Scientific Investigation
Definitions of Research
When you say that you are undertaking a research, you imply that the process:
1. Is being undertaken within a framework of a set of philosophies.
2. Uses procedures, methods and techniques that have been tested for their validity and reliability.
3. Is designed to be unbiased and objective.
What is a research
An organized, systematic, data-based, critical, objective, scientific inquiry into a specific problem that needs a solution.
Research is a purposive investigation resulting in an original contribution to the existing knowledge.
“…the systematic process of collecting and analyzing information (data) in order to increase our understanding of the phenomenon about which we are concerned or interested.”
“Careful, systematic, patient study and investigation in some field of knowledge, undertaken to establish facts or principles.” (Grinnell 1993:4)
Grinnell further says ‘research is a structured inquiry that utilizes acceptable scientific methodology to solve problems and creates new knowledge that is generally acceptable.’
Scientific Study
Yields desired results.
Facilitates Critical decision making.
Then, what does the term scientific means?
Not based on hunches, experience, and intuition.
Pursues an organized, logical, step-by-step process.
Purposive & rigorous .
Accuracy and confidence
Objective
• Identify problems• Gather data• Analyze data• Draw valid conclusions
Scientific means
Three essential requirements
• Validity• Reliability• Unbiased and Objective
• Validity is applied to any aspect of the research process.– Validity refers to the application of correct
procedures to find answers to a question.– If the correct procedures applied to find the
answer, then the findings are valid.• Reliability refers to quality of
measurement procedures that provides;– Repeatability– accuracy
• Unbiased and objective – Means that you have taken each step in an unbiased
manner.– Drawn conclusions to the best of your ability and
without introducing your vested interests.– Subjectivity is an integral part of your way of
thinking shaped by:• Educational background, discipline, philosophy,
experience and skills.
– Bias is a deliberate attempt to either conceal or highlight something.
The methods adopted to carry out a research have to fulfill these three essential requirements.
Hallmarks of Scientific Research
• Purposiveness• Rigor• Testability• Replicability
• Precision and confidence• Objectivity• Generalizability• Parsimony
How employee commitment to the organization can be increased?
Purposiveness
• Start with a definite aim/objective/purpose/ focus.
• Research is a purposive investigation.– Focus: increasing the commitment of
employees to the organization.– This will in turn be beneficial in many ways.
• Less turnover• Less absenteeism• Increased performance level
Rigour• A good theoretical base and a sound
methodological design add regour to a purposive study.
• Rigour connotes;– Carefulness (right sample, right respondents, right
information, minimum degree of bias)– Scrupulousness (conformity to high standards of
ethics/excellence)– Degree of exactitude in investigation – Methodological sophistication to ensure;
• Relevance• Appropriateness• Justifiability
• Suppose manager asks 10 or 12 employees to indicate what would increase their level of commitment to the organization. Based on the responses manger reaches several conclusions on how employee commitment can be increased.
• Is this process scientific or unscientific?
How employee commitment to the organization can be increased?
Unscientific. Lacks rigour.
• Lacks rigour due to;1. Conclusions being incorrectly drawn based on
few responses. Responses do not represent entire work force. (lacks methodological sophistication)
2. Manner of framing and addressing the questions could have introduced bias (incorrectness in responses). (lacks methodological sophistication)
3. There can be other reasons that impact on commitment which the respondents could not verbalize during the interviews (hence, researcher failed to include them). (lacks good theoretical foundation)
Testability• Need to formulate hypothesis. (How?)
– You can talk to a randomly selected sample of people (Inductive)
– You can refer to previous studies done on organizational commitment (Deductive)
• Hypothesis should be tested using appropriate statistical tests to the data collected.– Correlation analysis– Chi-square tests– t- test
Employees who receive greater opportunities for participation in decision making will have a higher level of commitment.
Deductive Reasoning
Under deductive method researcher begins with a hypothesis derived from the theory and thereafter gathers empirical evidences to test the hypothesis in order to establish if the hypothesis can be confirmed.
Associated with quantitative research.
Deductive method
StepsWorking theory or argument (RQ)
Hypothesis or proposition
Empirical
verification
Working theory or argument
Deductive method - Example
Motivation – Reinforcement Theory
Hypothesis drawn form the theory
The deductive reasoning
Human beings tend to repeat those actions that results in positive rewards and avoid those actions which results in negative actions.
Rewards followed by performance tend to generate higher efforts leading to increased performance.
Performance Rewards Higher efforts
Increased performance
Inductive ReasoningInquiry is begun with the aid of past observation of phenomenon which might give a clue or a hunch for the researcher to begin the research.
The researcher arrives at a theory or generalization through the accumulation and summation of a variety of past observation / experience concerning the phenomenon being studied.
Generalization is done on past experience or observation.Associated with qualitative research.
• Inductive method– Starts from a specific observation.– Derives general conclusions therefrom.
– You observe the ‘unit’ and conclude on the ‘whole’ based on the observations on the ‘unit.’
All swans which have been observed are white in colour.
Therefore, all swans are white.
• How do you conclude that all the participants (say 30) attending this course are from public sector?
Inductive Reasoning - Example
Unemployment problem and educated youth and youth revolts in Sri Lanka.
How do we look at the youth revolts in Sri Lanka? (1971, 1989)
What are our past experience / observations in relation to youth revolts in Sri Lanka?
Inductive Reasoning - Example
Our past experience and observations:
When unemployment among educated youth increases educated youths rise against the state.
This has happened two times (1971 and 1989) in the history of Sri Lanka.
Researcher generalizes according to past observations (repeated observations) when unemployment increases among educated youth they revolt against the state.
Inductive ReasoningSteps
Past experience or Observation
Hunch
Repeated
observation
Generalization
When the unemployment among the ……..
They revolt against the state
It has happened twice
When the unemployment among the ……..
Replicability• Replicability refers to the occurrence of the
same result in other studies as well.• E.g. based on the results, let us suppose,
that the manager concludes it is the participation in decision making that influences most on the commitment of employees to the organization among other factors.– What if this result is repeated in other studies
using the same method adopted?• The hypo is supported again and again by other
studies.
Precision and Confidence• Difficulty in drawing ‘definitive’
conclusions in research. (Why)– Unable to study the universe of items,
events, population.– Have to base on a sample from the universe.– Sample does not show exact characteristics
of the universe.– Measurement errors.– Biases
Yet, our findings need to be close to the reality!Yet, our findings need to be close to the reality!
Precision– Closeness of findings to the “reality” based on
a sample.– Degree of accuracy or exactitude of the results.– E.g. Researcher estimates that number of
production days lost due to absenteeism is between 30 and 40 as against the actual figure 35.
• Closer to reality• If the estimate is between 25 and 50?
– So, precision is confidence interval in statistics.
Confidence– Being precise is not enough.– Confidence refers to probability of our the
estimate being correct.– We should confidently claim that 95% of the
time our results our results will be true.– Known as confidence level.– In social science research, 95% of confidence
level is accepted as conventional (significance level).
– Scientific sampling design achieves both precision and confidence.
Objectivity• Conclusions/recommendations drawn
should be based on data analysis and interpretations.
• Subjective and emotional values have no room.
• E.g. Greater participation in decision making will increase organizational commitment.
• What if data prove otherwise and researcher wants to prove it?
• Findings answer the problem. (Objectivity)
Generalizability
• It is the scope of applicability of research findings to other settings.
• Wider applicability is expected.• In applied research,
generalizability is limited.– Applied research– Basic research
Parsimony• Simplicity is preferred to complexity.
– Data collection– Data analysis– Generating solutions
• Manageability of research.– Not many variables.
Controlled– Exploring causality in relation to two
variables.– But, in real life there may be more
situations affecting the relationship.– In your research, you have to set up your
study in a way that minimizes the effects of other factors affecting the relationship.
– In social science research we only attempt to quantify their impact.
Systematic– Procedure adopted follows a certain
logical sequence.– Each step should relate to the next step.– Not in haphazard way.
Valid and verifiable– What ever you conclude on the basis of
findings is correct and can be verified by others.
• Empirical– Conclusions drawn are based on hard
evidence gathered from information.
• Critical– Scrutiny of procedures used.– Process of investigation must be foolproof
and free from any drawbacks.– Process adopted can withstand critical
scrutiny.
Scientific Method
The scientific method of research
• Scientific method consists of following elements.
• Concepts –
• Variables -
names or labels used to identify and understand phenomena eg. Power
special kind of concepts containing within them a notion of degree or differentiation. eg. Degree of power distance
The scientific method of research
• Hypothesis -
• Measurements –
• Theories -
proposes a relationship between two or more variables that can be empirically verified. eg. Increased poverty leads to crime.
process of verification of the proposed relationship between variables through field research
consist of a set of related proposition or affirmed hypothses that suggest why event occur in the manner they do.
Step 1 Identification of variables to be studied
• Variables are:
• Independent Variable (IV)
• Dependent Variable (DV)
• Extraneous variable (EV)
Variables
• Social science involves the study of variables and attributes.
• Attributes are characteristics/qualities that describe an object.
Object Characteristics
Male
Female
Extravert
Gleeful
Farmer
A person
• So, variables are logical groupings of attributes.
Gender/SexMale
FemaleVariable
Occupation
Variable
It is this relationship (association, causality) between variables and attributes that science tries to explain.
Step 2 A hypothesis about the relation of one variable to another
• Deductive approach requires the hypo.• Conjectural statements (tentative theories)• Hypo should demonstrate the relationship
between the variables and the concept.• Hypo demonstrates either positive or
negative relationships.• Variables included in the hypo must be clearly
stated and measured.
Step 3 Testing the hypo. to see the field reality• Researcher engaging in an empirical exercise
does the following.• Define the population (who will be covered)• Select a representative sample from the
population• Study the sample using methods to measure the
characteristics of the sample• Process and analyze data gathered to measure if
the hypothesized relationship between the variables can be established.
Step 4 Comparison of measured relationships with the original hypothesis
• This is done to draw generalizations• Depending on the result, the hypo gets
established or changed.
Step 5 Suggestions about the theoretical significance of the
findings
• Findings reveal whether the theory can still be held valid or not.
Hypothetico-deductive method• Developed by Austrian philosopher
Karl Propper.• This method involves seven steps.
1. Identify a broad problem area2. Define the problem statement3. Develop hypothesis (testable, falsifiable)4. Determine measures5. Data collection6. Data analysis7. Interpretation of data
Types of Research• Applied research• Basic research• Action research• Case research (Studies)
Research Strategies
• Survey strategy• Experimental• Action• Case• Ethnographic