysing the ¢ + drawing inference from the collected data using various tech os AL present it is impossible to identify even a sin isle field of human activit activity whether it is biol logy, agriculture, business management, economies, demoy " . demography, redicines--which cannot do without statistic. MY: eowraphy, education, pharmacology prkbece ote ore ee Scanned with CamScanner1.2.2 History of Biostatistics + Adolphe Quetelet (1796-1874), the Bel stical methods for the first i Jgian astronomer and mathematician used sta : 1 {0 the problems of biology, medi J developed statistical theory as a general ind sociolo; method of researc plicable to any observational science. Jed the *Father of Biostatistics and ics, anthropometry, and psychology + Francis Galton (1822-1911), first cousin of Charles Darwin, is ugenies’, He made notable Our present underst to hil contributions in the fields of he edity, cus landing of correlation, the measure of agreement between two variables, ty also credited + Karl Pearson (1857-1936) applied statistical foundation for the “descriptive a heredity rests on statistical b; 1 of natural selection and laid the wethods in the demonstr ind correlational statistics’. He also emphasised that the whole doctrine of + The term biometry was coined by WER. Weldon (1860-1906), a zoologist at University Coll + Ronald A. Fisher (1890-1962) was a domina mple theory are used in almost .e, London nd biometry. Hiy contributions to small 1.2.3 Applications of Biostatistics Application of biostatisties is not restricted 1 cert of these applications are as follows: ain experi yenty but is used. wide variety of comests. Son ies The birth of science of zenctics is based on the statistical use by Mendel to analyse the results of F, a ‘The statistical methods generally applied in classical or Mendelian square tests are used to determine the deviation between observed and expected frequencies of a phenotype in generation. It can also be ascertained whether the outcome of a cross confirms to the resi occurrence and e ties are ehi-squ ts oF Mendel's laws. The imation of percentage can also be worked out with the aid of statistical procedures 2. Statistics and Population Population geneties is con nges occurring in it over generations. The evolutionary changes that occur in a population are due to forces of evolution such as mut migration, isolation and selection, The frequencies of different genes sand changes in their freq} effect of these forces ed with the application of different statistical methods. ‘There are also effects related to small population size and consequent inbreeding, The application of statistical methods have made it possible to trace the consequences of these effects in a population through abstract models ies wed with studying genetic stn ire of populations and ion, ies due to th be estin To study the behaviour of genes in a population which is concerned with changes in the frequency of genes in the population statistical methods are applied, ¢.g.. the relationship between allelic and genotypic freq 1g binomial equation. tics and Statistics ‘The traits exhibiting continuous variation, which are often controlled by two or more genes are termed as polygenic and where several genes make additive contribution to the phenotype, the trait is known as quantitative character and it exhibits continuous variation, To study such traits various statistical methods are used by the geneticists, like in twin studies, Weinberg's differential method is used to estimate the frequencies of monozygotic or dizygo 4. Statistics in Molecular Bi Statistical methods are used molecules and amino acid sequ evolutionary and phylogenetic rel normalities in different human populations. molecular biology and bioinformatics for sequence analysis of DNA and RNA ing in proteins from different specimens. Such statistical data help in tracing lationships, in identifying genetic disorders and in distribution of congenital Scanned with CamScanner™ arious activities im a rinology vari and kndoceia 4 inference of or the ¢ jon analysis are contributed for th sin a eh pistey, Physinloss 5. Statistics Rive timation Statistical techniques have areal Correlation and rear uation st at and enilo ad in cell Po sn of effect of Vari bi hare teats are extensiVelY ae Ful in the estimation of eT ests sel 1 physiolo endocrinologi factors on fe analysis is found in endocrine gland. ical experiments. Multiva cular cell, function of a system Or a resented in tabular form. Such tab ture and Forestry inultiple use in the mar i ds, forest produce and yie fees ction per year al are called yield tables, These provide an idea of production pel ATM crop production, forest production, animal husbandry, fish production, i © Use ¢ construction of yield tables. ee aa cit ia i isthe apa a ession function with tree volume a mn dep " “ 7 y y ume yield of a tree is the output of a regression fu ction of a set of polynomial or diameter andor height ay independent variable(s), In practice the best suited fun ial models selected is based on some goodness of fit criteria. ee test the effect of different types of manures, levels of irrigation, varie knowledge is required. 6, Statistics in Agri from animals are TEP ind have 1k production, etc, The standard regression exponen T analytical data, a thorough statis 7, Statistics and Forest Mensuration In forestry, the objective is to measure characteristics of trees, ie, length, area some cases, measuring the characteristics may be easy but quite difficult in many other attributes which are difficult 10 be measured directly in terms of easily measurable characteristics is the crux of ‘mensuration problem. Since the prediction is probabilistic in nature, the science of statistics plays a major role in this field of canopy. volume and wei ses. The prediction of those 8. Si Biomass refers tothe total mass of living material in a given locality or a given are: The estimation of biomass has ‘sumed considerable importance in recent years. The common procedure of estimation is throu; zh the u cauations anu! andard tables. In biomass estimation many theoretical refinements Ye Stepwise regression, non-linear regression, ete, are used extensively, istics and B ¢ of regression weralised test. square 9. Statistical Ecology Ecologists frequently m habitats, Studies on veg ure the environmental variable atiables aloy ° ; the environ ne with the observations on the orga ; al versus environment are amendable to such Gomelation analysis is called multiple discriminant anal Variates [ten the unis for observations are discrete, a simple tes fo sample is provided by ehisquare value computed out ae Studied together, the results are not straight tors Two major aspects of interest with many speet of directivity. The species relat § ab ‘ lations are stud uundance relations and measurement and geometrical distribution. These funct ox-normal, nen i imal frome dsb al, negative binomia aa mami of in duals and make i possible o pre P between a number speci the observed number of species in the samples my i lation from in the sample. This is o, c e ple. This ie vn ‘es in the whole population from isms in their is I analysis. An extension of canonical * operates on several sets of ilar itatively si of two specie . the conti : Pecies oF groups within the pontingency table. When more than two species are Ping of the frequency classes. 10, Statistical mental Mana The use of statistical methods is environmental biolo statistical technique. Scanned with CamScannerlogy, biostatistics is helpful in finding out () the action of a drug, (i) the comparison of the action of two different drugs or two successive dosages of the same drug, (dif) in evaluating the relative potency ofa new drug with respect to a standard drug, and (iv) the efficiency of « newly manufactured drug or injec is tested by using the test or test of significance. Statistics is used in drug: development to cure diseases and 10 determine people's reaction to those drugs through null hypothesis. y and Public Health Community and public health include epidemiology, health services research, clinical trial, health eare, environmental health, health management and biomedical engineering. Health managers are expected to take sound decisions on the basis of available information from various clini - al trials. on the © effect of use of sera and vac * comparative data obtained by comparing pereentage of deaths and attacks among vaccinated persons with the percentage of deaths among unvay + evaluation of spread of a disease ‘+ cure t0 various deadly diseases = genetic defects or genetic ma sal communi level to check spread of epidemics, ‘nated ones. Iconditions responsible for specific diseases, All above information are made available to the health man organise data and analyse differences ger by the field officers who collect samples, 13. Medical Biostatisties Everything in medicine be it research, diagnosis, treatment, epidemiology. use of treatment techniques or percentage of success depends on data collected through counting or measurement, This is called medical statisties or quanti medicine. The purpose of data collection is to" + compare various medical parameters like blood pressure, pulse rate, RBC count, WBC count, haemoglobin estimation of human beings at different age and in different conditions ‘+ compare the efficiency of a particular drug, operation or line of treatment, the pereentage cured, relieved or died in the experiment and control groups. + find an association between two attributes such as cancer and smoking or blood pressure and life style. identify signs and symptoms of a disease or syndrome. Proportional incidence of one symptom or another indicates whether it is a characteristic feature of the disease. «success of a particular drug in the treatment of a disease, and side effects of the drug. ‘The raw data when organised, compared, analysed and interpreted, help medical practitioners and researchers to reach some conclusion. Since medical observer gives his opinion or develop a regime of treatment of human problems, he needs to be well conversant with statistical analysis, For professional interpretation of his results, he or she should learn the application of statistical methods himself, Medical statistics supports following different fields: © Health services including both public and community health. « Medical treatment related to cause of disease, environmental conditions responsible for the spread of a disease. living and feeding habits, efficacy of drug, method of treatment, etc. «Vital statistics in demography including events of birth, deaths, infant mortality and level of activeness in geriatic persons, 1.2.4 Limitations of Statistical Methods « Statistical methods cannot be applied to all kinds of phenomena and cannot answer all our doubts. They have certain limitat ‘© Statistical laws are not exa ms. ct laws, like mathematical or chemical laws, They are based on the average, Scanned with CamScannerot as precise measur or estimates and not as ps Statistical data can be treated as eee . ey not the individual fact + Statistical data ca eee Statist th the aggregate ° for every individual, Thus » * Statistical nee oa patie erate eaten 7 + They are derived by ta inferences are uncertain, Pr aareat * Statistical technique deals with the cvantiatve unless given a numerical shape intelligence, honesty, etc. They cannot + the argument that “in a country 15,p9) noes qualitative aspects, like conclusions, * Statistical results might lead to fallacious co eae SSeS ecination is useless’ ‘ons died of small-pox, therefore, vacci Sgt the percentage of persons who were not vaccinated and died. * Statistical technique is same for the social as well as phy ape * It is only a too! and not an end itself. It cannot help in formulating policy throw light on the phenomenon, no. * The greatest limitation of statistics is that only one who has a sound Knowledge of statistical meth efficiently handle statistical data, Persons with Poor expertise knowingly or unknowingly: can conclusions, sciences, Od ea raw faulty * Manipulation of figures ean lead to wrong conclusions, * Statistics neither proves anything nor disapproves anything, 1.2.5 Role of Biostatistics in Modern Research Modern researchers in biology cannot Compiling data, calculating averages, variables and then finally do without biostatistics. Al] Tesearches are data based. Biostatistics 1S used in ponraying variabilities, calculating. percentile, tracing relationship hetwcen "eaching Some conclusion by matching the observations with biostatistical ¢ |. Calculating averages: Averages Provide a picture of the performance of 2. Variability of measurements Using the average as a point variations and modifications about the central point. This * Graphs, tables, linear Fepresentat ‘alculations, ‘Arious groups under study of teference, one can determine 8 called variability of measu ions and figures the spread of ments, are prepared to portray the sen different variables is called correlation ® It helps to Understand the relationship raw inference of growth index of tracing between dosage and response relationship between h characters of a large popu, 8. Determining the ‘ation can provide information about the MtiON can, be drawn, » Correlating their scares in these tests lligence tests are expected to obtain highest Acader 9. Determining rel determined by between these Wo sets of data fe 10. Comparison of performance: ‘tne Performan differences can be tested, Nee OF 10 op The Felativity of als With s, Measurement instruments is sme Parallel devices. The correlation same ‘alculated, By comparing the performance of ong SPOUP With another. difference ean be determines. Scanned with CamScanneraT STATISTICAL TERMS 1.3.1 Population ‘The word ‘population’ has a diffe represents the number of people living totality of individuals of a given species per given time in all the individuals of a species that interbreed, But a statistical population refers to any well defined group of individuals who are being studied or the total number of observations of a particular type about which inferences are to be made. More simply, a group of study elements is called population, For example, all university students in Delhi could be a population. All patients of a hospital suffering from tuberculosis, patients suffering from breast cancer and treated with a new drug or haemoglobin levels of all adult women of a community may be considered as a population. aning in statistics than in ordinary usage. Ordinarily, the word population an area, a region or a country. Biological definition of population is th: population means a group of ven area, In geneti 1.3.2 Sample Sample is a small group or subset of a population selecte can be used for investigat researchers want {0 find out some specific Feature about a population, but it is not possible to study every single individual in the population. They select a small number of individuals from the population, study them and use that information to draw conclusions about the whole population. This is called sample. For example, we want to study the average height of students studying in Class IX of different Senior Secondary Schools in Calcutta or Bangalore. It is not necessary to observe the height measurements of all the students. In fact, we can take a small representative sample of a few students from different schools for measurements represents all the attributes of entire population and and can give results. Sampling may be Random or Biased 1. Ina simple random sample every member of the population has an equal chance of being selected. No one in the population is favoured over other in the selection process 2. Biased sample or non-random sample is the sample is drawn with a purpose. It does not provide equal ‘opportunity to all members of the population of being selected. 1.3.3 Unit Unit is the smallest object or individual or group th small subareas of land, individual patients, etc. Units are expressed in two ways: can be investigated as the source of basic information, e.g.. (a) Sampling units during surveys. (b) Experimental units during experiments. 1.3.4 Parameter Parameter is any numerical property, characteristic or fact that is descriptive of @ population. Population mean, median and mode are the ‘parameters of location’. These relate the numerical values of a variable with the general location of a population. Usually all the characteristics of a population can be specified in terms of a few e-ga if the characteristic is length and the measurements of length are variable then the mean length can be regarded as a parameter. ‘Since parameters are descriptions of the popula 0 symbols are used to represent same parameters in populations meanings and methods involved in the calculation of these parameters. jon, a population can have many parameters, Moreover, different and samples because there are differences in the Scanned with CamScannerAk Pc ion Parameters and Samp) sed to Indicate Population Param i BLE 1.2 Symbols u: i | le Statistics [ Symbol for Population Symbol for Sampr Term _Sombel for Poputtion _Srmbel for Sample # ‘Mean f Variance "i # Standard deviation ———_—____ aI 1.3.5 Treatment a In statistical studies, the samples or groups of eeces data are analysed, each sample is referred as a treatment ve being compared are called treatnen, © hen 1.3.6 Data or Statistical Data «situ palate a set of facts expressed in quantitative or qualitative form in a manner autabi Based on the typeof sources used in their collection the data can be classic under |: Primary data: Data collected by the investigator from personal experimental studies or surveys, for presentation ang Analysis the following iyo types: Examples: + Interviews of people on smoking and alcoholism. * Measurement of height of students in a class, * Effect of a drug on a group of patients, 2. Secondary data: Data obtained from the records or secondary sources such as journals, magazines, newspapers or internet, Variable is a quality or a Characteristic which is being observed Temeasured and can vary from one individual to another. For example, animals of same species may differ in their length, weigl Sex, etc. These characteristics The pnables. Thus, variable can be defined seni gauti¢s by which individuals dit Measurement of age, categories, Includes: 1 Discrete variables * Continuous variables 2. Qualitative Variable is on, can be expressed i 3. Random variable: Whenever the hey leaves, ete as a value of the respective variable When 7 individual is determined, the result is referred Med a random variable. Values oby ta # TeSult Of chance factors, the variable is of following two types: Sedu ae deni ae Scanned with CamScannerpossible value ‘or example, 0 + of patients each rof «Discontinuous oF discrete random» Dominus ov eee random vara «ne which maple of whi al ot daly paticnts visiting a peneral hptal a day is represented by whole alues or absence of values in between. F fe the numbe 275 or $.97 oF the numbe i a discrete random variable, be wumber, such as 5, 4, 10, 20 and not as 1S rooms in a house oF the number of persons 23.4 Serr elel aie Gta aT oa See ee ae eee al values such as 2 «Continuous random variable is one which can ta nee \ can take any numeral value within a certain range, be. Ht does ws from hot possess any gaps oF interruptions. The W30 brn to 150 P The height of a child at various ages when he assumes all possible values within the limit even in fractons: Value, they exhibit variation When observations within the data for a partic Variation between observations ean be due put it may also be due to diet or disease, A variable can be dependent or independent, A. i Went, A variable is called dependent variable when it ts of interest. It changes in response to some intervention a - 1 variable do not have the »: many factors. For example, the vanation in human hei ‘The independent variable is that variable which is itself being: manipt 1.3.8 Subscripts and Summations Subscripts and suffixes are used to distinguish the different values of a variable. If the weight of ane species of fish is symbolised by X, then the weight of different fishes of the same species may be denoted by Ny. Nox Nye oS ‘The total weight of all the fishes caught from a pond will be X, +X, +X + This may be simplified by wr EX, where (¥) symbolises the summation of all the observations. Wt tells us to ‘um up from first value of X ap to the last value of X (X,). This can be simplified by writing the sum as EX 1.3.9 Inference Inference is simply a conclusion about a popul 1. Subject matter inference is the inference from a sample of units to the popufations of units Statistical inference is a conclusion about a population on the basis of information contained in 4 Wwomen aged 40 to 49 are subject matter inference nferences about their heart tical inferences, lation. It can be of two types ge inferences about Indi: rates as recorded in the study are We proceed from a sample of observations to popu to a population of units form a population of observations byw whet inference and wv of observations by way of a statisti, nce. ‘of @ subject matter infer to its tue value, 1.3.10 Accuracy ess of a measured oF Accuracy shows the closene computed valu 1.3.11 Precision Precision is the closeness of FP {precision are ted measurements oF the same quantity used synonymously in everyday speech, but in st Meld inaccurate but precise weight, Simi, an ynvensitive Seale smh tess there is bids in a measuring instrument, precision will Teal ics they are used Usually, accuracy differently. A biased but sensiti rate but imprecise rea result in-an ac accuracy. We need, to be mainly conce ned with the former. 1.3.12 Parametric Methods nce that are based on the assumption that the population has a certain probability thods. By these The methods of statistical infer distribution, the resulting collection oF Inethods iaference is made about ane OF MO Gt tatstniion and Fedistibution are associated with the distribution. stntcal tests and procedures are referred as parametric Gf the probability distribution parameters. For example, values of parameters of an assumed normal probability 1.3.13 Nonparametric Methods gi do not Fr nanparame! jes are nonparametric. assumptions of any form of probability distribution from which frie methods. These are also called distribution free methods, For edlures: thi re known as jequeney techniq The statistical pre its E0ITe, chi-square £1 Scanned with CamScannereae nt in parametric statistics cannot always be met in — logic inhere! se assumptions are: Generally, the thee assumptions that 2% °° cessary. These assump studies. This makes the use of nonparan hat are normally distributed. ‘Samples are drawn from populations that here is homogeneity of variance, ie., the Variance by ee ears « assumed that ther on. The F-max test can be used to qe + Inthe analysis ofthe variant is sa ee ariation. The determin groups are homogeneous within the the homogeneity of variance. ith equal intervals. inuous measures Wi + The measures to be analysed are continuous ™. NOTATIONS USED IN BIOSTATISTICS = fF population. Roman letters are used for statistics of samples and Greek for the parameters of pop' z Summation of observations. It is capital P(x) of X__ Probability of happening Greck letter pronounced as sigma. 1 Total number of observations (sample x Mean or arithmetic mean or sample mean. size) (read as x bar), AM Arithmetic mean Population mean GM Geometric mean a Equal to HM Harmonic mean * Not equal to SE 5 or > Greater than ee Standard error of standard deviation < Lesser than [ oy SEM Standard error of mean re as test Student's test or “1° ratio E . . Fa Deviation (obtained from actual mean) Conelation coetficient x’ Deviation (obtained from actual mean) . Number of standard deviations from M, naa the mean s ; on 3 Mean deviation (MLD, g Sample standard deviation s Population’s standard deviation d Variance of sample o Population variance Deviation from the mean o Assumed mean yi Modulus of Wi way of class-interval YG Logarithmic base = 2.718 © ‘orrection - Infinity v Frequency v Square root oe Degree of freedom I Union of ‘ Number of groups oF classe f eee ia Expected number * m i Section of sets, o Observed number if 7 value of class i i Ut lass boundary q i PPer class boundary x Q Class width o class interval y Estima value or the dependen Q Pestie deviation N a number of observationsTotal aan’ Qe Lower quanite sample R Upper quantile (Population sizey = yy n Rank Pis3igy Scanned with CamScanner