Classification and sources of experimental errors.

The error is defined, as we had said, as the difference between the true value and the one
obtained experimentally. Errors do not follow a specific law and their origin is due to multiple
causes.

Depending on the causes that produce them, errors can be classified into two large groups:

A systematic error is called an error that is constant throughout the entire measurement process
and, therefore, affects all measurements in a defined way and is the same for all of them. These
errors always have a certain sign and the probable causes can be:
- Instrumental errors (devices); for example, instrument calibration error.
- Personal error: This is, in general, difficult to determine and is due to personal limitations. Such
as, for example, parallax errors, or visual problems.
- Measurement method errors, which correspond to an inappropriate choice of the measurement
method; which includes three different possibilities: the inadequacy of the measuring device, the
observer or the measurement method itself.

Accidental errors are those that are due to small variations that appear between successive
observations made by the same observer and under the same conditions. The variations are not
reproducible from one measurement to another and it is assumed that their values are subject
only to the laws of chance and that their causes are completely uncontrollable for an observer.
Accidental errors mostly have a very small absolute value and if a sufficient number of
measurements are made, as many positive as negative deviations are obtained. And, although
corrections cannot be made with accidental errors to obtain values that are more consistent with
the real ones, statistical methods can be used, through which some conclusions can be reached
regarding the most probable value in a set of measurements.

https://www.ugr.es/~andyk/Docencia/TEB/Errores.pdf

One way of measuring is to take the normal curve, the Gaussian bell curve, as a reference. Our
measurements will fall somewhere on the curve; If the results are very different, we will speak of
an abnormality. It is valid that the value of two consecutive standard deviations be considered as
the cut-off point for normality. According to this statistical criterion, the distribution of normal and
abnormal is constant for any variable.  Normality and abnormality.  Sources of variation. 
Random error.  Systematic error. Error sources

https://es.slideshare.net/RossiGarciaRoman/errores-en-la-mediciones-y-fuentes-de-error-
41628013

The arithmetic mean is the average of a set of numbers, a 1 , a 2 , a 3 , . . . , a n , obtained

by adding all the numbers and dividing it by n. This is a way to find a representative value
for a set of numbers.

http://www.mathematicsdictionary.com/spanish/vmd/full/a/arithmeticmean.htm

It is mainly used to know the deviation that the data present in their distribution with
respect to the arithmetic mean of said distribution.
https://okdiario.com/curiosidades/2017/07/13/desviacion-tipica-1133576

Root mean square error (RMSE) measures the amount of error between two sets of data. In other
words, it compares a predicted value and an observed or known value.
https://acolita.com/que-es-el-error-cuadratico-medio-rmse/
The square root of the variance of the distribution of the statistic is called Standard Error
and provides a measure of reliability for the statistical data obtained from a sample since
the standard deviation is a measure of the sampling error (of the variation in the sample ).
http://www.fca.proed.unc.edu.ar/mod/book/view.php?id=3273&chapterid=187
In statistics , a histogram is a graphical representation of a variable in the form of bars, where
the area of each bar is proportional to the frequency of the values represented. They serve to
obtain a general "first view", or panorama, of the distribution of the population, or the sample,
with respect to a quantitative and continuous characteristic.

https://es.wikipedia.org/wiki/Histogram

The process to construct a histogram is very simple.

 Two axles are built.

 On the horizontal axis, or abscissa axis, the data or modalities obtained
are represented. In our case: 18, 19, 20, etc.
 On the vertical axis, the ordinate axis, the frequencies of each data or
modality are represented with numbers.
 Bars or rectangles with the same base are raised on the horizontal axis
up to the height of the numerical value of the frequency of each
modality. In our case: 18 until 2, 19 until 3, 20 until 14, etc.
 The bars are drawn next to each other.

https://marcelomendizabal.wordpress.com/2017/02/09/que-es-y-como-se-construye-un-
histgrama/
ERROR IN DIRECT MEASUREMENTS
SYSTEMATIC ERRORS:
These are errors related to the skill of the operator.
 PARALLAGE ERROR (EP), this error has to do with the position taken by the operator
to read the measurement.
 ENVIRONMENTAL AND PHYSICAL ERRORS (Ef), when climatic conditions
change, they affect the physical properties of the instruments: expansion, resistivity,
conductivity, etc.
Also included as systematic errors are calculation errors, errors in automatic data acquisition,
and others.
Most systematic errors are corrected, minimized, or tolerated; In any case, its management
depends on the skill of the experimenter.
MEASURING INSTRUMENT ERRORS:
They are errors related to the quality of measuring instruments :
 MINIMUM READING ERROR (ELM), When the numerical expression of the
measurement turns out to be between two marks on the scale of the instrument's reading.
The uncertainty of the value is corrected by taking half of the minimum reading of the
instrument.
Example: minimum reading of 1/25 mm
Elm = ½ (1/25mm) = 0.02 mm
 ZERO ERROR (E0), is the error of uncalibrated instruments.
Example: when the minimum and main reading scales do not coincide, the reading will be seen
to be deviated to one side of the zero of the scale. If this deviation were less than or
approximately equal to the minimum reading error, then

RANDOM ERRORS:
They are errors related to interaction with the environment , with the system under study, they
appear even when the systematic errors have been sufficiently minimized. balanced or
corrected.
Random errors are quantified by statistical methods. If n-measurements are taken of a physical
quantity x, the readings being x1, x2, x3,…,xn; The estimated value of the physical magnitude x
is calculated by taking the average as follows.
https://www.monografias.com/trabajos82/mediciones-errores-laboratorio-fisica/
mediciones-errores-laboratorio-fisica.shtml

The significant figures of a number are those that have real meaning
and, therefore, provide some information. All experimental measurements
are inexact and must be expressed with their significant figures. Let's look
at a simple example: suppose we measure the length of a table with a
ruler graduated in millimeters. The result can be expressed, for example,
as:
Length (L) = 85.2cm
This is not the only way to express the result, as it can also be:
L = 0.852 m
L = 8.52 dm
L = 852mm
etc…
 Regardless of how it is expressed, the result has three significant
figures, which are the digits considered true in the measurement. They
meet the definition because they have a real meaning and provide
information. Thus, a result like
L = 0.8520 m
It does not make sense since the instrument we have used to measure is
not capable of resolving to ten-thousandths of a meter.
http://www.escriboscientificos.es/trab21a40/cifrassignificativas/00cifras.htm