RESEARCH DESIGNS FOR
QUANTITATIVE STUDIES
What is a research design?
 A researcher’s overall plan for obtaining
answers to the research questions or for
testing the research hypotheses is referred
to as the research design.
Aspects of research design
 Intervention
 Comparison
 Controls of extraneous variables
 Timing of data collection
 Research sites and settings
 Communication with the study participants
The following questions would be addressed
What exactly is the intervention?
What are the procedures to be used for
both groups?
Who will receive the intervention and who
will not? How will each group be selected?
What is the dosage of the intervention?
INTERVENTION
INTERVENTION – Cont…
How long will the intervention be
administered? And when will it begin?
Who will administer the intervention?
Will those administering the
intervention are fully informed about
the study? Will study participants be
fully informed?
Under what conditions will the
intervention be withdrawn or altered?
COMPARISONS
 In a quantitative study, the researcher specifies the
nature of the comparisons in advance, where as in a
qualitative study, comparisons may suggest
themselves during the course of data collection.
TYPES OF COMPARISONS IN RESEARCH STUDIES
 Comparison between two or more groups
 Comparison of a single group at two or more points in
time
 Comparison of a single group under different
circumferences or experience
 Comparison based on relative rankings
 Comparison with samples from other studies
CONTROLS FOR EXTRANEOUS VARIABLES
The researcher must make decisions
about which extraneous variables to
control.
TIMING OF DATA COLLECTION
RESEARCH SITES & SETTINGS
COMMUNICATION WITH THE STUDY
PARTICIPANTS
Overview of Research Design Types
 Structured versus flexible designs Eg. Quantitative study and
qualitative study
 Between –subjects and within –subjects designs Eg. Testing
the efficacy of a drug
1. One group administered
one group do not administered Between subjects design
2. Same group before, after
treatment Within subjects design
 The time dimension
There are four situations in which it might be appropriate to
design a study with multiple points of data collection.
1. Studying time related processes - Eg. Healing, earning,
growth etc
2. Determining time sequences - Eg. Infertility results
in depression
3. Developing comparisons - Eg. Smoking and
lung
4. Enhancing research control - Eg. Two groups with
alternative
interventions
Categorization of studies
• Cross sectional design
• Longitudinal design
Longitudinal design
Types
– Trend studies
– Cohort studies
– Cross-sequential design (Control studies with a
cross –sectional approach)
– Panel studies
– Follow-up studies
Retrospective versus prospective
designs
Characteristics of good design
Appropriateness to the research question
Lack of bias
Precision
Power
Tips on designing research
Make a written list of the pros and cons of each
Balance a number of considerations, such as time, cost,
ethical issues and the integrity of the study.
Anticipate alternative findings and consider whether design
adjustments might affect the results.
Seek the advice of research experts in deciding the design
Write out a rationale for your choices one you have made
your design decisions.
Experimental Research designs for quantitative studies
Characteristics of true experiments
Manipulation – Involves doing something to at least
one group of subjects
Control – Refers to a group of subjects without
any manipulation
Randomization – involves the Placement of subjects
in groups on a random basis
Research designs
Experimental designs
Quasi experimental designs
Non experimental designs
Types of experimental designs
 Basic experimental design
 Solomon four group design
 Factorial design
 Repeated measures design
 Randomized clinical trials
Basic experimental design
Types
 After only design or post test –only design (R-X-O)
 Before after design or pretest –posttest design
(R-O1 X O2)
(R-O1 – O2)
Solomon four group design
Group Date collection
Before After
Experimental –with
pretest
X X
Experimental –
without pretest
- X
Control-with pretest X X
Control-without
pretest
- X
This design prevents, the post test measures
getting affected not only by the treatment but
also by exposure to the pre-test
Factorial design
Types of stimulation Auditing tactile
A1 A2
A1 B1 A2 B1
A1 B2 A2 B2
A1 B3 A2 B3
15min B1 This design permits the testing
of multiple hypotheses in a
single experimental
Daily exposures
45m
B3
30min B2
Repeated measures design
Randomized clinical trials
 Advantages and disadvantages of experimental studies.
Advantage – Confidence with which causal
relationships can be inferred
– Most powerful method available to
scientists
 Disadvantage – Hawthorne effect limitation
– Number of like characteristics of human
beings, finance etc
QUASI, EXPERIMENTAL RESEARCH
Quasi-Experimental designs
Non equivalent control group designs
Types
1.Non equivalent control group pre-test post –test
design
(Quasi experimental)
O1 X O2
O1 - O2
2. Non equivalent control group post test only design
(Pre experimental)
X O
O
Time series designs
Types
1. Time series design (Quasi –experimental)
O1 O2 O3 O4 X O5 O6 O7 O8
2. Time series non-equivalent control group design (quasi-
experimental)
O1 O2 O3 O4 X O5 O6 O7 O8
O1 O2 O3 O4 - O5 O6 O7 O8
3. Time series with intensified treatment (quasi-experimental)
O1 O2 X O3 O4 X+1 O5 O6 X+2 O7 O8
4. Time series with withdrawn and reinstituted treatment
(quasi-experimental)
O1 O2 X O3 O4 (–X) O5 O6 X O7 O8
Advantages and disadvantages of the
quasi-experimental research
Advantages Feasibility
Generalizability some
extent
Practicality
Disadvantage Absent of control
Non Experimental Research
Reasons for under taking non-experimental
research in nursing
o Human characteristics are inherently not subject
to experimental manipulation (eg. blood type
personality etc) and the effects of these
characteristics on some phenomenon of interest
cannot be studied experimentally.
o For ethical reasons.
o Not practical to conduct a true experiment
o Experimental design is not appropriate to some
research questions
Types of Non Experimental
Research
 Ex post facto or correlational research
 Descriptive research
1. Ex post facto design on correlational
research
Schematic diagram
Ex post facto design
Group A O
Group B O
Descriptive research
Purpose
- To observe, describe and document
aspects of situation as it naturally
occurs
- To serve as a starting point of for
hypothesis generation or theory
development.
Types
• Descriptive correlational study – No
control over the independent
variables
• Univariate descriptive studies – Not
focused on only one variable but to
describe the existing factors in terms
of frequency of occurrence
Advantages:
 An efficient and effective means of
collecting a large amount of data
about a problem area
 Short time only
 No biophysiologic or other measures
are needed
 Has an intrinsic appeal for the
solution of many practical problems.
Disadvantages:
Weak in its ability to reveal causal
relationships
Susceptible to the possibility of faulty
interpretation
Preexisting differences may be a
plausible alternative explanation for
any observed differences on the
dependent variable of interest.
SURVEY RESEARCH
EVALUATION RESEARCH
NEEDS ASSESSMENT
 OUT COMES RESEARCH
 SECONDARY ANALYSIS
 META –ANALYSIS
 DELPHI SURVEYS
 METHODOLOGIC RESEARCH
 METHODOLOGIC RESEARCH
 CONTENT ANALYSIS STUDIES
ADDITIONAL TYPES OF
QUANTITATIVE RESEARCH
research-designs-for-quantitative-studies-nursing-research-ppt-1.pptx

research-designs-for-quantitative-studies-nursing-research-ppt-1.pptx

  • 1.
  • 2.
    What is aresearch design?  A researcher’s overall plan for obtaining answers to the research questions or for testing the research hypotheses is referred to as the research design. Aspects of research design  Intervention  Comparison  Controls of extraneous variables  Timing of data collection  Research sites and settings  Communication with the study participants
  • 3.
    The following questionswould be addressed What exactly is the intervention? What are the procedures to be used for both groups? Who will receive the intervention and who will not? How will each group be selected? What is the dosage of the intervention? INTERVENTION
  • 4.
    INTERVENTION – Cont… Howlong will the intervention be administered? And when will it begin? Who will administer the intervention? Will those administering the intervention are fully informed about the study? Will study participants be fully informed? Under what conditions will the intervention be withdrawn or altered?
  • 5.
    COMPARISONS  In aquantitative study, the researcher specifies the nature of the comparisons in advance, where as in a qualitative study, comparisons may suggest themselves during the course of data collection. TYPES OF COMPARISONS IN RESEARCH STUDIES  Comparison between two or more groups  Comparison of a single group at two or more points in time  Comparison of a single group under different circumferences or experience  Comparison based on relative rankings  Comparison with samples from other studies
  • 6.
    CONTROLS FOR EXTRANEOUSVARIABLES The researcher must make decisions about which extraneous variables to control. TIMING OF DATA COLLECTION RESEARCH SITES & SETTINGS COMMUNICATION WITH THE STUDY PARTICIPANTS
  • 7.
    Overview of ResearchDesign Types  Structured versus flexible designs Eg. Quantitative study and qualitative study  Between –subjects and within –subjects designs Eg. Testing the efficacy of a drug 1. One group administered one group do not administered Between subjects design 2. Same group before, after treatment Within subjects design  The time dimension There are four situations in which it might be appropriate to design a study with multiple points of data collection.
  • 8.
    1. Studying timerelated processes - Eg. Healing, earning, growth etc 2. Determining time sequences - Eg. Infertility results in depression 3. Developing comparisons - Eg. Smoking and lung 4. Enhancing research control - Eg. Two groups with alternative interventions Categorization of studies • Cross sectional design • Longitudinal design
  • 9.
    Longitudinal design Types – Trendstudies – Cohort studies – Cross-sequential design (Control studies with a cross –sectional approach) – Panel studies – Follow-up studies Retrospective versus prospective designs
  • 10.
    Characteristics of gooddesign Appropriateness to the research question Lack of bias Precision Power Tips on designing research Make a written list of the pros and cons of each Balance a number of considerations, such as time, cost, ethical issues and the integrity of the study. Anticipate alternative findings and consider whether design adjustments might affect the results. Seek the advice of research experts in deciding the design Write out a rationale for your choices one you have made your design decisions.
  • 11.
    Experimental Research designsfor quantitative studies Characteristics of true experiments Manipulation – Involves doing something to at least one group of subjects Control – Refers to a group of subjects without any manipulation Randomization – involves the Placement of subjects in groups on a random basis Research designs Experimental designs Quasi experimental designs Non experimental designs
  • 12.
    Types of experimentaldesigns  Basic experimental design  Solomon four group design  Factorial design  Repeated measures design  Randomized clinical trials Basic experimental design Types  After only design or post test –only design (R-X-O)  Before after design or pretest –posttest design (R-O1 X O2) (R-O1 – O2)
  • 13.
    Solomon four groupdesign Group Date collection Before After Experimental –with pretest X X Experimental – without pretest - X Control-with pretest X X Control-without pretest - X This design prevents, the post test measures getting affected not only by the treatment but also by exposure to the pre-test
  • 14.
    Factorial design Types ofstimulation Auditing tactile A1 A2 A1 B1 A2 B1 A1 B2 A2 B2 A1 B3 A2 B3 15min B1 This design permits the testing of multiple hypotheses in a single experimental Daily exposures 45m B3 30min B2
  • 15.
    Repeated measures design Randomizedclinical trials  Advantages and disadvantages of experimental studies. Advantage – Confidence with which causal relationships can be inferred – Most powerful method available to scientists  Disadvantage – Hawthorne effect limitation – Number of like characteristics of human beings, finance etc
  • 16.
    QUASI, EXPERIMENTAL RESEARCH Quasi-Experimentaldesigns Non equivalent control group designs Types 1.Non equivalent control group pre-test post –test design (Quasi experimental) O1 X O2 O1 - O2 2. Non equivalent control group post test only design (Pre experimental) X O O
  • 17.
    Time series designs Types 1.Time series design (Quasi –experimental) O1 O2 O3 O4 X O5 O6 O7 O8 2. Time series non-equivalent control group design (quasi- experimental) O1 O2 O3 O4 X O5 O6 O7 O8 O1 O2 O3 O4 - O5 O6 O7 O8 3. Time series with intensified treatment (quasi-experimental) O1 O2 X O3 O4 X+1 O5 O6 X+2 O7 O8 4. Time series with withdrawn and reinstituted treatment (quasi-experimental) O1 O2 X O3 O4 (–X) O5 O6 X O7 O8
  • 18.
    Advantages and disadvantagesof the quasi-experimental research Advantages Feasibility Generalizability some extent Practicality Disadvantage Absent of control
  • 19.
    Non Experimental Research Reasonsfor under taking non-experimental research in nursing o Human characteristics are inherently not subject to experimental manipulation (eg. blood type personality etc) and the effects of these characteristics on some phenomenon of interest cannot be studied experimentally. o For ethical reasons. o Not practical to conduct a true experiment o Experimental design is not appropriate to some research questions
  • 20.
    Types of NonExperimental Research  Ex post facto or correlational research  Descriptive research 1. Ex post facto design on correlational research Schematic diagram Ex post facto design Group A O Group B O
  • 21.
    Descriptive research Purpose - Toobserve, describe and document aspects of situation as it naturally occurs - To serve as a starting point of for hypothesis generation or theory development.
  • 22.
    Types • Descriptive correlationalstudy – No control over the independent variables • Univariate descriptive studies – Not focused on only one variable but to describe the existing factors in terms of frequency of occurrence
  • 23.
    Advantages:  An efficientand effective means of collecting a large amount of data about a problem area  Short time only  No biophysiologic or other measures are needed  Has an intrinsic appeal for the solution of many practical problems.
  • 24.
    Disadvantages: Weak in itsability to reveal causal relationships Susceptible to the possibility of faulty interpretation Preexisting differences may be a plausible alternative explanation for any observed differences on the dependent variable of interest.
  • 25.
    SURVEY RESEARCH EVALUATION RESEARCH NEEDSASSESSMENT  OUT COMES RESEARCH  SECONDARY ANALYSIS  META –ANALYSIS  DELPHI SURVEYS  METHODOLOGIC RESEARCH  METHODOLOGIC RESEARCH  CONTENT ANALYSIS STUDIES ADDITIONAL TYPES OF QUANTITATIVE RESEARCH