FORMULATING THE RESEARCH DESIGN

History search for knowledge and truth

Broadly, it involves any appeal to past experience to help in knowing what to do in the present and future. It is concerned with describing past events or facts in a spirit of inquiring critically for the whole truth.

Historical observations help broaden our experiences and make us more understanding and appreciative of our human nature and uniqueness.

History can give us ideas about what ideas and strategies were poorly and successfully tried previously.

Because of this, history can assist us to determine possibly better ways of deciding and doing things.

Historical Research

DEFINITION AND AREAS OF HISTORY

DESCRIPTIVE RESEARCH

Views on the Value of Historical Research

a research design wherein past events are studied and related to the present or in the future time. (Castillo, 2002)

Its purpose is to reconstruct the past objectively and accurately.

METHODS OF HISTORICAL RESEARCH

Formulating your problem

Gathering your source materials

Criticizing your source materials

GATHERING YOUR SOURCE MATERIALS

CRITICIZING YOUR SOURCE MATERIAL

FORMULATING YOUR RESEARCH PROBLEM

1. External Criticism

2. Internal Criticism

EXTERNAL CRITICISM

involves finding out it the source material is genuine and if it possesses textual integrity.

Deals with the data of form and appearance rather than the meaning of the data.

There are several motivations for undertaking a historical research. One of these is your DOUBT about some event, development or experience in the past.

Another reason for your historical study may be your discovery of new source materials the meaning of which will supply answers about past events when you make your interpretations.

Classification of Historical Sources:

1. Primary Source - "best evidence"

2. Secondary Source

3. Deliberate

4. Inadvertent

INTERNAL CRITICISM

involves determining the meaning and trustworthiness of the statements found in the documents.

Finding out the real meaning of statements, the observer's competence, as well as his truthfulness and honesty.

Another source of your problem may be a question regarding an old interpretation of an existing data.

TYPES OF DESCRIPTIVE RESEARCH

Case Study

Surveys

Two Scopes of Surveys

In such studies, you do not take into account the relationships between variables. Your main intention is to use data for problem solving rather than for hypothesis testing.

CENSUS is a survey that covers the entire

population of interest

SAMPLE is a survey that deals only with a

portion of the population.

FOUR CATEGORIES OF SURVEY

(Accdg. to scope and subject matter)

c. Sample Surveys of Tangibles- you obtain your information from large groups where the variables are concrete.

FACTORS THAT DETERMINE ITS USEFULNESS:

a. Appropriateness of sampling procedure

b. Accuracy of your data-collecting method

c. Relevance of the information you gather

a. Census of Tangibles- covers a small population where the variables are CONCRETE e.g. no. of students in school, desks in a classroom, dental chairs in a dental infirmary.

b. Census of Intangibles- The variables that you measure are not directly observable such as pupils' achievement, attitude of parents

This means that you are going to gather data which include the past experiences and present condition of the subject as well as his environment.

You will attempt to find out how these factors relate to one another and how they have affected the subject.

E.g. researches in the field of guidance- personality problems, etc.

If you intend to gather relatively limited data from a relatively large number of cases.

It involves determining information about variables than about individuals.

Employed to measure the existing phenomenon without inquiring into why it exists.

d. Sample Survey of Intangibles- deals with the measurement of psychological or sociological constructs, as well as comparing a large population where the variables are not directly observable.

include observation, surveys and interviews, standardized tests and case studies.

It is a research design wherein events are recorded, described, interpreted, analyzed and compared. (Castillo, 2002.)

Its objective is to describe systematically a situation, condition or area of interest factually and accurately.

Developmental Study

Follow- Up Studies

• Case Study
• Surveys
• Developmental Studies
• Follow- up Studies
• Documentary Analyses
• Trend Analyses
• Correlational Studies

Employed when you intend to investigate the subsequent development of participants after a specified treatment or condition.

Generally, you may use these studies to evaluate the success of particular programs like the guidance program, instructional program, administrative program, etc.

Developmental studies may investigate progression along a number of dimensions such as intellectual, physical, emotional or social development.

2 COMPLEMENTARY TECHNIQUE:

1. Longitudinal- extended period of time

2. Cross- Sectional- studying participants at the same point in time.

Nature of Descriptive Research

Ways of Obtaining Descriptive Information

Trend Analysis

Documentary Analysis

To describe the nature of a situation as it exists at the time of the study and to explore the causes of a particular phenomena.

It is employed in studies which aim to project the demands or needs of the people in the future.

1. By Asking- personal interview, survey, questionnaire, opinionnaire

2. By Observation- activity analysis, product analysis, situation analysis

3. By Use of Devices or descriptive survey insruments

Involves gathering information by examining records and documents.

Often referred to as content analysis.

For example, you can analyze business records, or determine the extent that books carry lessons on character education.

Correlational Studies

These studies are designed to help you determine the extent to which different variables are relate to each other in the population of interest.

The critical distinguishing characteristic is the effort to estimate a relationship, as distinguished from simple description.

Correlational studies do not require you to have a large sample. It is assumed that if a relationship exists, it will be evident in a sample of modest size provided that the sample you use is representative of the population you are investigating and provided that the instrument you use is reliable and valid.

RESEARCH DESIGN

THREE BASIC RESEARCH DESIGNS

DEFINITION

Experimental Research

Under this section, the method of research that will be or was used (descriptive, experimental, etc.) should be explained and defined briefly.

The procedural part of the method, its appropriateness tot he study, and some of its advantages should be properly and briefly discussed.

A Research design encompasses the methodology and procedures employed to conduct scientific research.

- Wikipedia

A detailed outline of how an investigation will take place.

It constitutes the blueprint for the collection, measurement and analysis of data.

1. Descriptive Research

2. Historical Research

3. Experimental Research

PURPOSE

To ensure that the evidence obtained enables us to answer the initial question as unambiguously as possible

Experimental Designs

3 Characteristics of Experimental Design

There is no perfect design: enhancing one type of validity tends to jeopardize the other so that you must attempt an adequate compromise.

There are two major classes of experimental design: single-variable and factorial designs.

1. An independent variable of is manipulated

2. All other variables except the dependent variables are held constant

3. The effect of the manipulation of the independent variable on the dependent variable is observed or measured.

a research design wherein the cause and effect relationship of a treatment on a variable is determned (Castillo, 2002).

This can be further divided into true experimental design and quasi-experimental design.

CHARACTERISTICS OF A SOUND DESIGN

Three General Types

(Campbell and Stanley)

Quasi- Experimental Designs

True Experimental Design

Pre- Experimental Design

The One-Spot Case Study

This is employed when it is not possible to randomly assign subjects to groups, the quasi-experimental designs are employed.

It consists of just one group which is exposed to a treatment and then given a posttest.

It is devoid of controls whatsoever.

where: x = treatment

O = Posttest

All of the true experimental designs have one characteristic in common which is RANDOM ASSIGNMENT OF SUBJECTS to groups.

All of the true experimental desins have a control group.

The One-Group Pretest-Posttest Design

There is no randomization done.

One- Spot Study Case

One-Group Prettest-Posttest Design

Static-Group Comparison

Time Series Experiment

It involves one group with pre and posttests.

Threats to internal validity may be history, maturation, testing, instrumentation and statistical regression.

where: O1= pretest

x = treatment

O2 = posttest

Counterbalanced Designs

The Nonequivalent Control Group Design

These designs refer to those that allow the subjects to try out all experimental treatments but in a different sequence.

They are sometimes called rotation experiments, cross- over designs, or Latin-square designs.

• It is an elaboration of the one-group, pretest-posttest.
• Involving just one grop, it is pretested four times and then the same group is posttested four times.
• The number of times of pretesting and posttesting may be more.
• If a group scores essentially the same on a number of pretests and then significantly improves following a treatment,you have more confidence in the effectiveness of the treatment than if just one pretest is administered.

• This design is considered as one of the most widely used designs in educational research.
• Consisting of two groups, the design gives pretest and posttest to each group but only one gets the treatment.
• There are no randomization done.
• This design is commonly used with groups with participants naturally assembled such as those in classrooms.

The Static-Group Comparison

The Pretest- Posttest Control Group

The Solomon Four- Group Design

There are four comparisons to be made:

O1 and O2

O2 and O4

O5 and O6

O3 and O5

This is a design in which a group which has experienced X is compared with one which has not for the purpose of establishing the effects of X.

The design involves TWO GROUPS, one experimental, the other control, and both are posttested.

X O1 Where: x= Treatment

O2 O1= Exp.Posttest

O2= Control Posttest

This design involves two groups, both of which are formed by random assignment.

Both groups are pretested and posttested but only one group is given treatment.

This design has the highest prestige since it has the GREATEST VALIDITY.

It is the extension of the pretest-post test control group design

Two groups are pre- and posttested; two groups are posttested; two groups have treatments and two groups do not have experimental treatments.

The two independent variables are the treatment variable and the pretest variable. While the design enjoys preference there is one difficulty with it, there might not be enough subjects to form four groups.

DATA ANALYSIS

2. If the average differences between posttest scores of the experimental group and those of the first control group between the second control group and those of the third control group are the same, then the experiment must have had a comparable effect on the pretested and on unpretested groups.

The data generated by this design are computed using the two way ANOVA on posttest scores. The results of the computation are analyzed as follows:

1. If the posttested mean of the experimental group is significantly greater than the mean of the first control group. And if the control posttest mean is significantly higher than that of the second control group, there is evidence that the experimental treatment is effective.

The Posttest- Only Control Group Design

• There are two groups, one of them receiving the experimental treatment, both are posttested but are not pretested. It is commonly confused.
• This design is exactly the same as the prestest-posttest control group design except that there is no pretest.
• The subjects are randomly assigned to groups, exposed to the independent variable, and posttested.
• Posttest scores are then compared to determine the effectiveness of the treatment.

REFERENCES:

Research Methods by Consuelo G. Sevilla, Ed.D

Methods of Research by Merle E. Salmorin, MA

Clearly and explicitly specify hypotheses central to the problem selected

Review previously published literature dealing with the problem area.

Identify the problem clearly and justify its selection

METHODS Vs. DESIGN

Describe the methods of analysis which will be applied to the data in determining whether or not the hypotheses is false.

Clearly describe the data which will be necessary for an adequate test of the hypotheses and explain how such data will be obtained.

RESEARCH DESIGN

RESEARCH METHODS

A Research design involves determining how your chosen method will be applied to answer your research question.

A design of your study can be thought of as a blueprint detailing what will be done and how will this be accomplished.

Generalized and established ways of approaching research questions. (e.g. qualitative and quantitative)

Not all methods can be applied to all research questions, so the choice of method is limited by the area of research that you wish to explore.

Choice of research methods and design should be thought of as a reciprocal process extending well into your study.

FORMULATING THE RESEARCH DESIGN