w.e.l.c.o.m.e
good morning /an
mm bagali
JAIN University
CMS Business School / Bangalore
"Quantitative and Qualitative for Research"
Department of Management Studies
Cambridge Institute of technology
November 29, 2013.
• Setting
• Objectives
• Hypothesis
• Sampling
• Questionnaire
• Analysis
• Results
elements of research study
Focus for the day
H y p o t h e s i s
Some thinking
– Do we require always
– How big X small the statement should be
– Should it always be proves X disproved
– When to develop X construct
Some statements
That, children's in US watch an average of 3hrs of TV / week
Most people who come to courtroom are innocent
The Tax law have an effect on the Revenue
That larger firms are more efficient in conducting R and D
hy·poth·e·sis
/hīˈpäTHəsis/
Noun
A supposition or proposed explanation made on the
basis of limited evidence as a starting point for further
investigation
A proposition made as a basis for reasoning, without
any assumption of its truth
Synonyms
supposition - assumption - presumption
Hypothesis is derived form the Greek words
 “hypo” means under
 “tithemi” means place
Hypothesis
Def.in.ition
A statement of the predicted relationship between two or
more variables
• Tentative theory or supposition set up and adopted
provisionally as a basis of explaining certain facts or
relationships and as a guide in the further investigation of
other facts or relationships
• A hypothesis is written in such a way that it can be disproven
(null) or proven (alternative) by valid and reliable data
meaning
Under known facts of the problem to explain relationship
between
......... a guess but experienced guess based on some facts
…...is a hunch, assumption, suspicion, assertion or an idea
about a phenomena, relationship, or situation, the reality of
truth of which one do not know
Researcher calls these assumptions, assertions, statements, or
hunches hypotheses and they become the basis of an inquiry.
Results observed X Results you expect
thus,
Written statement
Drawn from experience/observation
Constructed / formulated
Data analysis
Questionnaire
Purpose
• Allow theoretical propositions to be tested in the real
world
• Guide the research design
• Dictate the type of statistical analysis for the data
• Provide the reader with an understanding of the
researchers expectations about the study before data
collecting begins
precautions
properly formulated Ho and H1
one tailed or two tailed
Null or Alternative
Rejection or Acceptance
characteristics
a tentative proposition
unknown validity
specifies relation between two or more variables
simple, specific, and contextually clear
capable of verification
related to the existing body of knowledge
prove X disprove
accept X reject
advantage Hy
• Bringing clarity to the research problem
 provides a study with focus
 signifies what specific aspects of a research problem is to
investigate
 what data to be collected and what not to be collected
 enhancement of objectivity of the study
 formulate the theory
 enable to conclude with what is true or what is false
 The format of the questionnaire
The rationale or sources of hypothesis
• From the researchers own experiences
• From previous research studies
• From theoretical propositions
• Literature available
• Observation
• Discussions
• Historical studies and evidences
Ethical Issue
Hypothesis should always be written before the study and
should not be changed after the study results are
examined
– Don‟t change
– Don‟t alter
– Don‟t add
Variables
– Independent
– Dependent
– Controlled
Terms to know
– M= Mean
– DV= Observed phenomena
– A= significant level
– S= Sample Standard Deviation
– T= t test for degree of freedom (normal population)
– X= Sample mean
– a= alpha-level of significance
– B= beta
Hy
Null
Rejection region
Significance
Sampling distribution
Independent variable
Dependent variables
Level of significance
Types of Hypothesis
Descriptive Hy: The magnitude, trend or behaviour of
population under the study:
• Eg: The attrition rate in BPO is almost 40%
• The literacy rate in Blore is 90%
– Rational Hy: States the expected relationships between two
variables, i.e.: increase, decrease, less than or more
than….
Eg: Higher the exhaustion / stress experience by BPO professionals,
higher the turnover intention
Stating
Hy are used to state the relationship(s) between two
variables and may be stated as :
– Null Hy (one tailed)
– Non Directional
– Directional (Two tailed)
Formulating Null and Alternative Hy
Directional Hy:
The population parameters is structured to be Greater than / Equal
to / Less than / called as ONE tailed test(one sided)
Non Directional Hy:
The population parameter is structured to be equal to a specified
value called as TWO tailed test(two sided)
Criteria while designing hypothesis
• Declaration form
• Uni-dimensional (two variables at a time)
• Measurable
• Based on literature / theories
• Statistical testing
Classifications of hypothesis
Typologies
Simple or complex:
A Simple hypothesis: concerns the relationship between
one independent( cause) and one dependent variable
(effect).
A complex hypothesis:
Concerns a relationship where two or more independent
variables, two or more dependent variables, or both, are
examined in the same study (multivariate)
Hypothesis are used to state the relationship
between two variables and may be stated as
Null hypotheses (no relationship between two variables).
Nondirectional hypotheses (we don‟t know or won‟t speculate
about the direction of the relationship between two variables).
Directional hypotheses. We state the direction of the relationship
between two variables.
Null and research hypothesis
Null hypothesis (Ho)= Statistical hypothesis; predict that
no relationship exists between variables (Rejection
intention)
Research hypothesis(H1)= Alternative hypothesis; state
the expected relationship between variables
(Acceptance intention)
Steps in Testing Hypothesis
• As researchers and management professionals, one must
understand the principles and concepts behind the use of
various statistical methods
• Generalizations from data may be based on models that
require assumptions that may not be appropriate to the
situation
• Understand the role of ‘uncertainty’
How statistics helps research
• Understand the effects of variability and chance
• How many subjects to study
• How long to study a situation
• Are my findings consistent with my hypotheses or can they be explained by
chance or variation
• Obtain estimates of important parameters
• Summarize quantifiable information
• Describe - with precision and accuracy
• Build the evidence of which relationships are likely not due to chance
Examples:
• lifetime of light bulbs
• quality of textile garments
• is training effective?
• what factors predict a successful micro-lending?
• how do I discover fraud in credit card transactions?
So what do these 5 examples have in common?
• Dealing with quantifiable information
• Information obtained on several instances/subjects, not just a single one
• Admit the presence of variability among instances
• Have uncertainty from not observing the entire population of
subjects/instances
• Presence of chance is acknowledged
• Models (approximations of reality) are used
• Models of association: correlation, time series, linear multiple variable
regression
• Data must fulfill some assumptions/requirements for each model
Step 1: State the H0 and H1
» Rejection X Acceptance
» Write / state / construct in such a way that, Null gets
rejected / Alternative gets accepted
» And, null is the basis for argument
» We either "Reject H0 in favour of H1" or "Do not
reject H0"; we never conclude "Reject H1", or even
"Accept H1".
Alternative Hy
• "Do not reject H0", this does not necessarily mean that the
null hypothesis is true, it only suggests that there is not
sufficient evidence against H0 in favour of H1.
• Rejecting the null hypothesis then, suggests that the
alternative hypothesis may be true.
Step 2 : Significance level and Sample size
» 0.05 level / 5 % level
» Big or small or what???
Fixed probability of wrongly rejecting the
null hypothesis H0, if it is in fact true.
Step 3: Determination of a test statistics
» Correlation
» Regression
» Multivariate
» Time series
» Survival analysis
» Students „t‟ test
» Z test (normal distribution)
Step 4: Determination of a Critical Region(CR)
» Rejection region (RR)
» Try to reject null hy
Step 5: Computing the value of the test statistics and
collect the data
Independent and dependent and controlled samples
variables
Computing the value of the test statistics and collect the data
• The scale of measurement determines how a variable is described,
analyzed and interpreted ,
Description
• Tell possible values, or range of values
• Tell likely values to observe in a population
• Tell the central tendency, variability, shape in a sample
• Tell the observed frequency of values in a sample
• Quantify the relationships with other variables
• Analysis
• Infer characteristics of the population from the sample values
• Compare groups with respect to their distribution of this variable
• Establish how it relates to other variables
• Interpretation
• Are characteristics and relationships meaningful / important?
• Are they statistically significant?
Step 6: Making Decision and Conclusions
Rejections
Acceptance
How you conclude results
errors
Type 1 error
In a hypothesis test, a type I error occurs when the null
hypothesis is rejected when it is infact true; that is, H0 is
wrongly rejected
Type 2 error
A type II error occurs when the null hypothesis H0, is not
rejected when it is in fact false.
Relationships specify
How the value of one variable changes in relation to
another
May be either positive, negative, or the two variables may
not have any relationship to one another
Level of Significance
The level of significance for rejecting the statistical null hypothesis
should always be stated before data are collected
The level of significance usually set at (.05). this means that the
researcher is willing to risk being wrong 5% .
Generally the aim of the researcher is to reject the null hypothesis
because this provides support for the research hypothesis.
Fix it to : 0.05 level
Test Statistics
Mathematical formula to test Null Hy
p value
Significance level
variance
Standard Deviation
• P value
– The observed level of significance, is the smallest level at which Ho
can be rejected
– The decision rules for rejecting Ho in the p-value approach are :
• If p-value is greater tha or equal to „a’ , you do not reject the null hy;
• If p-value is less than ‘a’, you rekect the null hy
Thus, Hypothesis Criteria
• Is written in a declarative sentences
• Is written in the present tense. There is a positive
relationship between the number of times children have
been hospitalized and their fear of hospitalization
• Contains the population
• Contains the variables
• Is empirically testable
A relook
Does the study contain a hypothesis or hypotheses?
Is each hypothesis clearly worded and concise?
Is the hypothesis written in a declarative sentences?
Is each hypothesis directly tied to the study problem?
Does each hypothesis contain the population and at least two
variables?
Is it apparent that each hypothesis contain only one
prediction?
if the study contains research questions, are the questions
precise and specific?
Do the research questions further delineate the problem area
of the study?
Example of hypothesis formulation
– Title : Employee empowerment
– Objective: The investigation is an empirical research
work undertaken to understand how a model company
can be created with innovative workplace programme
and policies.
– It was also intended to understand the impact of such
innovative practices on empowerment and how such
processes could change the very face of the organisation
and help it remain at the top of the business
Hy formed
• Ha1 Individual and organisational achievements can be
gained through the sense of belonging;
• Ha2 A sense of Organisational life through climate shapes
behavior and moulds positive attitude towards organisational
growth and development leading to employee empowerment;
• Ha3 Access to information about the mission, value, goals and
objectives of an organization is positively related to
empowerment;
• Ha4 If an organization aspires for fundamental changes, it
must change the fundamentals; and
• Ha5 Empowerment at workplace makes leaders redundant.
Supported by ……
– Data collection
– Questionnaire formulation
– Style of data collection
– Analysis
– Conclusions
• Concluding remarks
• If you know the principles of statistics, you will understand how it can
help you improve the management of processes that are subject to
uncertainty – from variability, sampling, chance
• If you know the methods of statistics, you will know that there are
multiple options and methods to address the same issue – all are
based on models, and thus all are incorrect – but some models are
more useful than others
• If you are clever, you will know that cheaters like to cheat others – but
you will not be cheated !!
Thank you, all
Any questions or comments about the presentation can be sent to
mm.bagali@jainuniversity.ac.in

Hypothesis....Phd in Management, HR, HRM, HRD, Management

  • 1.
    w.e.l.c.o.m.e good morning /an mmbagali JAIN University CMS Business School / Bangalore "Quantitative and Qualitative for Research" Department of Management Studies Cambridge Institute of technology November 29, 2013.
  • 2.
    • Setting • Objectives •Hypothesis • Sampling • Questionnaire • Analysis • Results elements of research study
  • 3.
    Focus for theday H y p o t h e s i s
  • 4.
    Some thinking – Dowe require always – How big X small the statement should be – Should it always be proves X disproved – When to develop X construct
  • 5.
    Some statements That, children'sin US watch an average of 3hrs of TV / week Most people who come to courtroom are innocent The Tax law have an effect on the Revenue That larger firms are more efficient in conducting R and D
  • 6.
    hy·poth·e·sis /hīˈpäTHəsis/ Noun A supposition orproposed explanation made on the basis of limited evidence as a starting point for further investigation A proposition made as a basis for reasoning, without any assumption of its truth Synonyms supposition - assumption - presumption
  • 7.
    Hypothesis is derivedform the Greek words  “hypo” means under  “tithemi” means place
  • 8.
    Hypothesis Def.in.ition A statement ofthe predicted relationship between two or more variables • Tentative theory or supposition set up and adopted provisionally as a basis of explaining certain facts or relationships and as a guide in the further investigation of other facts or relationships • A hypothesis is written in such a way that it can be disproven (null) or proven (alternative) by valid and reliable data
  • 9.
    meaning Under known factsof the problem to explain relationship between ......... a guess but experienced guess based on some facts …...is a hunch, assumption, suspicion, assertion or an idea about a phenomena, relationship, or situation, the reality of truth of which one do not know Researcher calls these assumptions, assertions, statements, or hunches hypotheses and they become the basis of an inquiry. Results observed X Results you expect
  • 10.
    thus, Written statement Drawn fromexperience/observation Constructed / formulated Data analysis Questionnaire
  • 11.
    Purpose • Allow theoreticalpropositions to be tested in the real world • Guide the research design • Dictate the type of statistical analysis for the data • Provide the reader with an understanding of the researchers expectations about the study before data collecting begins
  • 12.
    precautions properly formulated Hoand H1 one tailed or two tailed Null or Alternative Rejection or Acceptance
  • 13.
    characteristics a tentative proposition unknownvalidity specifies relation between two or more variables simple, specific, and contextually clear capable of verification related to the existing body of knowledge prove X disprove accept X reject
  • 14.
    advantage Hy • Bringingclarity to the research problem  provides a study with focus  signifies what specific aspects of a research problem is to investigate  what data to be collected and what not to be collected  enhancement of objectivity of the study  formulate the theory  enable to conclude with what is true or what is false  The format of the questionnaire
  • 15.
    The rationale orsources of hypothesis • From the researchers own experiences • From previous research studies • From theoretical propositions • Literature available • Observation • Discussions • Historical studies and evidences
  • 16.
    Ethical Issue Hypothesis shouldalways be written before the study and should not be changed after the study results are examined – Don‟t change – Don‟t alter – Don‟t add
  • 17.
  • 18.
    Terms to know –M= Mean – DV= Observed phenomena – A= significant level – S= Sample Standard Deviation – T= t test for degree of freedom (normal population) – X= Sample mean – a= alpha-level of significance – B= beta
  • 19.
    Hy Null Rejection region Significance Sampling distribution Independentvariable Dependent variables Level of significance
  • 20.
    Types of Hypothesis DescriptiveHy: The magnitude, trend or behaviour of population under the study: • Eg: The attrition rate in BPO is almost 40% • The literacy rate in Blore is 90% – Rational Hy: States the expected relationships between two variables, i.e.: increase, decrease, less than or more than…. Eg: Higher the exhaustion / stress experience by BPO professionals, higher the turnover intention
  • 21.
    Stating Hy are usedto state the relationship(s) between two variables and may be stated as : – Null Hy (one tailed) – Non Directional – Directional (Two tailed)
  • 22.
    Formulating Null andAlternative Hy Directional Hy: The population parameters is structured to be Greater than / Equal to / Less than / called as ONE tailed test(one sided) Non Directional Hy: The population parameter is structured to be equal to a specified value called as TWO tailed test(two sided)
  • 23.
    Criteria while designinghypothesis • Declaration form • Uni-dimensional (two variables at a time) • Measurable • Based on literature / theories • Statistical testing
  • 24.
    Classifications of hypothesis Typologies Simpleor complex: A Simple hypothesis: concerns the relationship between one independent( cause) and one dependent variable (effect).
  • 25.
    A complex hypothesis: Concernsa relationship where two or more independent variables, two or more dependent variables, or both, are examined in the same study (multivariate)
  • 26.
    Hypothesis are usedto state the relationship between two variables and may be stated as Null hypotheses (no relationship between two variables). Nondirectional hypotheses (we don‟t know or won‟t speculate about the direction of the relationship between two variables). Directional hypotheses. We state the direction of the relationship between two variables.
  • 27.
    Null and researchhypothesis Null hypothesis (Ho)= Statistical hypothesis; predict that no relationship exists between variables (Rejection intention) Research hypothesis(H1)= Alternative hypothesis; state the expected relationship between variables (Acceptance intention)
  • 28.
    Steps in TestingHypothesis
  • 29.
    • As researchersand management professionals, one must understand the principles and concepts behind the use of various statistical methods • Generalizations from data may be based on models that require assumptions that may not be appropriate to the situation • Understand the role of ‘uncertainty’
  • 30.
    How statistics helpsresearch • Understand the effects of variability and chance • How many subjects to study • How long to study a situation • Are my findings consistent with my hypotheses or can they be explained by chance or variation • Obtain estimates of important parameters • Summarize quantifiable information • Describe - with precision and accuracy • Build the evidence of which relationships are likely not due to chance
  • 31.
    Examples: • lifetime oflight bulbs • quality of textile garments • is training effective? • what factors predict a successful micro-lending? • how do I discover fraud in credit card transactions?
  • 32.
    So what dothese 5 examples have in common? • Dealing with quantifiable information • Information obtained on several instances/subjects, not just a single one • Admit the presence of variability among instances • Have uncertainty from not observing the entire population of subjects/instances • Presence of chance is acknowledged • Models (approximations of reality) are used • Models of association: correlation, time series, linear multiple variable regression • Data must fulfill some assumptions/requirements for each model
  • 33.
    Step 1: Statethe H0 and H1 » Rejection X Acceptance » Write / state / construct in such a way that, Null gets rejected / Alternative gets accepted » And, null is the basis for argument » We either "Reject H0 in favour of H1" or "Do not reject H0"; we never conclude "Reject H1", or even "Accept H1".
  • 34.
    Alternative Hy • "Donot reject H0", this does not necessarily mean that the null hypothesis is true, it only suggests that there is not sufficient evidence against H0 in favour of H1. • Rejecting the null hypothesis then, suggests that the alternative hypothesis may be true.
  • 35.
    Step 2 :Significance level and Sample size » 0.05 level / 5 % level » Big or small or what??? Fixed probability of wrongly rejecting the null hypothesis H0, if it is in fact true.
  • 36.
    Step 3: Determinationof a test statistics » Correlation » Regression » Multivariate » Time series » Survival analysis » Students „t‟ test » Z test (normal distribution)
  • 37.
    Step 4: Determinationof a Critical Region(CR) » Rejection region (RR) » Try to reject null hy
  • 38.
    Step 5: Computingthe value of the test statistics and collect the data Independent and dependent and controlled samples
  • 39.
    variables Computing the valueof the test statistics and collect the data • The scale of measurement determines how a variable is described, analyzed and interpreted , Description • Tell possible values, or range of values • Tell likely values to observe in a population • Tell the central tendency, variability, shape in a sample • Tell the observed frequency of values in a sample • Quantify the relationships with other variables • Analysis • Infer characteristics of the population from the sample values • Compare groups with respect to their distribution of this variable • Establish how it relates to other variables • Interpretation • Are characteristics and relationships meaningful / important? • Are they statistically significant?
  • 40.
    Step 6: MakingDecision and Conclusions Rejections Acceptance How you conclude results
  • 41.
    errors Type 1 error Ina hypothesis test, a type I error occurs when the null hypothesis is rejected when it is infact true; that is, H0 is wrongly rejected Type 2 error A type II error occurs when the null hypothesis H0, is not rejected when it is in fact false.
  • 42.
    Relationships specify How thevalue of one variable changes in relation to another May be either positive, negative, or the two variables may not have any relationship to one another
  • 43.
    Level of Significance Thelevel of significance for rejecting the statistical null hypothesis should always be stated before data are collected The level of significance usually set at (.05). this means that the researcher is willing to risk being wrong 5% . Generally the aim of the researcher is to reject the null hypothesis because this provides support for the research hypothesis. Fix it to : 0.05 level
  • 44.
    Test Statistics Mathematical formulato test Null Hy p value Significance level variance Standard Deviation
  • 45.
    • P value –The observed level of significance, is the smallest level at which Ho can be rejected – The decision rules for rejecting Ho in the p-value approach are : • If p-value is greater tha or equal to „a’ , you do not reject the null hy; • If p-value is less than ‘a’, you rekect the null hy
  • 46.
    Thus, Hypothesis Criteria •Is written in a declarative sentences • Is written in the present tense. There is a positive relationship between the number of times children have been hospitalized and their fear of hospitalization • Contains the population • Contains the variables • Is empirically testable
  • 47.
    A relook Does thestudy contain a hypothesis or hypotheses? Is each hypothesis clearly worded and concise? Is the hypothesis written in a declarative sentences? Is each hypothesis directly tied to the study problem?
  • 48.
    Does each hypothesiscontain the population and at least two variables? Is it apparent that each hypothesis contain only one prediction? if the study contains research questions, are the questions precise and specific? Do the research questions further delineate the problem area of the study?
  • 49.
    Example of hypothesisformulation – Title : Employee empowerment – Objective: The investigation is an empirical research work undertaken to understand how a model company can be created with innovative workplace programme and policies. – It was also intended to understand the impact of such innovative practices on empowerment and how such processes could change the very face of the organisation and help it remain at the top of the business
  • 50.
    Hy formed • Ha1Individual and organisational achievements can be gained through the sense of belonging; • Ha2 A sense of Organisational life through climate shapes behavior and moulds positive attitude towards organisational growth and development leading to employee empowerment; • Ha3 Access to information about the mission, value, goals and objectives of an organization is positively related to empowerment; • Ha4 If an organization aspires for fundamental changes, it must change the fundamentals; and • Ha5 Empowerment at workplace makes leaders redundant.
  • 51.
    Supported by …… –Data collection – Questionnaire formulation – Style of data collection – Analysis – Conclusions
  • 52.
    • Concluding remarks •If you know the principles of statistics, you will understand how it can help you improve the management of processes that are subject to uncertainty – from variability, sampling, chance • If you know the methods of statistics, you will know that there are multiple options and methods to address the same issue – all are based on models, and thus all are incorrect – but some models are more useful than others • If you are clever, you will know that cheaters like to cheat others – but you will not be cheated !!
  • 53.
    Thank you, all Anyquestions or comments about the presentation can be sent to mm.bagali@jainuniversity.ac.in