Introduction
Research is a process that, by applying the scientific method, seeks to obtain relevant and reliable information (worthy of faith and credit), to understand, verify, correct or apply knowledge.
In order to obtain some results in a clear and precise way, it is necessary to apply some type of research, research is closely linked to human beings, it has a series of steps to achieve the stated objective or to reach the requested information.
Research is based on the scientific method and this is the method of systematic study of nature that includes observational techniques, rules for reasoning and prediction, ideas about planned experimentation and ways of communicating experimental and theoretical results.
In addition, it has a series of characteristics that help the researcher to govern effectively in it and it is so compact that it has forms, elements, processes, different types, among others.
It is essential for the student and for the professional, because it is part of the professional path before, during and after achieving the profession; she accompanies us from the beginning of studies and life itself. For all kinds of research there is a precise process and objectives.
Research helps us to improve the study because it allows us to establish contact with reality in order to get to know it better, the purpose of this lies in formulating new theories or modifying existing ones, in increasing knowledge in order to elaborate theories.
The research activity is efficiently conducted through a series of elements that make the object accessible to knowledge and on whose wise choice and application the success of the research work will depend to a great extent.
That is why the objective of this documentary research is carried out, in order to give information about the series of procedures and steps that are needed to carry out a thesis: in the first section are the preliminary parts of the theoretical framework It covers from the delimitation of the topic to the determination of the variables, in section number 2; It describes what is an experimental investigation, the types of methods to be used and the classification of graphs to represent the statistical data of the tests or instruments to be used in an investigation.
With this work, you can refer to the importance of research as a learning process; Since it has a range of fundamental characteristics and which are narrowed in a very compact way in order to capture the information or to achieve the proposed objectives, it must be remembered that scientific research is a rigorous method in which a series of objectives is obtained. previously proposed and in a very technical way, and the research is the one that aims to expand scientific knowledge, without pursuing, in principle, any practical application and research is an action to clarify.
Theoretical framework
Theoretical framework is the synthesis of the general context (local, national and global) in which the topic of the proposal is located, current state of knowledge of the problem, gaps that exist and the void that one wants to fill with the project; why and how the proposed research, unlike previous research, will contribute, with probability of success, to the solution or understanding of the problem posed.
The main functions of the theoretical framework are as follows:
to. It guides on the knowledge of the type of research
b. It broadens the horizon of study, since it gives options of points of view
c. Leads to establishing hypotheses
d. Inspires lines of research
and. It foresees the way to interpret the data from the investigation
«A theory is a set of concepts, definitions and propositions related to each other, which present a systematic point of view of phenomena specifying relationships between variables, with the purpose of explaining and predicting phenomena »
The functions of theories
1. The most important function of the theory is to explain why, how, who, when, who where a phenomenon occurs.
2. Explain different manifestations of the phenomenon.
3. Describe how the phenomenon originates, evolves and affects.
4. Systematizes isolated and diffuse knowledge.
5. Makes predictions about the future of the problem.
6. Explain the relationships with different variables.
How is a theory evaluated?
All theories contribute knowledge. The most common criteria to evaluate it are:
to. Description capacity.
b. Logical consistency
c. Perspective
d. Heuristic fruiting - generating new questions
and. Parsimony - simplicity
1.1. Subject limitation
All research is limited by various factors of a social, political, economic nature… but we will particularly point to human and material resources.
In human resources, according to what is related to the researcher: Their ability to investigate, consider all parts of the problem, intellectual and human aptitude, acquisition of basic knowledge, use of methods and techniques, time available for research and consulting some specialist in the research topic.
Material resources: Bibliographic sources, access to libraries, archives or to any research system, and tools required by the application of the documentary research technique.
Delimitation of the topic.
- Accuracy.- The subject must be precise, have a very limited outline that makes it particular.Extension limited.-Select a single perspective or part of what was initially chosen.Originality.- That it is new as a subject, interpretation or focus..- Having certain analysis techniques must be taken with the environment and the necessary bibliography, have the necessary time and have the guidance of a good specialist guide on the matter.
Theme Features
Interest.- An important factor for the researcher to stay in the research process and to make the effort required to tackle it, avoiding abandonment due to tiredness or boredom.
Originality.-Ingenuity must be put into practice to create or raise themes with new approaches, avoiding imitations or copies.
Relevance.-That research contributes something to science, to humanity, or to the researcher himself.
Precision.- Precision avoids generality that will lead to superficial and confusing results.
Objectivity.- When posing a problem, it must be done faithfully to the object of study and for it to be objective, a closed attitude should not be assumed.
1.2. Problem Statement
In the definition of the study problem, it is essential to clearly identify the question to be answered or the specific problem whose solution or understanding will contribute to the execution of the research project. Therefore it is recommended to make a clear, precise and complete description of the nature and magnitude of the problem and justify the need for research in terms of social development and / or contribution to global scientific knowledge.
The approach to the problem of a social investigation must leave well established:
1. The description of the problem.
2. Define the subject and the object of the investigation.
3. Delimitations of theory, time, space and scope.
4. Research justification
5. Feasibility
6. Research objectives
7. Establishment of research questions
description of the problem
Planning a problem is to fine-tune its structure formally. The problem should be described in concrete, explicit and specific terms, so that the arguments can be investigated using scientific procedures.
A correctly posed problem is partially solved, the greater the accuracy, the better the chances of obtaining a satisfactory solution. The researcher must be able not only to conceptualize the problem but also to verbalize it in a clear, precise and accessible way.
Sometimes you know what you want to do but you do not know or can communicate it to others and you need to make an effort to translate your thoughts into terms that you understand and accept so that you can later communicate it to others.
The problem statement must be exposed with the following criteria:
to. The problem must be clearly formulated; describe the facts, situations, participants, characteristics of the phenomenon, places, dates, conflicts, dramas, difficult situations, outcomes, etc.
b. Express the problem and its relationship with one or more variables.
c. Express the possibilities of conducting empirical tests.
d. Point out the observable and measurable aspects.
Expected impact
The expected impact is not a reformulation of the results but a description of the incidence of the results from the point of view of the strategic issues or problems defined in society.
They are mainly related to the solution of local, regional, national or global problems, and / or to the development of the country, in academic, socioeconomic, environmental, productivity, etc. terms. Potential direct and indirect users of the Research results
The researcher must clearly identify the institutions, unions and communities, national or international, that will be able to use the results of the research to develop their objectives, policies, plans or programs.
1.3. Research objectives
The objectives of the investigation; they refer to clear and precise statements of the purposes for which the investigation is carried out. When we select a topic we must know why we selected it, when we identify why at that moment, we will be defining the objective of our theme, this objective may refer to our personal interest, the interest of other people, for example: our bosses or the in charge of our public or private institution.
Considerations that must be taken into account when setting objectives:
1. Its formulation must include concrete results in the development of the investigation.
2. The link of the objectives must be within the possibilities of the researcher.
Classification of objectives:
Overall objective. It consists of stating what you want to know, what you want to find and what you want to know. It also consists of what we intend to do in our research; that is, the clear and precise statement of the goals pursued in the research to be carried out.
Specific objectives: the general objectives give rise to the specific objectives that indicate what is intended to be done in each of the stages of the investigation. These objectives must be evaluated at each step to know the different levels of results.
Methodological objectives: There are also methodological objectives which help us achieve the proposed objectives at either of the two levels ("general and specific"), the methodological objective helps us clarify the meaning of the hypotheses and collaborates in the operational achievement of the investigation.
How to formulate objectives: A well formulated objective is one that manages to convey what the researcher is trying to accomplish; that is, what the best statement of an objective aims to obtain as a result includes a greater number of possible interpretations of the objective to be achieved.
1.4. Justification and feasibility of the investigation
Justifying an investigation is; state the reasons why you want to do it, because all research must be done with a defined purpose, you must explain why the research is convenient and what or what benefits are expected with the knowledge obtained.
The researcher must know how to "sell the idea" of the research to be carried out, so he must emphasize his arguments on the benefits to be obtained and the uses to which the knowledge will be put.
To this end, the research advisor establishes a series of criteria to evaluate the usefulness of a proposed study; such criteria are:
to. Convenient, in terms of academic purpose or social utility, the sense of urgency. What will it do and who will it serve.
b. Social relevance. Transcendence, utility and benefits.
c. Practical implications. Does the information really have any use?
d. Theoretical value, is any knowledge gap to be filled?
and. Methodological usefulness, will a new model be used to obtain and collect information?
The feasibility of research
The viability of research is closely related to the availability of material, economic, financial, human, time and information resources. For each of these aspects, a critical and realistic question must be asked with a clear and defined answer, since any doubt about it may hinder the purposes of the research.
1.5. Hypothesis formulation
Hypotheses are announced propositions to tentatively answer a problem, it can also be tested to determine its validity and can be developed from different points of view, it can be based on a conjecture, on the result of other studies, on the possibility of a similar relationship between two or more variables represented in a study, or it may be based on a theory by which an assumption of deductive process leads us to the claim that if certain conditions exist, certain results can be obtained, that is, the cause-effect relationship.
The importance of the hypothesis derives from the link between theory and empirical reality between the formalized system and research. They are working instruments of theory and research insofar as they introduce coordination in the analysis and guide the choice of data.
The hypotheses cover the following functions:
• Initial explanation. The elements of a problem may seem obscure, therefore, through the formulation of hypotheses the data could be completed, detecting the possible meanings and relationships of them, and introducing an order between the phenomena.
• To stimulate research. The hypotheses specify and summarize the problems found, serving as an impetus for achieving the inquisitor process.
• From methodology source. This formulation leads us to an analysis of the variables to consider and, consequently, to the methods necessary to control them.
• Of criteria for evaluating research techniques. Often the hypotheses establish in their statement some set of conditions that make possible a critical judgment on the technical procedures followed to satisfy the specified conditions.
• From organizing principles. The hypotheses are mainly organizers around which mayas of relations can be formed between the known facts, pertinent to the problem, and others not so directly connected.
Types of hypotheses
The following classification presents a first group of two different formulations, a second group classified by an object and extension, and a third group of loose or different denomination hypotheses.
First group:
• General or empirical. She is the leader of the research, she tries to approach the problem as a basis for the search for data, she cannot cover more than what is proposed in the objectives of the research or disagree with them.
• Work or operational. Once the general hypothesis has been formulated, the working hypothesis is formulated. It is called work because it is the indispensable resource for the precise and definitive achievement of the objectives proposed in the research.
• Null hypothesis. In formulating this hypothesis, the intention is to deny the independent variable, that is, it is stated that the cause determined as the origin of the problem fluctuates, therefore, it must be rejected as such.
• Research hypothesis. It identifies with the general one and responds broadly and generically to the doubts presented in the formulation of the problem.
• Operational assumptions. He presents us with the general hypothesis of the investigation around the phenomenon to be studied and the instruments with which the variables are to be measured.
• Statistical hypothesis. It is the one that expresses the operational hypothesis in the form of a mathematical equation, therefore it must be precise in order to facilitate a statistical relationship. The most accurate of the statistical hypotheses is called the null hypothesis, which denies the relationship between the variables in a study.
Second group:
• Descriptive hypotheses. They refer to the existence of exchange relations in the structure of a given phenomenon that is studied.
• Causal hypotheses. It is a tentative proposition of the factors that intervene as a cause in the phenomenon under study.
• Unique hypotheses. In this hypothesis, the proposition presented is located in spatio-temporal terms.
• Statistical hypotheses. It indicates to us that a greater proportion of people or elements with certain or certain characteristics have such or such other characteristics.
• Restricted general assumptions. In this hypothesis, the proposition refers to the totality of its members, being restricted either to a specific place or period of time.
• Unrestricted universal assumptions. They are verified by a certain science, and which form the basis of its laws and theories.
Third group:
• Alternative hypotheses. When a problem is answered, it is convenient to propose other hypotheses in which independent variables appear other than the first one we formulated. Therefore, it is necessary to find different alternative hypotheses in response to the same problem and to choose among them which ones and in what order we are going to treat them.
• Particular hypotheses. Only those that are deduced and articulated from a basic hypothesis.
• Empirical hypotheses. They are isolated assumptions without theoretical foundation but empirically proven.
• Plausible hypotheses. They are theoretically based assumptions, but without empirical contrast.
• Ante-facto hypothesis. This type of hypothesis introduces an explanation before observation. Orient and proceed to discovery.
• Post-facto hypothesis. It is deduced from the observation of a phenomenon or a fact.
• Validated hypotheses. They are well-founded and empirically tested hypotheses.
1.6. Determination of the variables
Variables are characteristics, attributes, qualities or properties that occur in individuals, groups or objects and their validity systematically depends on the theoretical framework that underpins the problem and from which it has been derived, and on its direct relationship with the hypothesis that supports it.
In the process of elaboration of a variable it is recommended to take into account the following:
• The nominal definition of the variable to be measured.
• The actual definition or dimension of the variable.
• The operational definition or indicators of the variable.
Finally, the index is indicated, which is nothing more than the result of the combination of values obtained by an individual or element in each of the indicators proposed to measure the variable. Variables are classified according to their capacity or level at which they allow us to measure objects, that is, the most common and basic characteristic of a variable is to differentiate between the presence and absence of the property that it states.
The classification of the variables is:
• Continuous variable. It occurs when the phenomenon to be measured can take quantitatively different values. For example the chronological age.
• Discrete variables. They are those that establish categories in non-quantitative terms between different individuals or elements.
• Individual variables. They present the characteristic or property that characterizes certain individuals, and can be: Absolute, Relational, Comparative and Contextual.
• Collective variables. They present the characteristics or properties that distinguish a specific group or group, and can be: Analytical, Structural and Global.
• Background variable. It is the one that is assumed as the antecedent of another, that is, that there are variables that are antecedents of others.
• Independent variable. It is the variable that precedes a dependent variable, which is presented as the cause and condition of the dependent variable, that is, they are the conditions manipulated by the researcher in order to produce certain effects.
• Dependent variable. It is the effect produced by the variable that is considered independent, which is managed by the researcher.
• Intervening or alternate variable. It is the variable that is between the independent and dependent variables, so that it can replace the independent variable, which has been formulated, or it can also act as a relative factor in the variable relation.
• Strange variables. When there is an independent variable not related to the purpose of the study, but which may have effects on the dependent variable, we have a strange variable.
• Dichotomous variables. They only allow division into two categories. Example: day and night.
• Inter Variable. They are those that make comparisons between groups.
• Intra variables. They are those that can study the same group in different periods.
2.1. Experimental research
Experimental research involves manipulating an untested experimental variable, under tightly controlled conditions, in order to describe how or for what cause a particular situation or event occurs.
It is an experiment because precisely the researcher provokes a situation to introduce certain study variables manipulated by him, to control the increase or decrease of that variable, and its effect on the observed behaviors. The researcher deliberately handles the experimental variable and then observes what happens in controlled situations.
The experimental research follows the following stages:
1. Delimit and define the object of the investigation or problem. It consists of clearly determining the objectives of the experiment and the questions to be answered. Then the independent variables, the dependent variables, the constant parameters and the necessary precision in the measurement of the variables are indicated. The existing bibliography, the region in which the results are of interest, the available equipment and its precision, and the time and money available are taken into account.
2. Present a working hypothesis. To do this, you must be certain of the type of work to be carried out: if it is a question of verifying a hypothesis, a law or a model, it is not necessary to propose a working hypothesis; if the work is a complement or extension of another, it is possible that the hypothesis of the original work could be used or some small modification could be made; If the problem to be investigated is new, then it is necessary to propose a working hypothesis.
All investigation begins with an assumption, a presentiment or idea of how the phenomenon can occur. These ideas must be clear enough to anticipate a tentative result of how this phenomenon can occur: this tentative result is the hypothesis.
3. Prepare the experimental design. Once the nature of the problem is known (if it is an investigation, extension or confirmation), the desired precision, the appropriate equipment and the proposed working hypothesis, it should be analyzed whether the answer to our problem is going to be the interpretation of a graph, a value or an empirical relationship; This will point us to the experimental procedure, that is, how to measure, in what order, and what precautions to take when doing so.
Once these stages have been determined, the experiment is designed using the following steps: Determine each and every one of the components of the equipment, couple the components, carry out a test experiment and tentatively interpret the results and check the precision, modifying, if it is necessary, the procedure and / or equipment used.
4. Carry out the experiment. Once the test experiment and the tentative interpretation of results have been carried out, carrying out the final experiment is almost reduced to filling columns, prepared in advance, with readings of the measurements, to detecting any anomaly that occurs during the development of the experiment and to plotting the pertinent graphs or calculate the value or values that will answer the problem.
5. Analyze the results. The analysis or interpretation of results, be they values, graphs, tabulations, etc., must answer as clearly as possible the question or questions posed by the problem.
In general terms, the analysis includes the following aspects: 1) If the experiment seeks to confirm a hypothesis, law or model, the results must show whether or not there is agreement between theory (the hypothesis, law or model) and the results of the experiment..
It may happen that the agreement is partial; if so, it must also be presented in which parts there is, and in which not; 2) If it is an experiment that discriminates between two models, the results must allow the discrimination to be made in a clear way and provide the reasons to accept one and reject the other; 3) If what you are looking for is an empirical relationship, it must be at least in graphic form; the ideal is to find an analytical expression for the graph, that is, find the equation.
This equation is called empirical because it was obtained through an experiment and as an analytical expression of a graph. It should be taken into account that in a graph each experimental point has a margin of error and that in case of doubt when the curve is not well determined, a better adjustment should be made using least squares. It should be noted that the simplest curve to analyze is the line and that if we did not obtain it by graphing our points, we should try to obtain it, either by changing variables or graphing on semi-log paper.
6. Draw conclusions. Once the results of the experiment have been achieved, the researcher must apply his scientific criteria to accept or reject a hypothesis or a law; You may also make some conjecture about one model, or propose creating a new one, which would lead to a new problem. The following criteria generally apply:
1) Rejects a hypothesis, law or model, when it experimentally verifies that it is not fulfilled. It is enough that there is a single phenomenon that cannot be explained to discard it.
2) Accept as true but not absolutely true a hypothesis, law, theory or model, as long as there is no proof of failure in the explanation of any phenomenon.
3) It may happen that the hypothesis or model agrees only partially with the experiment, then it is necessary to speculate about the possible reasons for the difference between the theory and the experiment, and try to make new hypotheses or modifications to the existing one, which leads to a new problem.
In the conclusions the questions posed in the experiment are answered clearly, to check whether or not our working hypothesis or the proposed model is valid. If there are unanswered questions, establish why or if warrants, conjecture about the hypothesis or model that describes the phenomenon studied.
7. Prepare a written report. Its parts will be:
1) the definition of the problem;
2) the experimental procedure;
3) results;
4) conclusions.
The writing of the writing under the conventions of a research report.
Controlled experiment refers to selecting two random samples: one subject to a special variable and the other not subject to the same variable. The final characteristics of both are compared and then the effect of the experiment is determined.
If there is a significant difference between them, the hypothesis is analyzed and the experiment is performed again. The difficulty lies in achieving uniformity of characteristics in the experimental sample, and the control one requires precision in the calculation of the characteristics.
2.2. Methodology
It must be shown, in an organized, clear and precise way, how each of the specific objectives proposed will be achieved. The methodology should reflect the logical structure and scientific rigor of the research process from the choice of a specific methodological approach (questions with corresponding supported hypotheses, sample or experimental designs) to the way in which the analysis, interpretation and presentation of the results.
The procedures, techniques, activities and other methodological strategies required for the investigation must be detailed. The process to be followed in the collection of information should be indicated, as well as in the organization, systematization and analysis of the data.
Keep in mind that the methodological design is the basis for planning all the activities required by the project and for determining the required human and financial resources.
A vague or imprecise methodology does not provide elements to assess the relevance of the requested resources. For National Programs that require it, the researcher must describe the ethical considerations. Additionally, the Institution's letter of approval of the thesis project must be attached. In the case of research on people or human groups, it is essential to start from the principles of institutional ethics.
Stages that the researcher must carry out to carry out an experimental investigation.
* Presence of a problem. For which a bibliographic review has been carried out.
* Identification and definition of the problem.
* Definition of hypotheses and variables and their operation.
* Design of the experimental plan.
*Research design.
* Determination of the population and sample.
* Selection of measuring instruments.
* Development of instruments.
* Procedures for obtaining data.
* Data reliability test.
* Carrying out the experiment.
*Data treatment. Here at this point we must bear in mind that one thing is the raw data, another the processed data and another, the data that must be given as definitive.
2.3. Statistics in research
The statistical application process involves a series of steps:
Selection and determination of the population or sample and the contained characteristics to be studied. In the event that you want to take a sample, it is necessary to determine its size and the type of sampling to be carried out (probabilistic or non-probabilistic).
Obtaining the data. This can be done by directly observing the elements, applying surveys and interviews, and conducting experiments.
Classification, tabulation and organization of the data. The classification includes the treatment of the data considered anomalous that can, at a given moment, falsify an analysis of the statistical indicators. Tabulation involves summarizing the data in statistical tables and graphs.
Descriptive analysis of the data. The analysis is complemented by obtaining statistical indicators such as the measures: central tendency, dispersion, position and shape.
Inferential analysis of the data. Data treatment techniques are applied that involve probabilistic elements that allow conclusions to be inferred from a sample to the population (optional).
Frequency Distribution. It is a grouping of data into mutually exclusive categories giving the number of observations in each category. The relative frequency is obtained by dividing the class frequency by the total data (n). The percentage frequency is obtained by multiplying the relative frequency by 100.
2.4. Measures of central tendency
This type of measurement allows us to identify and locate the point (value) around which the data tend to be collected (“Central point”). These measures applied to the characteristics of the units of a sample are called estimators or statisticians; while applied to populations they are called parameters or statistical values of the population. The main methods used to locate the center point are the mean, median, and mode.
1. Average
It is the most widely used, best known, and easiest measure of central position to calculate, mainly because its equations lend themselves to algebraic handling, which makes it very useful.
Its main disadvantage lies in its sensitivity to changing one of its values or to extreme values that are too large or too small. The mean is defined as the sum of all observed values, divided by the total number of observations.
2. Median
With this measure we can identify the value that is in the center of the data, that is, it allows us to know the value that is exactly in the middle of the data set after the observations have been placed in an ordered series. This measurement indicates that half of the data is below this value and the other half is above it. To determine the position of the median the formula is used:
Some characteristics of the median are:
-All ordinal, interval, or ratio data sets have a median.
-A data set only has a median.
-The median is not affected by extremely large or extremely small values, so it is especially useful when you have these values.
3. Fashion
The modal measure indicates the value that is most often repeated within the data; that is, if we have the ordered series (2, 2, 5 and 7), the value that is repeated the most times is the number 2 who would be the mode of the data.
It is possible that in some occasions two values are presented with the highest frequency, which is called Bimodal or in other cases more than two values, which is known as multimodal.
Some characteristics of fashion are:
-The mode can be determined in data groups at all levels (nominal, ordinal, interval and ratio).
-There can be more than one mode for each group of data.
-Fashion is not affected by extremely large or extremely small values, so it is especially useful when you have these values.
2. 5. Tabulation and graphical presentation
Tabulation is a common way of presenting associations between two or more variables. A table has the advantage that a large amount of data can be well arranged in it and exact figures are preserved. One downside is that a large table is not illustrative: it rarely reveals more than the most obvious regularities or interdependencies between data. Some conventional abbreviations used in tables are presented under the heading Classify.
Graphic presentation
A table is a table that consists of the joint, ordered and normally totalized arrangement of the total sums or frequencies obtained in the tabulation of the data, referring to the categories or dimensions of a variable or of several variables related to each other.
The tables systematize the quantitative results and offer a numerical, synthetic and global vision of the observed phenomenon and the relationships between its various characteristics or variables. In it, the qualifying phase of quantitative research ends and is finally finalized.
Having the definition of what a table is, we can then work each of the types of tables requested:
Data entry table: This is a table in which only the data obtained from the scientific research or experiment appear. It is the simplest table and is used when no further information about the data is needed, these tables are constructed by tabulating the data, this procedure is relatively simple, to perform it we deal with a set of statistical data obtained by recording the results of a series of n repetitions of some experiment or random observation, assuming that the repetitions are mutually independent and performed under uniform conditions, it is important to say that the result of each observation can be expressed numerically, for this type of table data entry can work with one or more variables,so that our statistical material consists of n observed values of the variable X.
Frequency tables: A frequency table is made up of the categories or values of a variable and their corresponding frequencies. This table is the same as a frequency distribution. This table is created by means of tabulation and grouping, which is a simple method as we had begun to see in the data table, the same tabulation procedure previously described is performed if the number of observed values for the variable is works with a single variable, discounting the repeats are small, if there are repeats the frequency f is the number of repetitions of a given value of X.
However, when the data set is larger, it is laborious to work directly with the observed individual values, and then some sort of grouping is generally carried out as a preliminary step, before any other data processing is started.
The rules for grouping are different depending on the variable, discrete or continuous, for a discrete variable it is usually convenient to make a table whose first column contains all the values of the variable X represented in the material, and in the second, the frequency f with which each value of X has appeared in the observations.
Double-entry tables: Also called contingency tables, are those data tables referring to two variables, formed, at the headings of the rows, by the categories or values of one variable and at those of the columns by those of the other, and in the squares of the table, by the frequencies or number of elements that simultaneously meet the two categories or values of the two variables that cross each other.
For the tabulation of a grouped material of simultaneous observations of two random variables we will need a table described as previously described, the rules for grouping are the same as in the case of a single variable.
This type of tables provide statistical information on two interrelated events, it is useful in cases where the experiments are dependent on another experiment, more applications of bivariate statistical analysis appear below.
Graphic methods:
A diagram is a kind of schematic, consisting of lines, figures, maps, used to represent either statistical data at scale or according to a certain proportion, or the elements of a system, the stages of a process and the divisions or subdivisions of a classification. Among the functions that the diagrams fulfill, the following can be pointed out:
- They make data, systems and processes more visible They reveal their variations and their historical or spatial evolution. They can show the relationships between the various elements of a system or a process and represent the correlation between two or more variables. They systematize and synthesize the data, systems and processes. They clarify and complement the tables and the theoretical or quantitative expositions. Studying their disposition and the relationships they show may suggest new hypotheses.
Some of the most important diagrams are tree diagram, area or surface diagram, band diagram, bar diagram, block diagram, circular diagram, circular polar diagram, point diagram, stem and leaf diagram, histograms and box and mustache graphics or boxplots.
Univariate graphs: To work univariate graphs we must first know what univariate statistical analysis is, it provides the analyst with representative measures of the distribution or averages, dispersion indexes of the distribution data, procedures to normalize the data, measures of inequality of some data in relation to others and finally measures of the asymmetry of the distribution.
Point graphics: It is a variation of the simple linear diagram which is formed by straight lines or curves, which result from the representation, on an axis of coordinates, of frequency distributions, this construct placing on the x axis the values corresponding to the variable and on the ordinate axis the value corresponding to the frequency for this value. It mainly provides information regarding frequencies. This is used when only frequency information is needed.
Stem and Leaf Plots: It is a quick way to obtain an illustrative visual representation of the data set, to construct a stem and leaf diagram you must first select one or more initial digits for the stem values, the final digit or digits are they turn into leaves, then a list of stem values is made in a vertical column.
Continuing to record the leaf for each observation along with the corresponding stem value, finally the units of stems and leaves are indicated somewhere in the diagram, this is used for large lists and is a summary method of displaying the data, it has the disadvantage that it only provides the data, and there is nowhere information on frequencies and other important data.
Bar diagrams: name given to the diagram used to graphically represent discrete non-grouped frequency distributions. It is so named because the frequencies of each category of the distribution are represented by lines or columns of proportional length, separated from each other.
There are three main classes of bar charts:
- Simple bar: they are used to graph unique facts Composite bars: in this graphing method the bars of the second series are placed on top of the bars of the first series respectively Multiple bars: it is highly recommended to buy a statistical series with Another, for this, uses simple bars of different colors or patterns on the same Cartesian plane, one next to the other.
The bar chart mainly provides comparative information and this is its main use, this chart also shows the information regarding the frequencies
Histograms: Used to illustrate samples grouped into intervals. It consists of rectangles joined to others, whose vertexes at the base coincide with the limits of the intervals and the center of each interval is the class mark, which we represent on the axis of the abscissa. The height of each rectangle is proportional to the frequency of the respective interval. This proportionality is applied by means of the following formula; Rectangle height = relative frequency / base length.
The histogram is used to represent continuous quantitative variables that have been grouped into class intervals, the disadvantage of which is that it does not work for discrete variables, otherwise it is a useful and practical way of displaying statistical data.
Pie charts: it is a chart that is based on a proportionality between the frequency and the central angle of a circle, in such a way that the central frequency corresponds to the 360 ° central angle. The following formula is applied to construct: X = relative frequency * 360 ° / S relative frequency.
This is used when working with data that have high frequencies, and the values of the variable are few, the advantage of this diagram is that it is easy to do and is easily understandable, the disadvantage is that when the values of the variable are many it is almost impossible or rather it does not inform much this diagram and it is not productive, it mainly provides information about the frequencies of the data in an understandable and simple way.
Bivariate graphs: To work the scatter diagrams, we must first know what bivariate statistical analysis is and its advantages. Bivariate statistical analysis is that analysis that operates with data referring to two variables and aims to discover and study their statistical properties.
It is mainly oriented to the normalization of the values or frequencies of the raw data, determines the existence, direction and degree of the joint variation between the two variables, which is done by calculating the relevant correlation coefficients, calculates the covariance or product of the deviations of the two variables in relation to their respective means and finally establishes the nature and form of the association between the two variables in the case of interval variables.
Scatter diagram: it is a diagram that graphically represents, in a space of ordinates, the points of said space that correspond to the correlative values of a joint bivariate distribution, these diagrams should be used when we have a bivariate statistical analysis, that is, a data table double entry, the advantage they have is that you can graph a joint bivariate distribution in a simple way and the main disadvantage is that it does not work if it happens that a pair repeats itself.
Conclusions
The research collects knowledge or data from primary sources and systematizes them to achieve new knowledge. It is not research to confirm or collect what is already known or has been written or researched by others. The fundamental characteristic of research is the discovery of general principles.
The researcher starts with previous results, approaches, propositions or answers about the problem at hand. To do this you must: carefully plan a methodology, collect, record and analyze the data obtained, but if these instruments do not exist, you must create them.
All research must be objective, that is, it eliminates the researcher personal preferences and feelings, and refuses to seek only those data that confirm his hypothesis; Hence, it uses all possible tests for the critical control of the data collected and the procedures used.
Finally, once the data is systematized, it is recorded and expressed through a report or research document, which indicates the methodology used and the procedures used to reach the conclusions presented, which are supported by the same research carried out.
In research, there must be a series of characteristics for it to be actually scientific: being planned, that is, having a previous organization, having data collection instruments that meet the criteria of validity, reliability and discrimination, as minimum requirements to achieve a scientifically valid report.
But above all to be original, that is, to point to a knowledge that is not possessed or that is in doubt and it is necessary to verify and not to a repetition reorganization of knowledge that they already possess. The researcher should try to eliminate personal preferences and feelings that could play or mask the result of the research work.
To carry out an investigation, the necessary time must be available in order not to rush information that does not respond, objectively, to the analysis of the available data. Aiming at numerical measures, in the report trying to transform the results into more easily representable and understandable quantitative data and more objective in the final assessment.
Offering verifiable results and verifying them in the same circumstances in which the investigation was carried out. Considering investigated particular situations, for which a sampling technique with the necessary scientific rigor is required, both in the selection method and in the sample quantity, in relation to the population in question.
It is very convenient to have a detailed knowledge of the possible types of research that can be followed. This knowledge makes it possible to avoid mistakes in choosing the right method for a specific procedure. Since although research is always present, it is always good to know the technical and scientific side of things, no matter how common and everyday they tend to be.
One of the most common failures in research consists in the lack of delimitation of the topic, that is, due to the lack of ambition of the topic, which is why it is essential to be very clear about the objectives and the path to be followed with the research so that it can end its path where it should.
When posing a problem, remember; formally structure the idea of the research, developing the three fundamental elements: objectives, questions and justification of the research. The objectives and questions must be consistent and feasible to answer and go in the same direction.
While in the justification he explains the reasons why it is necessary to make the investigative effort. On the other hand, the feasibility criteria are based on the availability of resources, social convenience, relevance, practical implications, theoretical value and methodological utility.
A problem statement should not include moral or aesthetic judgments. It must include aspects of professional ethics respecting confidentiality, intellectual work and practices that respect human dignity.
Bibliographic reference
Hernández Sampieri Roberto. (1991). Investigation methodology. Mexico: Mc Graw Hill.
Hernández, R.; C. Fernández and P. Baptisa. (nineteen ninety five). Investigation methodology. Mexico: Mc.Graw-Hill.
Münch, Lourdes., Ernesto, Ángeles. (2001). Methods and techniques of investigation. Mexico: Trillas.
Orna, Elisabeth., Gram, Stevens. (2000). How to use the information in research works. Barcelona: Gedisa.
Ortiz Uribe, Frida Gisela., Pilar García, María. (2003). Research methodology: the process and its techniques. Mexico: Limusa.
Reza Becerril Fernando, (1997). Science, methodology and research. Mexico: Alambra Mexicana, pp. 207-259.
Tamayo and Tamayo, Mario. (1998). The process of scientific research. 3rd ed. Mexico: Ed. Limusa SA
Tomeo Perucha V., Uña Juárez I. (2003). Descriptive Statistics Lessons. Publisher: Thomson.
Download the original file
