Formulating the Research Problem

Research Steps
Formulating the research problem Conceptualizing and Operationalizing Constructing an instrument for data collection

Selecting a sample
Writing a research proposal Collecting data Processing data Writing the report

The research problem

Some considerations

Researchable Complex enough Not absolute truth Two inter-related concepts

Problem definition
Most crucial Most difficult Foundation for a sound research Whats a research problem? A research problem can be viewed as a gap between the existing state of affairs and the desirable state of affairs.

 Identification of research problem is important.

Clear and economical plan is impossible in the absence of an identified research problem. Provides a good foundation for the research. Validity / reliability Have a clear idea about what you want to find out and NOT what you think you must find. Research objectives Research problem may take number of form from simple to complex.

 Clear identification of research problem facilitates the determination of:

The type of study design The type of sampling strategy The type of research instruments The type of analysis

Garbage in garbage out.

Choosing a topic
Think about
Public Admin.
Subject matter of your degree programme

Human Re. Mgt.

An interesting course unit

An attractive field unit within that unit

Training & Dev.

Labour turnover

A topic that interest you

Choosing a Research Problem

Expect the research questions to evolve and change during a qualitative study the assumption is of an emerging design. In quantitative designs, the research questions remain fixed throughout the study. At the beginning one may become confused. But, it is quite natural.
Confusion is often but a first step towards clarity.

Considerations when choosing a research topic

Choice you have Your motivation Regulations Expectations Previous research Size of the topic Time available Resources available Cost of research Need for support Access Methods for researching

In case you dont have a topic

Ask the supervisor, colleagues, friends Use previous research Develop on practice at work Relate to interests Just think of a title Start from a quote that changes you Follow hunches Draw picture, diagram Just start anywhere Be prepared to change direction

 To formulate a research problem, a knowledge in the subject area is a must.

Identification of the destination before undertaking the journey.

 Not all questions can be transformed into RPs.

Potential research questions may occur to us on a regular basis, but the process of formulating them in a meaningful way is not at all an easy task. Power, Meenaghan & Twoomy (1985:38)

To form an RP, considerable knowledge of both subject area and research methodology is required.

Sources of research problem

Four Ps.
People Problems Programmes Phenomena

Most research in the humanities revolves around four ps.

 Most of the research studies are based upon at least a combination of two ps.
How has the Samuirdhi Programme impacted in alleviating poverty in women?
People

Women

Problems

Impact
Phenomena

Poverty
Program Samurdhi

 People
Others Ps

Study Population
Subject areas

Every research has two aspects.

The study population

The subject area

Aspects of a research problem

Aspects of Study
Study population

About
People

Study of
- Individuals - Organizations - Groups - Communities - Issues - Situations - Associations - Profiles etc. - Contents - Structures - Outcomes - Attributes - Satisfaction - Service providers etc. - Course and effect relationships - Study of phenomena They provide you with required info or you collect info from them

Problem

Subject area

Programmes

Information that you need to collect to find answers to your research questions

Phenomena

Considerations in selecting a research problem

Interest Magnitude Motivation

Visualize the work involved Manageability/specificity/clarity Concept measurements

Indicators Measurement

Level of expertise
Adequate level

 Relevance
Relevant to the field of interest Up-to-datedness Timeliness Organizations objectives etc.

Availability of data
Accessibility Availability Form of data

Ethical issues
Anonymity Confidentiality/ privacy Relationship

Steps
Identification of a broad field/subject area of interest
Field of interest Field of employment ..

Dissection of broad area into sub-areas

Broad areas have many aspects. Hence, need to dissect.

Broad subject area

Sub-areas
Profiles of families in which DV occurs Profile of the victims of DV

Profile of the perpetrators

Domestic Violence

Reasons for DV Extent and type of DV Impact of DV on family Services available to the victims Effectiveness of services provided

 Selection of the most interesting area Not all areas are feasible to study. Select the area you really want to study. Forget areas you are not interested in. Think of;
Time Accessibility to data Resources etc.

 Raise research question

Ask the question what is it that I want to find out about in this area? List whatever questions you want to find answers to. Eliminate if there are many questions.

Research question Example

Research question is derived from the research problem.
Eg. Why is employee performance below expectation? Q1 Do the employees who are satisfied with their pay work better than those who are not satisfied with their pay?

Q2 Do Sri Lankan Employees perform better under conditions of strict supervision?

 Formulate objectives
What are your main objective (overall)? What are your sub-objectives (specific)? Develop objectives in line with research questions
Objectives grow out of the research question

Difference between objectives and research question (Only difference is the way it is written) Objectives to start with action verbs Objectives transform questions into behavioural aims

 Assess the objectives

Examine the objectives to ascertain the feasibility of achieving them. Consider them in terms of :
Time Resources Technical expertise

Double check
Are you still interested in the topic? Ready with the necessary resources? If no, reassess objectives

Suppose you want to conduct a study in the area of alcoholism. What steps you will take in formulating your research problem?

Step 1 Identify Alcoholism

Step 2 Dissect
Profile of alcoholics Causes of alcoholism Effects on family Treatment methods Community attitudes Process of becoming

Step 3 Select

Step 4 Raise Qs
What impact has alcoholism on material relations? How does it affect children? What are the effects on familys finances?

Effects of alcoholism on the family

Step 7
Double check
Still interested? Agree with objcs? Have resources? Have expertise?

Step 6 Make sure

Assess the objs. In the light of:
Work involved Time available Resources Technical expertise

Step 5 Formulate objectives

Main Obj.
To find out the effects of alcoholism on the family.

Specific Obj.
To Ascertain the impact of alcoholism on material relations. To find out ways in which alco affects childrens lives. To find out effects on the fin, situation.

Group exercise - Research areas

1. 2. 3. 4. 5. 6. People Empowerment Crime wave Unemployment Public management Political behaviour Transportation

Group work: Select an area of your interest to conduct a study. What steps you will take in formulating your research problem?

Step 1 Identify

Step 2 Dissect

Step 3 Select

Step 4 Raise Qs

Step 7
Double check

Step 6 Make sure

Step 5 Formulate objectives

Main Obj.
Specific Obj.

 If the study is descriptive;

Your main objective should clearly describe the major focus of the study. Mention the organization, location etc. if those are not confidential.
E.g. to describe the types of treatment programmes by(name of the organization) to drug addicts in (name of the place).

 If the study is correlational

Main objective should include the main variable being correlated in addition to the three properties.
E.G. to ascertain the impact of migration on family role. to compare the effectiveness of different teaching methods on the comprehension of students

 If the study is to test a hypothesis

In addition to the above characteristics, the wording of the main objective should indicate the direction of the relationship being tested.
E.g. to ascertain if an increase in youth unemployment will increase the incidence of street crime. Increase in youth unemployment will increase incidence of street crime.

Forming research problem

Interrogative (question) form Descriptive (statement) form

Research problem cont.

Example: Major area: Concept 1: Concept 2: Criminal Justice Death penalty Violent crimes

Research problem cont.

Forming the RP
1. The effect of .(concept 1) (concept 2)
2. The role of . (con 1) in.(con 2) 3. The use of(con1) in(con 2)

RP cont.
The effect of the death penalty on violent crime rates in Sri Lanka. How does death penalty affect the commitment of crimes in Sri Lanka?

RP Cont.
Always start the research question with How or Why (Implies somewhat broader investigation) Avoid using where, who, when (suggests narrow scope)

Features of RP
The right topic should be interesting, complex, and compelling. Problem is significant. It should be delineated. Should be able to obtain information required. Should be able to draw conclusions related to the problem. Should be able to state the problem clearly and concisely.

Concepts and Variables and Hypothesis

What is a variable?
An image, perception or concept capable of measurement (capable of taking different values) is called a variable. A concept that can be measured is a variable.
Try measuring these concepts.

Excellent Well Like wonderful

Effective Quality Waste of time Discriminated accountability

Concept and a variable - Difference

Concept Mental images Perceptions Meanings vary remarkably from person to person Cannot be measured Subjective impression Variable Measurable Varying degree of accuracy Can use unit of measurement

Measurability is the main difference between a concept and a variable.

Concepts
Effectiveness Satisfaction Impact Excellent Rich Domestic violence High-achiever Alcohol consumption

Variables
Gender (male/female) Attitude Age Income Weight Height Religion

Subjective impressions. Different understanding. Not measurable.

Measurable. But the degree of precision varies.

Concepts, indicators & variables

If you are using a concept in your study, you need to operationalize it (Showing how the concept is measured).
To operationalize a concept, you need to identify indicators (A set of criteria reflective of the concept). Criteria are then converted into variables.
Concept Indicator Variable

Rich

Criteria
Income Assets

Income per year Land House etc.

Job satisfaction of employees Concept (DV) Indicator Variable Measure

Commitment to Working hours work Punctuality Arrival time Departure time Increase in output Output produced A & D time No. of hours

No. of outputs

Job satisfaction

Reduced customer No. of reduced complaints customer complaints

Creative ideas

New ideas
Ideas accepted Ideas implemented Number

Types of variables
Can be viewed from three different ways.

1.Causal relationship 2.Design of the study 3.Unit of measurement

From the viewpoint of causation

Studies attempting to investigate causal relationship or association. (cause and effect) Four types of variables may operate.
1. Change variables (responsible for bringing about
change in a phenomena) 2. Outcome variables ( effects of a change variable)

3. Affect variables (affects the link between cause and

effect variables)

4. Linking variables (completes the relationship between

cause and effect variables)

Types of variables in a causal relationship

Linking or connecting variables 4 Intervening/confounding variables CV

Cause
Change variables 1 Independent variable IV

Effect
Outcome variables 2 Dependent variables DV

Variables affecting relationship 3 Extraneous variables EV

IV,DV, and EV in a causal relationship

Smoking Assumed cause Independent variable Cancer Assumed effect Dependent variable

Affect the relationship

Age of the person Extent smoking Duration of smoking Extent of daily exercise etc.

Extraneous variables

IV,DV, EV, and CV in a causal relationship

University Education Independent variable Availability of bursary Confounding variable
Parents professions

Student achievements Dependent variable

Students background Girl friends/boyfriends Political party affiliations etc.

Extraneous variables

From the viewpoint of study design

Two types of variables.
1. Active variables
Variables that can be manipulated, changed or controlled. ( different teaching models etc.)

2. Attribute variables
Variables that cannot be manipulated, changed or controlled.
age, sex, income etc.

Active and attribute variables

Study intervention
Different teaching models Experimental intervention Programme service etc. Active variables A researcher can manipulate

Study population
Age Gender Level of motivation Attitudes etc. Attribute variables A researcher cannot manipulate

From the viewpoint of the unit of measurement

Under this, there two ways of categorizing variables.
1. Categorical variables 2. Continuous, qualitative, and quantitative variables

Categorical variables
Measured on nominal or ordinal measurement scales.
Gender
Male Female
Nominal

Income
Above average Average Below average
Ordinal/Ranking

Categorical variables cont.

Categorical variables are of three types.
1. 2. 3. Constant Dichotomous Polytomous Variable can have only one value or category. (taxi, tree, water etc.) Variable can have only two categories (Good/bad, Yes/No, Rich/Poor etc.)

Constant variable

Dichotomous variable

Polytomous variable Variable can be divided into more than two categories.
Religion (Buddhism, Hinduism, Christianity) Race (Sinhala, Tamil, Muslim) Political parties (Democratic, Socialist, Liberal etc.)

Continuous, quantitative, qualitative variables

Continuous variable
Can take infinite number of values in an interval.
Height (4 7, 5 7, 59, 6 etc.) Weight Income

Quantitative variables
Measured on a numeric scale.
Temperature (C or F) Age (years, months)

Qualitative variables
Use either nominal or ordinal measurements.
Hair color (Black, Brown, Gray, Red, Yellow) Name (Mary, Nalin, Jeevani, Edward) Gender (Male, female)

Levels of Measurements
Four common levels of measurements. 1. Nominal scale
No ordering is implied, and addition/subtraction and multiplication/division would be inappropriate for a variable on a nominal scale.
Gender {Female, Male} Religion {Buddhist, Christian, Hindu, Muslim}

2. Ordinal scale
Set of ordered values.
This lecture is (Very Poor, Poor, Average, Good, Very Good)

3. Interval scale
Numerical scales (Ordered and equal intervals on the underlying trait.)
Temperature (f 30 and f 40, f 80 and f 90)

4. Ratio scale
Numerical scale with a true zero point.
Market share, income group, age group, etc.

Hypothesis

 A hypothesis is a specific, testable prediction about what you expect to happen in your study.
For example, a study designed to look at the relationship between study habits and test anxiety might have a hypothesis that states,
students with better study habits will suffer less test anxiety.

Unless your study is exploratory in nature, your hypothesis should always explain what you expect to happen during the course of your experiment or research.

 Tentative explanation about the relationship between two or more variable. Provide a possible solution to the problem. Helps researcher in exploring a research issue of interest to him/her. Initially, loosely set out but has to be precise later. Either proved or disproved. Should be simple.

Hypothesis and the research problem

Hypothesis
Provides a tentative answer to the problem. Suggests specific strategy to sole the problem. Can be shown to be false through research. Should have at least two or more variables.

Research problem
Raises how variables are related to one another. Suggests resolution through empirical research. No evidence would lead to rejection of the problem statement. Should have at least two or more variables.

Null Hypotheses
State that there is no relationship between two variables. The opposite of the non-directional research hypothesis. Example: Colors of clothing has no impact on peoples mood.

Alternative hypothesis
The alternative hypothesis is the hypothesis used in hypothesis testing that is contrary to the null hypothesis. It is usually taken to be that the observations are the result of a real effect. Examples: Colors of clothing affect peoples mood. (Nondirectional hypothesis)

Bright color of clothing uplifts peoples mood.

(Directional hypothesis)

Validity and Reliability

Validity
Validity is the
Extent to which a test measures what it claims to measure. It is vital for a test to be valid in order for the results to be accurately applied and interpreted. Validity is the extent to which a question or scale is measuring the concept, attribute or property it says it is.
face validity content validity construct validity

Reliability
Reliability is the extent to which an experiment, test, or any measuring procedure yields the same result on repeated trials. Refers to the consistency of a measure.
For example, if a test is designed to measure a trait (such as introversion), then each time the test is administered to a subject, the results should be approximately the same.

Assignment
Identifying complexities of problem definition
Broader Area: Educated Unemployed Tasks:

1. What is excluded in defining the research problem? 2. What are other concerns if the problem is to be defined more sharply? 3. Set objectives for the identified problem.

Answers
Exclusions: Educated who are employed

Unemployed who are uneducated

Objectives
To find out characteristics of educated unemployed To find out information to create employment opportunities To design training programmes to prepare them for specific jobs To arrange them in some order of priority to provide them with jobs when opportunities become available

Other concerns
Are we concerned with all educated unemployed? Concerned with only those who have attained a certain level of education? Educated in any subject? Educated in any particular subject? Concerned with under-employment? Include those who are completely without any job?

Unless: we think clearly and define more precisely our research problem and objectives,

we know from the start what questions we desire to answer eventually,

we will not know what specific information we will find useful for understanding the problem or for answering these questions.