RESEARCH TERMINOLOGIES & PRINCIPLES.

The purpose of this session is to introduce a new learner to definitions of common terminologies used in a research field.

Definition of research:

Research is the systematic investigation and study of phenomena in order to establish facts and reach new conclusions; it involves correcting and analysis of data. The facts obtained from research can be used to formulate principles, theories, laws, and prediction of the future. Research can also be defined as a systematic, empirical, and goal oriented study or analysis of phenomena (Leedy & Onnond, 2011).

Note that from the above definition we can answer the question, why research is considered to be a science?

1. i) It is a systematic process. ii) It is empirical in nature i.e. it relies on verifiable evidence, not hearsay.

iii) Its goal oriented. iv) It critically analyses issues to develop answers to complex questions.

1. v) It can be replicable and verifiable. vi) It follows the rules of logic and is consistent with known facts

• It uses Empirical tests under controlled environments like
• It can test hypothesis or assumptions. ix) It emphasizes validity and accuracy of the findings:

1. x) It leads to formulation of principles, theories, laws, and prediction of future.

Importance/purpose of research

1. Research avails the relevant information that acts as a basis for, policy formation, formulation of laws, theories and principles.
2. New products are introduced on market through research; new ways of doing things.
3. Research is an agent of change. It is a basis of great discoveries and inventions.
4. Research helps us to predict the future.
5. Research information is needed to make informed decision.
6. Research helps us to answer complex problems/questions on nature e.g. behaviour of heavenly bodies like sun.
7. Killer diseases have been contained through research.
8. Research gives us broad view of event.

Definition of a research problem.

A research problem is what the researcher is interested in finding out or studying.

It can be defined as a statement about an area of concern, a condition to be improved, or a difficult to be eliminated.

What are the Sources of research problem?

• Personal interest and experience.
• Use of intellectual curiosity. One can ask oneself: “How?” “Why?” and so on.
• Using prior research done by other researchers in case there recommendations given for further studies.
• During evaluation of a specific program one can identify a gap which needs research.
• Through direct observation of a current need in your community. Applied research often arises from specific needs.

Examples of research problem:

A researcher may ask questions like the following:

• What is the cause of cholera outbreak among people in Katanga?

In this case the researcher is interested in studying or finding out the cause of cholera outbreak among people in Katanga.

• Which age group is most affected by malaria in Mulago? Here the researcher is interested in studying or finding out the age group most affected by malaria in Mulago.

Note that, it’s important to mention the problem, the population affected and the study area (location).

Qualities/ characteristics of a good research problem:

• Acronym FINER
• F: feasible (one that is researchable e.g. in terms of cost, time, respondents, e.t.c)
• I: interesting (interesting enough to overcome the many hurdles & frustrations of the research process) Note; It is wise to confirm that you are not the only one who finds it interesting.
• Novel (Good research contributes new information) should be one which is not too common or has been already researched on.
• Ethical (For example the study should not pose physical risks to respondents or invasion of privacy).
• Relevant or significant to society (consider how your results might advance scientific knowledge and health policy).

It’s important to note that the above qualities also apply to a research question and research topic as we shall see later.

Steps in the formulation of a research problem:

• Identify a broad field or subject area of interest to you.
• Dissect the broad area into subareas.
• Select what is of most interest to you.  Raise research question.  Formulate objectives,  Assess your objects.
• Double-check (proper explanation of the points is required).

Definition of a variable.

A variable is a label or name that represents a concept or characteristic that varies (e.g. gender, weight, tribe, anxiety, intelligence, attitude towards something, marital status, income, occupation etc.).

Or we can simply define a variable as any item of interest that when manipulated can have more than one possible value and so we say that its value changes.

Types of variables

variables are classified according to the nature and purpose they serve

(A) Classification of variables according to purpose.

Independent & dependent variables (i.e. cause & effect)

(a) The independent variables act as the ’cause’ in that they precede, influence, and predict the dependent variable (effect/outcome). For example, poverty may lead to divorce and so poverty is an independent variable in that context. (Another name for independent variable is explanatory or casual variable).

(b) Dependent variable act as the effect (outcome) in that they change as a result of being influenced by an independent variable. The dependent variable also known as the criterion variable is the variable being predicted.  For example, poverty may lead to divorce and so divorce is a dependent variable in that context.

Note: We are not saying that the variable we have called independent in this example is always independent! For example, laziness can lead to poverty .Thus in this context, poverty is a dependent variable.

Antecedent variable

The term independent and antecedent are synonymous.  An antecedent variable is a variable that precede another variable or consequence in a temporal sense. Although the consequence may not be related, both occur in the same state of time with the antecedent variable occurring first.

For example, the ball hit the window and the old man had a heart attack. In this case, the ball hitting the window came before the heart attack but did not cause it.

Extraneous variable/ background variable

These are variables that influence the outcome (dependent variable) of an experiment (research), though they are not the variables that are actually of interest.

Or simply say these are variables that influence the relationship between the variables that an experimenter is examining.

These variables are undesirable because they add error to an experiment. A major goal in research design is to decrease or control the influence of extraneous variables as much as possible.

How to control extraneous variables?

• A researcher must anticipate and identify the possible extraneous variables, and take appropriate means to control them. The following are the ways of controlling them; Randomizing: by use of random sampling.
• Eliminating: if the extraneous variables are known or Identifiable, they can be eliminated.
• Holding constant: if variables are difficult to eliminate and yet Identifiable, they could be held constant as the desirable ones are manipulated.
• Balancing: a balancing technique can be used to control the variables. This is possible through random assignment of subjects into the control and experimental groups. 
• Statistical adjustments can be used in quantitative studies to control extraneous variables.

It’s important to note that if extraneous variables are not controlled, both internal and external validity are affected.

Confounding variables

Are extraneous variables that vary systematically with the independent variable and exert influence on dependent variable so that the results you get do not reflect the actual relationship between the variables under investigation!.

• Example 1:  Not using counsellors with similar levels of experience in a study comparing the effectiveness of two counselling approaches.
• Example 2:  You want to study whether bottle feeding (cause) is related to an increased risk of diarrhoea in infants (effect). It would seem logical that bottle-fed infants are more prone to diarrhoea since water and the bottle could-get contaminated, milk could go bad, etc. But if you were to conduct this study, you would learn that bottle-fed infants are less likely to develop diarrhoea than breast-fed infants. It would seem that bottle feeding actually protected against the illness. But the truth; you would have missed a very important confounding variable- mother’s education. If you take mother’s education into account, you would learn that educated mothers are more likely to bottle feed their infants, who arc also less likely to develop diarrhoea due to better hygienic practices of their mothers.
• In other words, mother’s education is related to both the cause and effect.
• Not only did the confounding variable suppress the effect of bottle feeding, it even appeared to reverse it- confounding results indicated

 Intervening variable

Is the one that occurs between the independent and dependent variables, it is caused by the independent variable and is itself a cause of the dependent variable. This type of extraneous variables cannot be observed directly or manipulated in any way.

Example 1: A higher education (independent variable) typically leads to a higher income (dependent variable). Occupation is an intervening variable here between education and income because it is casually affected by and itself affects income.

In other words more schooling tends to mean a better job, which in turn tends to bring higher income.

Example 2: Work pressure would increase work distress which in turn would increase drinking. In this case work distress.is an intervening variable.

• NB. Intervening variable and mediator variables are synonymous

Orgasmic Variables:

Are extraneous variables that also cannot be manipulated in any way but have an effect on result of study e.g. -emotions of respondents

Moderator variable

That factor/variable which is manipulated or selected by the experimenter to discover whether it modifies the relationship between independent and dependent variables.

Example: In a study of two methods of teaching reading, one of the methods of teaching reading may work better with boys than girls. Method of teaching reading is the independent variable and reading achievement is the dependent variable. Gender is the moderator variable because it moderates or changes the relationship between the independent (teaching method) and dependent variable (reading achievement)

Treatment variable same as independent variable

Assigned and active Variables

• Any variable that can be manipulated is called active. For example, the variable methods of teaching can be manipulated. One can use role play or lecture method.
• Any variable that cannot be manipulated is called assigned variable. E.g. all variables which are characteristics of respondents, such as: their age, sex, education level.

(B) Classification of variables according to their nature

Continuous, discrete & categorical variables

Continuous variable:

• Is one that may assume any value between two an upper and lower limit. Or simply say is one that can be quantified on an infinite scale E.g. body weight, income, age & 1.Q
• Continuous variables assume-values within a given range or numbers with decimals.

Discrete Variable:

• is a variable that can take only whole number of distinct value. 
• Or simply say a variable whose unit is limited to integer (e.g. number of cigarettes smoked per day, number of people attended a given function).You can also say, is one that can be quantified on a finite scale of integers.
• Some examples will clarify the difference between discrete & continuous variables;
• Suppose the fire department mandates that all fire fighters must weigh between 150 and 250 pounds. The weight of a fire fighter would be an example of a continuous variable; since a fire fighter’s weight could take on any value between 150 and 250 pounds.
• Suppose we flip a coin and count the number of heads. The number of head could be only integer value. We could not for example get 2.5 heads. Therefore the number of heads must be a discrete value.

Categorical variables:

• Are measured and assigned to groups on the basis of specific characteristics. E.g. Gender: male and female, Grade: A-D, social economic status: low, middle, high; Color: blue, black, red.

Quantitative & Qualitative variables

• Quantitative variable: Variables that are measured on a numerical or quantitative scale. Ordinal, interval and ratio scales are quantitative. E.g. A country’s population, a person’s shoe size, a car’s speed are all quantitative variables .Variables which are not quantitative are Qualitative variables.

Qualitative variables:

• Are variables with no natural sense of ordering.
• They are therefore measured on a nominal scale E.g. hair color (Black, gray, red, yellow)
• Qualitative variables can he coded to appear numerical but their numbers are meaningless as in male=1 female=2
• Qualitative variables are also known as categorical variables
• E.g. Gender, religion, and Eye color

Operation definition of variables

• We define a word by telling what actions or behaviours the word expresses. For example, the variable long distance may be defined as a distance of more than 100 miles. This is not a dictionary meaning and not all researchers will define this variable that way. Choose the best definition.

Definition of internal and external validity:

• Internal validity is the degree to which the investigator draws the correct conclusions about what actually happened in the study. 
• Internal validity is affected by flaws within the study it’s self-such not controlling some extraneous variables, or a problem with research instrument.
• “Findings can be said to be internally invalid because they have been affected by factors other than those thought to have caused- them” (Seliger & Shohamy 1989, 95).

Here are some factors which affect internal validity;

subject Variability.

• Time given for the data collection or experimental treatment.
• Instrument/task sensitivity.

• External validity is the degree/ extent to which you can generalize your findings (can be appropriately applied) to people and events outside the study.

“Findings can be said to be externally invalid because  {they} cannot be extended or applied to contexts outside those in which the research took place” (Seliger &Shohamy 1989,95).

Here are some important factors which affect external validity:

• The effect of the research environment (e.g. if experiment is carried out in a highly controlled environment, results may not be able to work out outside).
• Bias in subjects selection to sample (e.g. if you select only male, results may not work on female.)  Researcher or experimenter effects.
• Data collection methodology.
• The effect of time. 
• Sample size.