FERMIN TORO UNIVERSITY

Master's in Educational Management and Leadership

STATISTICS IN SOCIAL SCIENCES
BARINAS - BARINAS STATE

BONILLA, MARIA.

MOLINA, JOSÉ.

MORALES, FRANK.

BARINAS, JULY 22, 2005

 INDEX

INTRODUCTION

SAMPLING 5

TYPES OF SAMPLING 9

PROBABILITY SAMPLING 9

NON-PROBABILISTIC SAMPLING 13

CONCLUSION

BIBLIOGRAPHY
 INTRODUCTION

The present work is titled SAMPLING, in which the most will be addressed

highlighting of this statistical procedure, and aims to study the types and

existing sampling techniques.

Here we will find basic concepts of population, sample, statistics.

inferential; characteristics, advantages and disadvantages of the sample; finally the types

of sampling, and some examples of them.

In this sense, it is important to highlight that the significance of sampling lies in that almost

we can never investigate an entire population; because it is very large, due to

economic reasons, lack of qualified personnel or to provide greater speed in the

process of collecting and presenting data. So, these are obtained from

starting from a sample of the population to be studied. Therefore, in the selection of the

sample must be taken into account: the size and selection of it. In this latter

aspect, the theme to be developed will be based.

The type of research used is documentary; since the information is

obtained through books and internet pages; which was analyzed, discussed and

process, to then be presented in a simple manner.

 Finally, most of the knowledge we have about the world is

obtains through samples; although most of the time with samples

inadequate; so one must be very strict in the selection of samples.

 5

SAMPLING

Sampling is a statistical process in which a phenomenon is studied that

it affects a large group or population, based on a selected or random sample

of the same.

The population is the set of all individuals that carry information.

about the phenomenon being studied. The individuals correspond to people, objects or

other elements that can be schools, courses, families, and even units

territorial.

The most common problems that arise in practice, when one

proceed to investigate the characteristics of a statistical population is the

impossibility of studying each of the individuals or objects that compose it;

therefore we must rely on a subset or part of the population as a basis for

to arrive at hypotheses related to various characteristics or parameters of it. For

thus, in these cases it is more convenient for reasons of economy, time and

efficiency analyze a sample that is a determined number of individuals or

population objects and assume that they can be seen in them to a certain extent

trust what is desired to know about the universe from where they come.
 6

The study of these aspects falls under inductive or inferential statistics.

whose purpose is to obtain knowledge about some extensive dataset, to

starting from samples of the same, which allow for the inference of the results

statistics, which represent arithmetic calculations performed on the values

obtained from the sample of the population that has been rigorously selected.

In order for these statistical data to be useful, they must present two

basic characteristics:

1. They must be relevant; that is, they must have a relationship with the topic.

investigate.

2. They must be unbiased; that is, they should not have distortions or errors.

originating from the instruments used in the data collection.

CHARACTERISTICS OF THE SAMPLES

For a sample to be acceptable from the analysis point of view

The statistician must meet the following characteristics:

It should be selected in a way that is representative.1of the population of

which was obtained.

1
It is a sample of an appropriate size that has been chosen by random procedures and is
consider that the observed characteristics "represent" or correspond to the population from where she
it comes from.
 7

The sample not only allows for obtaining estimates of characteristics of

the population (totals, averages, percentages, among others) must

provide measures of the accuracy of such estimates.

The sample should be quite small to avoid unnecessary expenses and the

wide enough for the sampling error to be acceptable and must also

bring together the demands of efficiency, representativeness, security and

flexibility.

ADVANTAGES AND DISADVANTAGES OF SAMPLING

Broadly speaking, the application of sampling is advisable mainly in

the following cases:

1) When the population is infinite or so large that enumeration is impossible.

complete each of the elements or units.

When the population has the same characteristics such that a sample is one

acceptable image of the same population.

3) Regarding the operational and administrative technical part we have:

a. Economy of time and money

b. Expansion of the scope of a research.

 8

c. Better training and control of the personnel involved in the research.

d. Greater control, quality, speed, and efficiency in processing and

publication of the data.

e. It allows for post-investigation verification using a minor

personnel and other resources.

f. Greater reliability and accuracy of the obtained data.

On the other hand, the application of sampling has some limitations.

common among which we have:

1) It does not allow for calculations, projections, or tabulations regarding sectors.

small ones from a population or when information is required for each element of

the population, since the doubts would be so important that it is not possible to have a

accurate prediction based on the sample.

The calculation of the results is subject to the errors of the sampling itself and of

errors unrelated to sampling.

Complete investigations offer a more solid and efficient margin of

accuracy obtained based on the sample.

4) Based on a sample, the cross-tabulation of certain characteristics is complicated.

investigated.

The probabilities of error multiply in the different operations and in the

complications that arise from its own mechanism.

 9

6) The statistical-mathematical preparation of the sampler for the efficient design of the

shows in all its stages.

TYPES OF SAMPLING

There are two general types for selecting samples from the

populations; the first is probabilistic sampling: these are those where the

The sampling operation is controlled objectively, that is to say, they are selected samples.

at random. Then, we have non-probability sampling: these are those in which the

sampling operation is carried out subjectively, at least partially, it

what it means that the results obtained from the sample they might not be

representatives of the population.

Probabilistic Sampling

SIMPLE RANDOM SAMPLES: in this type of sampling, each one of

The individuals of the population have the same chance of being chosen. Which

it can be with replacement and without replacement. In this case, to provide the security of

to ensure that the random sample is not biased, a must be used for its constitution

random table or random electronic processes.

 10

This type of sampling is more economical than other random procedures.

it also ensures the equiprobability of the choice, however, its limitation

is based on not providing enough cases of minority groups.

For its application, the following steps are necessary:

To make a list of all the members of the population and enumerate them.

each one of them.

To select the subjects of the sample, use a table of numbers.

random or a similar procedure using electronic means.

To manage the data collection instruments for the subjects

selected. Those subjects who for some reason are not subjected to

observation will integrate the experimental mortality category2.

For example; in the EU 'William Tell' located in Barrancas, Municipality

Cruz Paredes, the Physics teacher wants to understand the comprehension on the part

of the 9th grade students section A, on the topic of transformations and

clearing developed in class, for this, you need a sample of 5 students

out of a total of 36, which are listed in the folder of the

attendance.

The teacher proceeds to search for random numbers using the calculator, from

where he gets the following results: 15, 2, 3, 27 and 19.

2
It refers to the subjects chosen to undergo observation in a sample under study and
they cannot be located, or it is not possible to get them to provide the necessary information.
 11

CLUSTER SAMPLING: it consists of replacing the units

physical, elementary or latest to which the study refers, by units of

sampling that includes a group of those. Therefore, it is a type of sampling.

more economical and faster than simple random since it facilitates the work of the

field researchers. However, they may have some loss of randomness.

in the procedure and decrease in the accuracy of its results, in the case

that the number of conglomerates is less than 30. The steps to follow are the

same as the previous group (simple random sampling).

For example, the Head of the Educational Zone - Barinas wants to analyze the

performance in the 9th grade Physics area in the Barinas Municipality; for this

You need to select 50 Educational Units out of a total of 200.

Then, a list of all Educational Units where classes are taught is created.

9th grade; then the steps from the previous example are used.

STRATIFIED SAMPLES: it consists of dividing the population into

subpopulations, within each of which a random selection is made

simple as in the previous cases. This type of sampling allows for a sample

more homogeneous. In this case, we find the proportional stratified sample

which consists of distributing the sample proportionally to the number of

elements of each stratum and the uniform stratified sample which distributes by

The sample is the same among all strata. For example, in the 9th B section of the

UE "Padre Felipe Salvador Gilij", the influence of the variable is attempted to be related.
 12

students "male or female", with the variable "course approval" in the area

of mathematics. The information is collected with the subject teacher and the

the distribution is as follows:

Stratified sample of male and female students from the section of

9th 'B', UE 'P. Felipe Salvador Gilij'

NUMBER OF % OF THE
SEX SAMPLE % OF THE STRATUM
STUDENTS POPULATION

Men 16 52 5 31

Females 15 48 5 33

Total 31 100 10 32

SYSTEMATIC SAMPLES: This type of sampling consists of choosing a

random unit and then select each unit that is found among units of

that one. This technique tends to be faster and is less prone to errors than the

from simple random sampling. Generally, it is used when one wants to

select elements from a stratum, in a stratified design. For example, we

wants to select a probabilistic sample of 16 schools from a list of 160

schools.

One way to obtain a sample is: N/n = IM; 160/16=10

 13

Then, we proceed to select a number between 1 and 10, we take the 3, so

a=3

1. a = 3

2. a + 10 = 13

3. a + 2(10) = 23

4. a + 3(10) = 33

5. a + 4(10) = 43; We continue in the same way to obtain the others

results.

The schools with the numbers: 3, 13, 23, 33, 43, 53, 63 will be included in the sample.

73, 83, 93, 103, 113, 123, 133, 143, 153

NON-PROBABILISTIC SAMPLES

CONVENIENCE SAMPLING: It is the sample selected according to

to the convenience of the researcher; the elements of this type of sample are

chosen in a simple manner because they are accessible. They lack validity

external and reliability, furthermore they present sampling biases by criteria

arbitrary selection of subjects (even if the researcher does not notice them). By

example; it is desired to study the discipline of the College 'Father José María Veláz';
 14

for this, Professor Carlos takes the 8th grade section A as a sample, where the

fulfills functions as a guide teacher

INTENTIONAL OR RATIONAL SAMPLES: It is that sample

selected through the judgment of an expert in the field of study. Through

In this approach, the specialists choose what they believe is the representative sample.

It is fast and low cost. It does not require interviewer training. It is very

useful for exploratory studies as they allow the optimization of instruments

of observation. In addition, they allow for the formulation of hypotheses. Their limitation lies

in which it lacks external validity and reliability when trying to generalize towards

larger groups. For example; in the EB "María Montessori" located in San

Cristóbal, Táchira State, the psychopedagogue wants to study the low performance.

academic, for this purpose it takes as samples those students who present

learning difficulties.

SAMPLES BY QUOTAS: Among the types of non-probability sampling is

the most employed. It is similar to stratified sampling, as both types

the universe that is going to be studied is divided into groups or strata and is distributed the

total sample among the groups. It is fast and efficient, and it can also serve for a

preliminary essay of said technique, however, the interviewer bias for

Choosing subjects is its most evident flaw. For example; at the University of

The Andes - Táchira wants to study single students, with children or

pregnant women, or male students with long hair and earrings. So, each
 15

interviewer selects on their own and interviews with the help of a list

(directed interview).
 CONCLUSION

In the development of this research, some characteristics of the

samples, in this sense it has been noted that in practice it has been emphasized that the

need to select a sample where each element of the population or

the universe has a known probability of being selected, to achieve this

we must choose a type of probability sampling in order to minimize the error of the

sampling is simply assumed by chance.

On the other hand, the objective of a sample is that it allows to carry out

estimates of the values of the universe based on measurements obtained from it, and to

at the same time they allow for calculations of the safety or reliability of such

more accurate estimates using hypothesis statistical tests

about the universe.

Finally, it is necessary to know the theory of sampling to decide the

convenience of taking or not taking samples from a population considering the nature of the

research design and the peculiar characteristics of the project to confront the

Problem. Just as deciding the type of sample for a research project.

taking into account the variables of interest to relate, the type of population and the

proportions of the individuals.

 BIBLIOGRAPHY

Espasa Multimedia Encyclopedia (2000). Spain: Planeta.

GABALDON M., Néstor (1980) Some concepts of sampling (3rd Ed.)

Caracas, Venezuela: Central University of Venezuela, Faculty of Sciences

Economic and Social, Publications Division.

GONICK, L. and SMITH, W. The Statistics in Comic. Madrid, Spain: Zendrera

Zariquiey.

Unable to translate the provided URL as it does not contain text to translate.

The provided text is a URL and does not contain translatable content.

html

Unable to access external links.

Unable to access or translate the content from the provided URL.

Unable to access external content including the provided URL.