Transcript Document

Dr. S. Nishan Silva
(MBBS)
Stages in the Research Process
Define
Problem
Planning a
Research Design
Conclusions
and Report
Planning
a Sample
Processing and
Analysing the Data
Gathering
the Data
Research Process
Phase I
Preparation Phase
Phase II
Implementation phase
Phase III
Outcome Phase
1. Select a problem for
research
7. Collect the data
9. Interpret research
findings.
2. Literature review
8. Analyse data
10. Report the study
3. Formulate research
question
4. Select research
approach and design
5. Select data collection
method.
6. Specify a population
Research Methods - Timeframe
Research Project
Day 1
Develop Research Proposal
and obtain approval
Develop and test questions
Develop and test tool
Obtain participants
Administer instrument(s)
Ongoing data collection and analysis
Final collection of data
Research Report
Day 344
1. Problem Identification and statement
of research problems
Sources to identify problems

Nursing experience



Nursing Literature


Personal, of collegues
Of hospital records
Nursing journals, books
Theory

Text books
2. Process of Selecting a Research
Problem

The topic is RESEARCHABLE

The question is “THEORY BASED”

The research is a feasible project

The researcher has the ability to
carry out the study
3. Writing a problem statement



Convert the topic in to a
“STATEMENT”
Question, statement or hypothesis
A hypothesis is a statement of
predicted relationship or difference
between two or more variables
A good research statement
should have,

Area of focus

Population

Research design

Setting of the study
Different levels of research
problems

Level 1
 One
variable
 One Population

Level 2
 Two
variables
 Relationship between them

Level 3
 Experimental
type designs, finding causes
 Manipulation of one variable to find its effect
on the other
4. Define Variables
What is a variable?
A characteristic, property or attribute of the
person or thing under investigation.
Types of variables

Dependant Variable



Independent Variable



Other variable (outside the research) that can interrupt
May be controlled – increase accuracy
Discrete variable


The variable that is manipulated
Extraneous variable


It is the researched, observed variable
It changes according to the manipulating variable
A variable that it finite ; a whole number – ex- days, patients
Continuous variable


That is infinite
Spans a range
Defining variables

Conceptual definition
Defining as it is understood
 Ex- Social class


Operational definition
Working definition
 Definition to be used in the research
 Ex – Father’s usual occupation as stated by
the Mother

5. Literature review
 Definition
It is a critical summery of
available theoretical and
research literature on
the selected research
topic.
Finding Literature

Library catalogues – manual and
electronic

Indexes and abstracts –
 Ex
– MEDLINE, CINAHL, PubMED
 e-Medicine
Guidelines in doing the Review
1. Search for existing literature in the
library and on the web;
2. Prepare a working bibliography. Record all vital details
concerning the books or research you are including in
your bibliography
• Write in index cards; group together references from
a. books
b. journals and periodicals
c. unpublished material
3. Examine each material, then decide which ones will
actually be included in your review
6. Population and sample

Population – define as much as possible



Time bound?
Geographically bound?
Process of selecting a sample – sampling


Bias?
Therefore
 Probability sampling
 Non probability sampling
1.
2.
3.
4.
5.
Define the Population of Interest
Identify a Sampling Frame (if possible)
Select a Sampling Method
Determine Sample Size
Execute the Sampling Plan
 Population
of interest is entirely dependent
on Management Problem, Research
Problems, and Research Design.
 Some Bases for Defining Population:
 Geographic
Area
 Demographics
 Usage/Lifestyle
 Awareness
Probability Sampling

Simple Random Sampling
Everyone has a chance of getting included
 Random numbers table


Stratified Random Sampling
Population divided in to strata – segments
 Then do simple random sampling for each
strata


Systematic Sampling

Using every -----th person.
Random numbers table
Non-probability Sampling

Convenience sampling


Purposive sampling



Judgemental sampling
Selects groups according to criteria
Quota sampling



Also called accidental. As you meet them.
Quotas from pre-decided characteristic groups
Convenience sampling within a group
Cluster sampling


Multistage sampling
Larger clusters and smaller clusters within
Multistage Sampling

Stage 1


randomly sample clusters (schools)
Stage 2

randomly sample individuals from the schools
selected
 How
many completed questionnaires do
we need to have a representative sample?
 Generally the larger the better, but that
takes more time and money.
 Answer depends on:
 How
different or dispersed the population is.
 Desired level of confidence.
 Desired degree of accuracy.
Other factors

Inclusion criteria?
 Who
gets in?
 How to filter?

Exclusion criteria?
 Who
stays out?
 How to determine?
7. Research Design
 Quantitative
Research
 Experimental
Designs
 Non-experimental Designs
 Descriptive
Design
 Exploratory Design
 Co-relational Design
 Retrospective Design
 Quasi-experimental
 Qualitative
Designs
Research
Experimental Designs

Researcher manipulates variable/s

The design uses control groups

The selection of sample is based on
random sampling
Non-Experimental Designs

Descriptive Design


Exploratory Designs


To find out relationship between dependant variable and
independent variable without any manipulations . (Observe
as it is)
Co-Relational research Design


Description of a data collection on several variables
Relationship between two variables in the same sample
Retrospective Design

Collect data on variables after they have happened (looking
back at the past).
Cross-Sectional Versus Longitudinal
Studies

Cross-Sectional Studies
A study can be done in which data are gathered just
once, perhaps over a period of days or weeks or months,
in order to answer a research question.
Longitudinal Studies
Studying people or phenomena at more than one point in
time in order to answer the research question.

Because data are gathered at two different points in
time, the study is not cross-sectional kind, but is carried
longitudinally across a period of time.

Quasi-Experimental Research
 Researcher
does manipulate the
independent variable
 But unable to randomly allocate
 Uses a convenient sampling method
to form the sample.
Qualitative Research
What? Why? How?
 Data – words / pictures etc
 The unfolding process determines the next
step
 Researcher is the key instrument of data
collection
 Open ended questionaires / Interviews /
Video recordings / Observations

Qualitative - Definition

… qualitative researchers study things in
their natural settings, attempting to make
sense of or interpret phenomenon in terms
of the meanings people bring to them.
(Denzin & Lincoln, 2000, p.3).
Qualitative Research Designs
Descriptive / Exploratory Design
 Interpretative Design

 Ethnography


Anthropological
Methods – Interviews/ Observations / Records / Life
history facts / news reports / Diaries
 Phenomenology


“What it is like to have a certain experience?”
Ask peoples real life experiences / use novels/ films
 Ground


Theory
Researcher formulates tentative theories – using
inductive reasoning
Follows up those ideas with further enquiry – deductive
reasoning
Data Collection methods
 Types
of data to be collected
Quantitative
Data
that is collected as numbers
Qualitative
As
Data
data
words, pictures, documents,
Photos
Data Collection Methods and
Instruments
Bio-Physical measurements
 Observations

 Questionnaires


Advantages?
Disadvantages?
 Interviews




Unstructured – Open ended questions
Structured – Close ended questions
Interview schedules and interview guides
Advantages ? Disadvantages?
Issues in research instrument

Suitable for use


Based on theory frame of the study




Accurate
Protocol
Simple directions for users


Should test the theory / not too much of other info.
Collect adequate info
Valid / Reliable / Un-biased


Language / Culture
Uncomplicated
Easy to administer

Not taking too much time / effort
Pilot Project
Smaller version
 Test the instrument (questionnaire)
 Small sample from the same or similar
population
 Sort out problems

 Understandability
 Validity
 Accuracy
Reliability and Validity

Reliability

Basic sources of inaccuracy



Deficiency (error)
Inconsistency between readings
Methods to test reliability

Test-retest method


Equivalent Test


Same test twice with a rest in between
Two tests given to two samples with different behaviors
Split half method

Separate scores for even numbered and odd numbered items
analyzed.
Reliability and Validity

Validity

Types of Validity

Predictive validity


Content validity


Ability to differentiate people based on a criterian
Construct validity


Adequacy of coverage
Concurrent validity


Ability of the instrument to predict future behavior
Whether the theory is measured or something else is?
Face validity

Whether it appears to be valid
Data Analysis

How to process the collection (?papers)

Master data sheets

Coding

Master tables

Statistics
Flowcharting the Research Process (2)
Survey (Interview, Questionnaire)
Experiment (Laboratory, Field)
Secondary Data Study
Observation
Collection of Data (Fieldwork)
Editing and Coding Data
Sample Design
Probability
Sampling
Non-Probability
Sampling
Data Processing and Analysis
Interpretation of Findings
Report
Homework
Read Research Statistics
Chapter
The MOST IMPORTANT TIME for the statistics
to
be involved with a research study is in the
very BEGINNING
STATISTICS CAN HELP OBTAIN THE MAXIMUM
AMOUNT INFORMATON FROM AVAILABLE
RESOURCES
HOW???
HELP WITH THE DESIGN OF THE EXPERIMENT
DETERMINE SAMPLE SIZE NEEDED
DEVELOP PROCESS OF COLLECTING DATA
DISCUSS VARIABLES TO BE MEASURED AND
HOW THEY RELATE TO THE OBJECTIVES OF
THE STUDY
PROVIDE METHODS OF ANALYZING THE DATA
HELP TRANSLATE STATISTICAL CONCLUSIONS
INTO SUBJECT MATTER CONCLUSIONS
Why Use Statistics?

Descriptive Statistics
•
•

identify patterns
leads to hypothesis generating
Inferential Statistics
•
•
distinguish true differences from
random variation
allows hypothesis testing
Describing the Data
with Numbers
Measures of Central Tendency
•
•
•
MEAN -- average
MEDIAN -- middle value
MODE -- most frequently observed
value(s)
Describing the Data
with Numbers
Measures of Central Tendency
•
•
•
MEAN -- average
MEDIAN -- middle value
MODE -- most frequently observed
value(s)
Histogram-Frequency Distribution
Charts
Number of Plants in each Class
35
30
25
20
Number of plants in each
class
15
10
5
0
0.0-0.9 1.0-1.9 2.0-2.9 3.0-3.9 4.0-4.9 5.0-5.9 6.0-6.9
This is called a “normal” curve or a bell curve
This is an “idealized” curve and is theoretical based on an infinite number
derived from a sample
The Normal Curve and Standard
A normal curve:
Deviation
Each vertical line
is a unit of
standard deviation
68% of values fall
within +1 or -1 of
the mean
95% of values fall
within +2 & -2
units
Nearly all
members (>99%)
fall within 3 std
dev units
Terms
confidence interval:
The range of values we can be reasonably
certain includes the true value.
95% Confidence Intervals
Khaja
(n=40)
Anderson
(n=50)
Kennedy
(n=250)
-.40 -.35 -.30 -.25 -.20 -.15 -.10 -.05 .00
.05
.10
.15
.20