Uploaded by Girish P

Questionnaire and hypothesis

advertisement
Hypotheses and Variables
OBJECTIVES
Example
Research problem
A survey carried out s ress level in a teaching hospital
OBJECTIVES FOR THE STUDY:
•To identify the numbers of patients with arthiritis
•To identify the type of t
•To discover the treatments being used
•To discover if the sores were improving, deteriorating or static
•To discover when the sores had occurred, i.e. prior to admission or on the ward
•To list any support systems in use
•To identify the degree of risk of pressure sore development of all patients in the
hospital
•To identify any factors which are of particular relevance to tissue breakdown.
Hypothesis
HYPOTHESIS IS NOT A QUESTION, BUT RATHER IT IS A
STATEMENT ABOUT THE RELATIONSHIP BETWEEN TWO
OR MORE VARIABLES.
To be complete a hypothesis must include three
components:
The variables
The population
The relationship between the variables
Hypothesis
• Characteristics of Hypothesis:
A hypothesis must be precise and clear. If it is not precise and clear, then the
inferences drawn on its basis would not be reliable.
A hypothesis must be capable of being put to test. Quite often, the research
programmes fail owing to its incapability of being subject to testing for
validity. Therefore, some prior study may be conducted by the researcher in
order to make a hypothesis testable. A hypothesis “is tested if other deductions
can be made from it, which in turn can be confirmed or disproved by
observation” (Kothari, 1988).
Hypothesis
Characteristics of Hypothesis:
A hypothesis must state relationship between two variables, in the case of
relational hypotheses.
A hypothesis must be specific and limited in scope. This is because a simpler
hypothesis generally would be easier to test for the researcher. And therefore,
he/she must formulate such hypotheses.
As far as possible, a hypothesis must be stated in the simplest language, so as to
make it understood by all concerned. However, it should be noted that simplicity
of a hypothesis is not related to its significance.
A hypothesis must be consistent and derived from the most known facts. In other
words, it should be consistent with a substantial body of established facts. That is,
it must be in the form of a statement which is most likely to occur.
A hypothesis must be amenable to testing within a stipulated or reasonable period
of time. No matter how excellent a hypothesis, a researcher should not use it if it
cannot be tested within a given period of time, as no one can afford to spend a lifetime on collecting data to test it.
Hypothesis
Testing of hypothesis:
1.
As a part of investigation, samples are drawn from the population and
results are derived to help in taking the decisions. But such decisions
involve an element of uncertainty causing wrong decisions.
2.
Hypothesis is an assumption which may or may not be true about a
population parameter. For example, if we toss a coin 200 times, we may get
110 heads and 90 tails.
3.
At this instance, we are interested in testing whether the coin is unbiased or
not. Therefore, we may conduct a test to judge the significance of the
difference of sampling or otherwise. To carry out a test of significance, the
following procedure has to be followed:
Hypothesis
Framing the Hypothesis:
To verify the assumption, which is based on sample study, we collect data and find out the difference
between the sample value and the population value. If there is no difference found or the difference is
very small then the hypothetical value is correct. Generally two hypotheses complementary to each offer
are constructed, and if one is found correct, the other is rejected.

Null Hypothesis: The random selection of the samples from the given population makes the tests
of significance valid for us. For applying any test of significance we first set up a hypothesis- a
definite statement about the population parameter/s. Such a statistical hypothesis, which is under
test, is usually a hypothesis of no difference and hence is called null hypothesis. It is usually
denoted by Ho. In the words of Prof. R.A.Fisher “Null Hypothesis is the hypothesis which is tested
for possible rejection under the assumption that it is true.”

Alternative Hypothesis:Any hypothesis which is complementary to the null hypothesis is called an
alternative hypothesis. It is usually denoted by H1. It is very important to explicitly state the
alternative hypothesis in respect of any null hypothesis H0 because the acceptance or rejection of
Ho is meaningful only if it is being tested against an opposite hypothesis.
.
POPULATION
A population is what we call the entire group of individuals or elements who
meet the sampling criteria.
A sample is representative of that population.
Example: If we were interested in looking at the number of childhood cancers
in 2006 in the United Kingdom (i.e. population), we obviously could not
survey the entire population of children with cancer in that year who live in the
United Kingdom, and so consequently we would look at a smaller sample
taken from all the children with cancer in 2006 who live in the United
Kingdom.
•The individual units of a population are what we call the elements.
•Now an element can be anything that we are studying, for example it could be
a person, an event, their behaviour, or indeed any other single unit of a study.
•However, when elements are actually human beings, then rather than calling
them elements, we call them subjects.
CORRELATION
CORRELATIONS
The degree of relationship between two or
more variables, or between two or more sets of data, is
called linear correlation.
The degree of relationship is expressed by the coefficient
of correlation, and is symbolised by r.
The closer r is to 1.00 (either negative or positive) the
stronger the relationship.
RELATIONSHIP
This means the relationship between one variable and
another, for example, smoking and lung cancer.
Characteristics & Qualities of a
good Hypothesis










Predict a relationship between two or more variables
Observable testable
Be justifiable (based on rationale/theory)
Simple
Clear
Relevant to the problem
Specific
Relevant to the existing techniques
Fruitful
Consistent
TYPES OF ERRORS IN TESTING
HYPOTHESIS
Types of errors in testing of hypothesis:
The inductive inference consists in arriving at a decision to accept or reject a null hypothesis (Ho) after inspecting
only a sample from it. As such an element of risk – the risk of taking wrong decision is involved. In any test
procedure, the four possible mutually disjoint and exhaustive decisions are:
 Reject Ho when actually it is not true i.e., when Ho is false.

Accept Ho when it is true.

Reject Ho when it is true.

Accept Ho when it is false.
TYPES OF HYPOTHESIS
A hypothesis can be classified into six types: simple, complex,
associative and causal, directional, non-directional and null. In research,
a hypothesis is characterized by three essential elements: variables,
population and the correlation between the variables.
Associative hypothesis
Casual hypothesis
Simplex hypothesis
Complex hypothesis
Logical hypothesis
Statistical hypothesis
Causal vs Associative Hypothesis
Associative hypotheses
Propose relationships between variables - when one variable changes,
the other changes.
 Do not indicate cause and effect.
Causal hypothesis
 Propose a cause and effect interaction between two or more
variables.
Causal :Dieting women participating in a formal exercise
regimen will have greater weight loss than dieting women
without an exercise program
Associative:There is a positive relationship between amount
of exercise and weight loss among dieting women
Simple vs. Complex
Simple vs. Complex
Simple hypothesis – There exists relationship between the two
variables ( independent variable and dependent variable)
Eg: Smoking leads to cancer
Complex hypothesis
There exists relationship among variables ( independent variable and
dependent variables are more than two)
Eg: Smoking and other drugs leads to cancer
 Infants born to heroin-addicted mothers have lower birth weight
than infants of non-addicted mothers
 Infants born to heroin-addicted mothers have lower birthweight,
more neurologic complications, and higher mortality than infants of
non-addicted mothers
Logical and Statistical
hypothesis
Logical hypothesis
Hypothesis is verified logically.Eg.agreement, difference etc.
Statistical hypothesis
Hypothesis is verified statistically
Directional Hypothesis VS
Non-Directional Hypothesis
Directional Hypothesis
One can access the direction of or effect of one variable on
the other.
There will be a difference between the performance of
children based on their age.
Non-Directional Hypothesis
One does predict kind of effect of one variable on the other
but can state the relationship between variable 1 and
variable 2.
There will be a difference between the performance of elder children
than younger children
Empirical Hypothesis Vs Null hypothesis
Empirical Hypothesis
Working hypothesis is applied in field . during formulation it
is as assumption after testing in the field it becomes a
working hypothesis
Null hypothesis
 It is contrary to the working hypothesis and states there is no relationship
between the independent and dependent variable.
Example
 There is no relationship between gender and knowledge of transmitted
diseases among teenagers
 Teenage boys are better informed about transmitted diseases than
teenage girls
Variables
Variables are'qualities, properties, and or characteristics of
persons, things, or situations that change or vary, and that can be
manipulated, measured, or controlled in a research study.' (Burns
& Groves 2005:755)
There are different types of variables, namely:
Dependent variables;
Independent variables.
A dependent variable is the response, the behaviour, or the outcome that is
predicted and measured in research.Changes in the dependent variable are
presumed to be caused by the independent variables.An independent variable
is the treatment, the intervention, or the experimental activity that is
manipulated or varied by the researcher during the research study in order to
create an effect (i.e. change) on the dependent variable.
Types of Variables?
Qualities or characteristics that vary between individuals,
things, or situations
Independent
Dependent
◦ Cause
◦ Action
◦ Intervention
Effect, outcome
Reaction
Response
Same Variable Can Be Independent or Dependent depending on the
research design and question
Influence of a nurse’s experience with Influence of a nurse’s FAITH in
death and dying
religious activity
nurse’s degree of FAITH religious
activity
Nurses attitude towards death and
dying
IV=Death experience
IV=Religious activity
DV=Religious activity
DV=Death experience
It Takes All Kinds
Research variables (which could be independent or dependent or
neither) are those being studied
Demographic variables describe the population (and may be
confounding variables)
Extraneous (confounding) variables influence relationships being
studied
Hypothesis Testing
Is also called significance testing
Tests a claim about a parameter using evidence (data in a sample
The technique is introduced by considering a one-sample z test
The procedure is broken into four steps
Each element of the procedure must be understood
Hypothesis Testing Steps
A. Null and alternative hypotheses
B. Test statistic
C. P-value and interpretation
D. Significance level (optional)
Questionnaire
Development &
Validation
QUESTIONNAIRE DEVELOPMENT
Questionnaire is a formalized set of questions to obtain certain
information from certain respondents.
Developing a good questionnaire is NOT EASY!
•It takes time, time, and more time.
•You may end up writing multiple drafts.
•It helps to work with others when developing a questionnaire.
2
5
Types of Questionnaire
1 Self-administrated
Computer assisted and Web-based
Paper and pencil
2 Interviewer- administrated
•Face to face
•Telephone
2
6
Self-administered Questionnaire
Advantages:
1. Cheap and easy to
Disadvantages:
1.Low response rate
administer
2. Preserves confidentiality
2.Questions can be
misunderstood
3. Completed at any time
4. No influence by
interviewer
5. Allow for instant data
coding
2
7
3.No control by interviewer
Interviewer administered Questionnaire
Advantages:
Disadvantages:
• Interviewer bias
2
8
Participation of illiterate
people
• Needs more resources
Clarification
• Only short questionnaires
possible especially on
telephone
Quick answers
• Difficult for sensitive issues
Designing Questionnaire
29
Types of questions ?
1 Open-ended question
2 Closed-ended question
3 Contingency questions.
30
1- Closed questions
. You ask the respondent to choose, among a possible number of answers,
the response that most closely represents his/her viewpoint
Advantages:
•The respondent is restricted to a certain set of responses,
•They are easy to answer ( no hesitations),
•Less time consuming,
•Easy to code
Disadvantages :
•You will force the respondent to choose between specific answers … BIAS
•You offer answers that otherwise would not have come to his mind,
•They don’t allow for creativity and generating ideas,
•They do not permit the respondent to qualify the chosen response or express a more complex or
subtle meaning,
•There is may a tendency for the respondent to tick systematically either the first or last answer,
•To answer all items in a list in the same way
•May Select what may be considered as the most socially desirable response,
31
Open ended questions
Free-response questions are not followed by any choices and the
respondent must answer by supplying a response, usually by entering a
number, a word, or a short text.
Advantages:
They allow respondents to express their ideas spontaneously in their own language.
They are less likely to guide the answer than closed questions .
They can add new information when there is very little existing information available about a topic.
Disadvantages :
They may be difficult to answer
They require effort and time from the respondent,
They require the development of a system of coded categories with which to classify the
responses..!
Can’t be coded !
They require the respondent to have some degree of writing ability
32
3- Contingency questions
Is a special case of a closed-ended question because it
applies only to a subgroup of respondents…!
33
Types of closed questions
?
Scales ?
34
Do’s and Don'ts in framing
a Questionnaire
Wording of the question
 Logic flow of the questions
 Simple language
 Mother language /tongue is the best
 Use concise and clear words
 And Avoid …..
35
Don'ts in framing a Questionnaire
Double-barreled questions
Single questions that ask for two things and therefore require two
answers.
1.
Do you have your own table or your own room to do your
homework?
2.
Do you think it is good idea for children to study geography
and history in primary school?
36
2- Double negatives
Either in questions or answers
1.All the following aren’t used in the TTT of … Except
? :D
37
3- Avoid overlapping response categories
Under 20
1
20-30
2
30-40
40-50
50-60
3
How old are
you?
4
5
QUESTIONNAIRE
Under
20
21-30
2
31-40
3
41-50
4
51-60
5
38
1
4- Leading questions
Would you say that you are not in favour of school
on Saturday morning??
39
5- Overload the respondent’s memory
Would you say that you are not in favour of school
on Saturday morning??
40
6- Long questions
Generally, it is recommended to hold
questions to 25 words or less.
If a longer sentence is used then it
should be broken up so that there will be
several shorter sentences.
41
Avoid
Hypothetical questions
1. Prediction of behavior and future
2. “Would you use this resource in your
class if it were available..?”
3. People are generally poor predictors.
42
Testing
44
1. Informal tests
2. Experts review
3. Small-scale tests (pre test)
4.Functional or technical testing
5.Large-scale pilots and trials
Informal Testing
45
Informal testing, also called pretesting, is a fundamental step in the process
of developing a questionnaire. Because it is a relatively easy and
inexpensive technique, it is used frequently to detect and correct problems in
a proposed questionnaire.
Testing
47
2. Experts review
Pre-testing the questionnaire
Pre-pilot
Focusing group
46
Pre-testing the questionnaire
a.Is each item producing the kind of information needed?
b.What role is the item going to play in the proposed analysis?
c.Are the questions meaningful to the respondents?
d.Are respondents easily able to understand the items?
e.Can respondents use the response format for each item?
f.Was the question order logical and did the interview flow smoothly?
h. Did some parts of the questionnaire arouse suspicion?
I. Did other parts of the questionnaire seem repetitive or boring?
j.Were interviewers able to read the questions without difficulty?
k.Were respondents able to follow all instructions?
l.Was the questionnaire too long?
47
Character of this group:
 Select a sample similar in socio-economic
background and geographic location to the one
that will be used in the main study.
 This sample will not be included in the final survey.
 Not to be a convenience sample.
48
Reliability and validity
Validity: Concerns the degree to which a
question measures what it was intended to
measure and not something else.
•Content (or face) validity
•Empirical (or predictive) validity
•Concurrent validity
49
Reliability: concerns the consistency of a
measure. That is, the tendency to obtain the same
results if the measure was to be repeated by using the
same subjects under the same conditions.
Kappa coefficient !
50
Covering letters and introductory
paragraphs
 You must explain the objectives of the
survey,
 In an interview, one of the tasks of the
interviewer is to persuade the respondent
to co-operate.
 In a self-administered questionnaire, the
covering letter is the only instrument for
overcoming resistance.
61
Covering letter Items
 Identify the organization conducting the study.
 Explain the purpose of the study.
 Assure the respondent that information provided will
be managed in a strictly confidential manner.
QUESTIONNAIRE
52
Covering letter Items
 Explain WHY it is important that the respondent
should complete the questionnaire.
 Brief detail on how the respondent was selected
(for example, ‘Your name was randomly selected ....’).
 Expression of appreciation for the respondent’s help.
53
Covering letter Items
Estimate of questionnaire completion time.
Provide the name and contact numbers of
the Principal Researcher(PI).
54
Check the Validity and Reliability of Questionnaires
65
•Types of Questionnaires
• Structured Questionnaires
• Unstructured Questionnaires
•Qualities of a good Questionnaire
• Questionnaire Reliability
• Test-retest reliability
• Inter-rater reliability
• Parallel form reliability
• Split-half reliability
• Questionnaire Validity
• Establish face validity
• Conduct a pilot test
• Enter the pilot test in a spreadsheet
• Use principal component analysis (PCA)
• Check the internal consistency of questions loading onto the
same factors
• Revise the questionnaire based on information received
Contact Me
THANK YOU
UNIT 2
MEASUREMENT, SCALING AND SAMPLING TECHNIQUES AND
RESEARCH REPORT PRESENTATION
Validity and Reliability-Definition, importance, types of validity, types
of reliability-- Construction and Validation of questionnaire,
Cronbach alpha test, Measurement – definition- significance – types
Nominal, Ordinal, Interval and Ratio, Scaling techniques. Sampling
methods- Probability sampling methods and Non - Probability
sampling methods, Report writing – importance , guideline to write
an academic report, Basics of report presentation- Content of an
Academic Research report, Content on a Research Article, Steps to
publish an article, Research Metrics: Significance of Journal Impact
Factor, SNIP, SJR, , Cite Score, Metrics: h-index, g index, i10 index
Intellectual property
Download