Statistics

 Statisticscan be defined as the methodology

of making inferences and decision making
by making a math-based analysis on the
available data or new data to be collected
about the research topic.
 When we look at the literal meaning, “The
job of collecting data methodically and
indicating it as a number in order to draw a
conclusion appears as enumeration”.
 In this course, statistics should be regarded
as a science / art that enables the analysis of
the data in the field of study of the scientific
research carried out and the result of the
scientific research in an objective, simple
and explainable manner.
Pre-Statistics Period

 Before statistics took its place on the stage

of history as a methodology, Humans had to
choose the way of explaining subjects with
fortune-telling and divine concepts, rather
than making analytical thinkings and
inferences about interpreting what
happened and predicting the future.
 In ancient Greece (circa 2000 BC), people
applied for a kind of fortune in which 4
anklebones were thrown to predict what
would happen to them.. Due to its geometric
structure, the anklebone can stand in
balance only on 4 sides. And the probability
of standing on these faces is determined to
be approximately 40% - 40% - 10% and
10%.
 Thebest future for an ancient Greek was
that the 4 different faces of the 4 anklebones
would overlap. This situation, which
corresponds to a probability of about 1/300,
was not calculated at that time. It has been
described as "Venus's Luck" in reference to
Venus, the goddess of chance and future.

Yarborough Duke
 From ancient times to the first age, the task
of interpreting what happened, making
sense and predicting the future passed to
the clergy. Clergymen, who have enormous
sanctions and influence over the people,
preferred to interpret people's destinies
according to their birth dates and
horoscopes.
 Moreover, they tended to associate the
dates of people's death with their
horoscopes. B.C. In 216, in the war between
the Roman Republic and the Carthaginians,
the Roman Army suffered one of history's
greatest defeats against Hannibal's army,
and approximately 75,000 Roman soldiers
lost their lives in a single day. This situation,
1.5 centuries after the war, was questioned
by the Roman statesman Cicero as follows;
"Were all those who died from the same
horoscope?"
The Birth of Statistics

 In order to understand gambling, which is his

passion, Geromalo Cardano wrote the book "Games
of Chance" in 1565, even though the "equal" (=)
sign was not established yet.
 Recognized by many of us for his execution order by
the Inquisition, Galileo Galileo, in 1583, at the order
of his boss, the Grand Duke of Tuscany, published
his article to answer questions such as 10 in total
when 3 dice are rolled, and why they come more
often than 9 in total.
 Named as the founder of statistics, John Graunt,
examining the birth and death records in London,
concluded that in 1662 the population of London
was around 384,000. Until then, the population of
London was thought to be 2,000,000.
 In 1654, at the insistence of his friend Antonie
Gombaud, a famous gambler, Blaise Pascal discovered
the "Binomial Distribution" while working to solve the
proposition known as the "Points Problem". This
invention, which forms the basis of probability, became
known as "Pascal's Triangle".
 Abraham De Moivre, in his book "Law of Chances"
written in 1733, interpreted the pascal triangle as the
distribution that best describes the "Law of Errors".
 This interpretation was taken into account by Carl
Freidrich Gauss in 1810. He defined a differential
equation parameterizing the Pascal triangle. Today, we
know this differential equation by the name "Normal
Distribution Function".
 Robert Price, considered the founder of actuarial
science, published the "Bayes Theorem" in 1763,
upon the will of his close friend, priest Thomas Bayes.
 Francis Galton, Charles Darwin's cousin, found that in
the last half of the 19th century, there was no
significant difference between the lifespan of kings
and church members and those of other professions,
and concluded that praying had no effect on life
expectancy. He also defined the law of "regression to
the mean", one of the most well-known properties of
randomness.
 Something caught his attention when Simon
Newcomb was working on books containing
logarithm tables in 1881; he found that the pages of
the books on numbers starting with 1 were more
dirty and wrinkled than the pages about numbers
starting with 2, and this continued until the pages
about numbers starting with 9, which looked
relatively clean and new. This situation, known as
Bedford Law, is currently used in the United States
to detect tax fraud.
 Ancient Egypt
 Al-Kındi
 Tyco Brahe – Johannes Kepler
 Edmond Halley
 Abraham Wald
 https://medium.com/@ilyasvarol/bir-%C3%BCr%C3%BCn-y%C3%B6neticisi-d%C3%BCnyan%C4%B1n-en-kanl%C4
%B1-sava%C5%9F%C4%B1n%C4%B1n-gidi%C5%9Fat%C4%B1n%C4%B1-de%C4%9Fi%C5%9Ftirebilir-mi-90d3841
a63be

 Ronald Fischer
 John Wilder Tukey
BIOSTATISTICS

 Mathematics and statistical techniques; is a science

that includes its applications in health sciences,
makes new adaptations specific to this field and
produces new methods.
 Biostatistics; It can also be defined as a branch of
science that deals with the identification, counting,
regulation and evaluation of the characteristics of
living things and all other elements affecting these
characteristics.
VARIABLE

 The observed properties of the units are called

variables.
 Properties that can take different values from a unit
to a unit and whose values can be expressed
numerically are called.
 The unit, which shows measurement differences
depending on random effects, is also called
variable.
 Variables are divided into two groups according to
the way of observation:
 A- Qualitative Variable: Values that can be specified
by counts, not by unit-based measurements, that
can be symbolized to carry (or not) a certain
structural and spatial feature.
 B- Quantitative Variable: These are objective
properties which can be determined and symbolized
by various measurement techniques and units.
 Variables are divided into two groups according to
the mathematical state of the measured values:
 A- Discrete Variable: These are variables that can
only have integer values within the set of values
and are expressed as rounded integers. Number of
children, number of days of hospital stay, number of
cigarettes smoked per day etc...
 B- Continuous Variable: Variable (integer and
fractional) variables within the set of values (within
the definition range). Height, weight, age, systolic
blood pressure, creatinine value etc…
 Subject: the unit in which experimental studies are carried out.
In experimental studies, the object, goods, experiment
animals or individuals are called subjects. It is the unit of
experimental study.
 Data: data is a set of values of one or more variables that
have been obtained and saved from two or more units.
 PARAMETER: Typical value of the studied variable in society.
The parameter is the calculated or estimated numerical value.
It is the typical value of the variable in society that
summarizes the set of observations. For example, values such
as population mean, variance, incidence, relationship
coefficient of a variable are parameters.
 Statistics: N are typical values calculated from the data
obtained from an example of a number of units. The
equivalent of the parameter in the example. Statistic is an
estimator of the parameter.
MEASUREMENT LEVELS
 From the point of view of the number system, there
are important distinctions between the scales due to
the features carried out by the numbers used and
the operations it allows. 3 main features that
distinguish between scale types:
 1- Real numbers are in a certain order relation.
(ROW)
 2- The distance between each real number is equal.
( DECEMBER )
 3- Real numbers have the starting point determined
by the zero point. (RATE)
 Classifier Measurement Level: It is the level of
measurement in which objects are generally
classified under certain headings and classes
according to their attributes. The data can be
displayed in different categories. Does not include
any of the number system properties. Blood groups,
gender, place of birth can be given as examples.
 Sorter Measurement Level: These are
measurements in which order of magnitude,
smallness and equality can be made and the class
names are determined in this order because of the
logical relations between qualitative characteristics
(sometimes quantitative concepts). Contains the
sequence property of the number system. Weak,
moderate, fat ranking; (I do not like, I like a little, I
like, I like the sort of "Likert Scale" is called.)
 Intermittent Measurement Level:
 It is a method of measurement in which the features to be
judged can be put in a specified order at precise intervals. In
this scale, the variable can be infinite between two specific
values. A value of 0 at this measurement level does not
indicate that there is no measured characteristic.
 In the same way, one of the measurement characteristics
cannot be expressed by multiple of the other. Includes range
and sequence properties.
 Temperature measurements can be given as examples. 0ºC
does not indicate the absence of temperature and 4º C is not
twice the temperature of 2º C.
 Proportional Measurement Level:
 The range of units of measurement can be infinitely
small, so that even the smallest differences can be
specified continuously and in units.
 All metric measurements are at this measurement
level.
 If the measured characteristic is 0, it indicates that
there is no characteristic. Similarly, a measured
characteristic can be expressed in multiples of the
other. The sequence includes all three properties of
range, ratio. Weight and length measurements are
examples.