Experimental &Quasi Experimental Design

download Experimental &Quasi Experimental Design

of 70

Transcript of Experimental &Quasi Experimental Design

  • 8/2/2019 Experimental &Quasi Experimental Design

    1/70

    Dr.(Mrs.) A.V.RamanDean

    Omayal Achi College of Nursing

    Avadi, Chennai

    EXPERIMENTAL DESIGN

  • 8/2/2019 Experimental &Quasi Experimental Design

    2/70

    INTRODUCTION TO RESEARCH DESIGNS

    Quantitative research design

    Research design of a study spells out the

    basic strategies that researchers adopt to

    develop evidence that is accurate andinterpretable

    The research design incorporates some of

    the most important methodologicaldecisions that the researchers make

    particularly in quantitative studies

  • 8/2/2019 Experimental &Quasi Experimental Design

    3/70

    ASPECTS OF QUANTITATIVE

    RESEARCH

    Intervention

    Comparisons

    Control of extraneous variables

    Timing of data collection

    Research sites and settings

    Communication with the subjects

  • 8/2/2019 Experimental &Quasi Experimental Design

    4/70

    INTERVENTION

    In some studies nurse researchers want to test the

    effect of specific intervention on dependent

    variable (for e.g.. the effect of specific

    intervention on labor and neonatal outcome)

    COMPARISONThe most common types of comparisons are

    1. Comparisons between two or more groups (The

    hypothesis that the drug tamoxefin reduced the

    rate of breast cancer in high risk women could be

    tested by comparing women who received

    tamoxefin and those who did not)

  • 8/2/2019 Experimental &Quasi Experimental Design

    5/70

    2.In certain instances it is desirable to make

    comparisons for the same study participants

    (e.g. studying patients heart rate before and afterintervention,

    Comparing low back pain for patients lying in

    two different positions)

    CONTROLS FOR EXTRANEOUS VARIABLES

    The complexity of relationships among human

    characteristics often makes it difficult to answer

    research questions unless efforts are made toisolate key research variables and control other

    factors extraneous to the research questions

  • 8/2/2019 Experimental &Quasi Experimental Design

    6/70

    TIMING FOR DATA COLLECTION

    In most studies data are collected fromsubjects at a single point of time (for e.g.the subjects might be asked on a singleoccasion to describe the health promotionbehavior)

    In some designs we contact the subjectsin multiple time (e.g. to observe the growthpattern 3 monthly) Hence the researcher

    must decide on the number of datacollection points needed to address theresearch question properly

  • 8/2/2019 Experimental &Quasi Experimental Design

    7/70

    RESEARCH SITES AND SETTINGS

    sites are the overall location forresearch

    Settings are the more specific placewhere data collection will occur

    sites and settings to be selected so as tomaximize the validity and reliability ofth data in designing a study. The setting

    should be natural and avoid anxietypromoting or foreign to the subjectsexperiences

  • 8/2/2019 Experimental &Quasi Experimental Design

    8/70

    COMMUNICATION WITH THESUBJECTS

    Decide how much information to begiven to the participants

    Full disclosure of the study?

    Consent Oral or written?

    Think about the reading andcomprehension level of the subjects

    Who will provide the information?

    What additional questions from thesubjects are expected?

    Debriefing sessions?

  • 8/2/2019 Experimental &Quasi Experimental Design

    9/70

    EXPERIMENTAL DESIGN

    Science answers questions with experiments

  • 8/2/2019 Experimental &Quasi Experimental Design

    10/70

    EXPERIMENTAL DESIGN

    An experiment is a scientificinvestigation in whichobservations are made and dataare collected according to a setof well defined criteria. In anexperiment, researchers areactive agents not passiveobservers.

  • 8/2/2019 Experimental &Quasi Experimental Design

    11/70

    The procedures developedby physical scientists were

    profitably adopted by

    biologists during the 19thcentury, resulting in many

    achievements in physiologyand medicine.

  • 8/2/2019 Experimental &Quasi Experimental Design

    12/70

    DEFINITIONExperimental method is an

    empirical research method used

    to examine a hypothesizedcausal relationship betweenindependent and dependent

    variables.

  • 8/2/2019 Experimental &Quasi Experimental Design

    13/70

    Begin by asking a question aboutyour topic

    One that is testable with thematerials at hand

    What is a good question for an

    experiment?

    DEFINE THE PROBLEM

  • 8/2/2019 Experimental &Quasi Experimental Design

    14/70

    Now we need a hypothesis toguide our investigation.

    What is a hypothesis?

    Your best thinking about how the change

    you make might affect another factor.Tentative or trial solution to the question.

    An if then statement.

  • 8/2/2019 Experimental &Quasi Experimental Design

    15/70

    CHARACTERISTICS OFEXPERIMENTAL DESIGN

    A true experimental

    design is characterizedby Manipulation,

    Control , andrandomization

  • 8/2/2019 Experimental &Quasi Experimental Design

    16/70

    Refers to the process by

    which the researchermanages the independentvariable in order to study the

    effect on the dependentvariable.

    MANIPULATION

  • 8/2/2019 Experimental &Quasi Experimental Design

    17/70

    EXAMPLE

    . Gentle massage is effective as a pain reliefmeasure for elderly nursing home residents

    The Independent variable is gentle massage

    which could be manipulated by giving somepatients the massage intervention andwithholding it for others

    Then it is possible to compare the pain level(dependent variable) in the two group to seeif differences in receipt of the interventionresulted in degree of pain levels

  • 8/2/2019 Experimental &Quasi Experimental Design

    18/70

    CONTROL

    The experiment must be organized

    so that the extraneous factors are

    prevented from operating and

    confusing the outcome which is

    to be appraised.

  • 8/2/2019 Experimental &Quasi Experimental Design

    19/70

    Control is acquired by

    manipulating by randomizing,by careful preparation of the

    experimental protocols, andby the use of a comparisongroup or groups.

  • 8/2/2019 Experimental &Quasi Experimental Design

    20/70

    EXAMPLE

    If a researcher wanted to supplement diet of

    the low birth baby with a particular nutrient

    for 2 weeks, the weight gain at the end of 2

    weeks would tell us nothing about the

    treatment effectiveness

    Suppose an average 1 kg. weight gain is

    noted does this gain support the conclusion

    that nutrition supplement (IV) caused the

    weight gain (DV)

  • 8/2/2019 Experimental &Quasi Experimental Design

    21/70

    The answer will be NO? because babiesnormally gain weight as they mature

    Without a control group- a group that does notreceive the nutritional supplements it isimpossible to separate the effects ofmaturation from those treatment

    Therefore the term CONTROL GROUPrefers to a group of subjects whose

    performance on a dependent variable is usedto evaluate the performance of theexperimental group on the same dependentvariable

  • 8/2/2019 Experimental &Quasi Experimental Design

    22/70

    RANDOMISATIONThe term random essentially means that every

    subject has an equal chance of being assigned

    to any group. Randomization is the process

    that first ensures every unit in the targetpopulation has an equal chance of being

    chosen for the study sample, and then

    ensures that each unit in the study sample

    has on equal chance of being assigned to

    either the experimental or the control group.

  • 8/2/2019 Experimental &Quasi Experimental Design

    23/70

    EXAMPLE

    The effectiveness of a contraceptive healtheducation programme for multiparous

    women. Two groups of subjects wereincluded .One group was givenintervention and other was not. Thewomen in the sample are likely to differ

    in age, education, and attitude Thiswould affect the woman's diligence inpracticing contraception

  • 8/2/2019 Experimental &Quasi Experimental Design

    24/70

    Although randomization is a preferredscientific method for equalizing the

    groups, there is no guarantee that thegroup will in fact be equal Therefore weneed a procedure to consciously controlof those characteristics of subjects that

    are likely to affect the outcome is calledMATCHING (e.g. if matching were used incontraceptive education, then theresearcher might ensure that if there wasa married 38 year old woman with sixchildren in exp. Group, then there will alsobe 38 year old woman with six children incontrol group

  • 8/2/2019 Experimental &Quasi Experimental Design

    25/70

    The antecedent event in a

    proposed causalsequence is called the

    independent variable.

  • 8/2/2019 Experimental &Quasi Experimental Design

    26/70

    The measured effect in

    the causal sequence iscalled the

    dependent variable.

  • 8/2/2019 Experimental &Quasi Experimental Design

    27/70

    Eight classes of extraneousvariablesmay especially interfere with research

    on human subjects, history,maturation, testing, instrumentation,

    statistical regression selection,mortality and interaction among these.1.History, the specific events occurring

    between the first and secondmeasurement in addition to the

    experimental variables

  • 8/2/2019 Experimental &Quasi Experimental Design

    28/70

    2. Maturation,process within the respondentsoperating as a function of the passage of time

    per se (not specific to the particular events),including growing older, growing hungrier,growing more tired and the like

    3. Testing,the effects of taking a test upon thescores of a second testing

    4.Instrumentation,in which changes in thecalibration of a measuring instrument or

    changes in the observers or scores used mayproduce changes in the obtainedmeasurements

  • 8/2/2019 Experimental &Quasi Experimental Design

    29/70

    5. Statistical regression, operating where

    groups have been selected on the basis of

    their extreme scores6. Biaser resulting in differential selection of

    respondents for the comparison groups ?

    7. Experimental mortality, or differential lossof respondents for the comparison groups

    8. Selection-maturation interaction etc.,

    which in certain of the multiple group

    quasi-experimental designs, might be

    mistaken for the effects of the

    experimental variable.

  • 8/2/2019 Experimental &Quasi Experimental Design

    30/70

    PRINCIPLE GUIDELINES OFEXPERIMENTS

    The MAX-MIN-CON. By Kerlinger 1986.

    The abbreviation stands for the following.MAX-Maximize experimental variable

    MIN- Minimize error variance.

    CON-Control extraneous variance.

  • 8/2/2019 Experimental &Quasi Experimental Design

    31/70

    MAJOR STEPS INEXPERIMENTAL DESIGN1. Delineate the population or universe to

    be studied (i.e. the set of subjects orobjects that share a common observable

    characteristic)

    2. Select a sample from the population by

    random sampling

    3. By random assignment, sub-divide thesample into two sub-samples

    4 Specify one sub sample the

  • 8/2/2019 Experimental &Quasi Experimental Design

    32/70

    4.. Specify one sub-sample, theexperimental group and other the controlgroup

    5. Before introducing the independentvariable, observe and record allimportant characteristics of the twogroups

    6. Introduce the independent variable intothe experimental group but withholds itfrom the control group

    7. After introducing the independentvariable, observe the dependent variablein both experimental and control group

    8 Compare the changes that occur in

  • 8/2/2019 Experimental &Quasi Experimental Design

    33/70

    8. Compare the changes that occur inthe experimental group with those

    that may have occurred in thecontrol group

    9. Record the difference

    10.Compare these values withstatistically computed values that

    judge the significance of the

    difference, and indicate whether ornot the observed differences couldhave occurred by chance.

  • 8/2/2019 Experimental &Quasi Experimental Design

    34/70

    TYPES OF EXPERIMENTALDESIGN

    The investigator planning an experiment

    has many experimental design option to

    choose. Experimental designs fall into

    three major categories.

    a)True or classical experimental design

    b)Quasi-experimental design

    c)Pre-experimental design.

  • 8/2/2019 Experimental &Quasi Experimental Design

    35/70

    TRUE OR CLASSICAL EXPERIMENTALDESIGN

    TRUE OR CLASSICAL EXPERIMENTAL DESIGN

    True or classical experimental designincludes six major designs. They are

    Pre-test___ post - test control groupdesign

    Solomon four group design After/ post - test __ only experimental

    design Factorial design Randomized block design Cross over design or repeated

    measures design.

  • 8/2/2019 Experimental &Quasi Experimental Design

    36/70

    The most commonly used in nursing studies, arediscussed as nomenclature used and definitionsdeveloped by Campbell and Stanley ( 1963).The standard notations that are helpful

    in under standing alternativeexperimental design are -

    x To denote the experimental

    manipulation. (Treatment orintervention)

    O To denote observations ormeasurement

    R To denote randomization

  • 8/2/2019 Experimental &Quasi Experimental Design

    37/70

    1.PRE-TEST POST-TESTCONTROL GROUP DESIGN

    Pretest

    Experimenta

    lTreatment

    Controlgroup

    Post

    testExperimental

    groupRandomassignme

    nt

    Pretest Posttest

  • 8/2/2019 Experimental &Quasi Experimental Design

    38/70

    EXAMPLE

    Effectiveness of antenatal

    nutrition education on

    pregnancy and labor outcome

    among primi para mothers

    attending out patient departmentof selected hospitals Madurai

    2007-2009

  • 8/2/2019 Experimental &Quasi Experimental Design

    39/70

    2. Solomon four group design is acomplex particularly useful in studies of

    developmental phenomena and permits theinvestigator to differentiate many effects.

    Pretest

    Experimental

    Treatment

    Controlgroup

    Experimental

    group

    Controlgroup

    Post testExperimentalgroup

    Randomassignment

    Pretest Post test

    ExperimentalTreatment

    Post test

    Post test

  • 8/2/2019 Experimental &Quasi Experimental Design

    40/70

    EXAMPLE

    Effectiveness of workshop on

    management of HIV to improve

    the nurses attitude toward

    patients with AIDS admitted in

    selected hospitals , Madurai2006-2008

  • 8/2/2019 Experimental &Quasi Experimental Design

    41/70

    3.Post- test only control groups designthis design, which is sometimes called after only controlgroup designs, is composed of two randomly assignedgroups, but neither of which is pre tested or premeasured in the before period of time.

    Control

    group

    Post

    -test

    Post

    -testExperimental

    group

    Experimental

    treatmentRandomassignment

  • 8/2/2019 Experimental &Quasi Experimental Design

    42/70

    EXAMPLE

    Effects of holding the newborns onpaternal bonding behavior

    The researcher randomly assigned 36

    first time fathers attendinguncomplicated deliveries of normalinfants to experimental and controlgroup (Those who held and did not

    hold their infants at delivery)12-36hours after the babies were bornbonding behavior frequencies were

    recorded and observed the outcome

  • 8/2/2019 Experimental &Quasi Experimental Design

    43/70

    4. Factorial design

    In this design testing of multiplehypotheses is done in a single experiment

    In factorial experiments subjects are assignedat random a specific combination of

    conditionse.g. Does auditory stimulations have a more

    beneficial effect on the development ofpremature infants than tactile stimulation?

    (effect is weight gain and cardiacresponsiveness)

  • 8/2/2019 Experimental &Quasi Experimental Design

    44/70

    EXAMPLE

    AUDITORY

    A1

    TACTILE

    A215 Min

    B1

    A1 B1 A2 B1

    30 Min.

    B2

    A1 B2 A2 B2

    45 Min.

    B3

    A1 B3 A2 B3

    Type of stimulation

    Daily

    Exposure

  • 8/2/2019 Experimental &Quasi Experimental Design

    45/70

    5. Randomized Block Design

    There are two factors (independent

    variables) but one factor is notexperimentally manipulated.

    e.g. Compare the effects of tactile

    versus auditory stimulation for femalevs male infants.

    It can be structured as 2x2 experiment,

    with type of stimulation as one factorand gender as the other factor

  • 8/2/2019 Experimental &Quasi Experimental Design

    46/70

    6. Cross over design (orRepeated

    measures design)It involves the exposure of the same

    subjects to more than one experimental

    treatmente.g. Effectiveness of two alternative

    back rest position ( flat, horizontal vs.

    30o elevation) on intracranial and

    cerebral perfusion pressures in brain

    injured adults

  • 8/2/2019 Experimental &Quasi Experimental Design

    47/70

    PRE-EXPERIMENTAL DESIGNS

    1. The oneshot case study orsingle case study

    The study design has a totalabsence of control, it is

    considered to be little value asan experiment.

  • 8/2/2019 Experimental &Quasi Experimental Design

    48/70

    2.One-GroupPretest-posttest

    design

    Only one group is observed before

    and after the independent variableis introduced. Loss of the controlgroup decreases the usefulness

    of the study but may be necessaryin cases where it is not possibleor feasible to have control groups

  • 8/2/2019 Experimental &Quasi Experimental Design

    49/70

    EXAMPLE

    Effectiveness of selected

    nursing intervention onknowledge and level of

    satisfaction among clientswith stroke

  • 8/2/2019 Experimental &Quasi Experimental Design

    50/70

    ADVANTAGES OF EXPERIMENTAL

    DESIGN

    1.True experiments are the most powerful

    method for testing hypothesis of cause and

    effect relationship between variable because of

    the rigours, precision and control properties ofexperimental design.

    2. This ifthen type of relationship is important

    to nursing and medical researchers because ofits implications for predication and

    explanation.

  • 8/2/2019 Experimental &Quasi Experimental Design

    51/70

    3.Though the control imposed

    by manipulation comparison

    and randomization,

    alternative explanation to acause interpretation can be

    ruled out or discredited.

  • 8/2/2019 Experimental &Quasi Experimental Design

    52/70

    DISADVANTAGES OF EXPERIMENTALDESIGN

    1. Some variables are not feasible or ethicalto manipulate e.g., assessing pregnantwomen to take a new drug found to bedangerous to fetal development would

    not be ethically possible.2. Randomization and otherwise equal

    treatment of control and experimentalgroup can occur in a laboratory, but these

    conditions do not resemble what goesunder real world conditions, andexperimental findings can therefore bebased on rather artificial circumstance

  • 8/2/2019 Experimental &Quasi Experimental Design

    53/70

    3. Experimental design attempt to

    reduce variable to measurableterms. Many of the phenomena thatare important to science in nursing

    are complex, multidimensional andholistic, and defy the reductionismthat has worked reasonably in thephysical or natural sciences.

  • 8/2/2019 Experimental &Quasi Experimental Design

    54/70

    QUASI EXPERIMENTALDESIGN

  • 8/2/2019 Experimental &Quasi Experimental Design

    55/70

    DEFINITION

    It is a research design in

    which the researcher initiates

    an experimental treatment butsome characteristics of a trueexperiment is lacking.

  • 8/2/2019 Experimental &Quasi Experimental Design

    56/70

    PURPOSES

    To provide alternative means ofexamining causality in situations

    not conducive to experimentalgroup.

    To facilitate the search for

    knowledge and examination ofcausality in situations in whichcomplete control is not possible.

  • 8/2/2019 Experimental &Quasi Experimental Design

    57/70

    RANDOM ASSIGNMENT

    It is a procedure used toassign subjects to

    treatment or controlgroups randomly.

  • 8/2/2019 Experimental &Quasi Experimental Design

    58/70

    COMPARISON GROUPS

    Comparison groups are not selected using

    random sampling and do not receive theexperimental treatment.

    Types of comparisons groups:

    Groups that receive no treatment.Groups that receive placebo treatment

    Groups that receive usual treatment.

    Groups that receive second experimentaltreatment.

  • 8/2/2019 Experimental &Quasi Experimental Design

    59/70

    TYPES OF QUASI EXPERIMENTALDESIGN.

    QUASI

    EXPERIMENTALDESIGN

    Interruptedtimes series

    design

    Non equivalentcontrol group

    design

  • 8/2/2019 Experimental &Quasi Experimental Design

    60/70

    NON EQUIVALENT

    COMPARISONGROUP DESIGNS

    1 ONE GROUP POSTTEST ONLY

  • 8/2/2019 Experimental &Quasi Experimental Design

    61/70

    1. ONE GROUP POSTTEST ONLYDESIGN

    It is referred to as pre-experimental ratherthan quasi experimental where manipulation ofIndependent variable is done and posttest isconducted to measure the outcome.

    Manipulation of

    Dependent variable

    Treatment

    Measurement of

    Independent variablePost test

    Example

    Outcome of Antenatal Education on pregnancy andlabor .

    2 POSTTEST ONLY DESIGN WITH A

  • 8/2/2019 Experimental &Quasi Experimental Design

    62/70

    2. POSTTEST ONLY DESIGN WITH ACOMPARISON GROUP

    In this design a comparison group is introducedto determine the efficacy of the treatment and aftermanipulation of independent variable only posttestis conducted.

    Treatment

    Experimental Group

    Post test

    Treatment

    Non equivalent control group

    Post test

    Introducing Computer education to improve staff morale

    3 O G P P D i

  • 8/2/2019 Experimental &Quasi Experimental Design

    63/70

    3. One Group Pretest-Posttest Design

    In this design the experimental group is

    expected to serve as a comparison group. Pretestand posttest scores are used to find the efficiencyof the treatment.

    Experimental Group

    Pre-test Treatment Post test

    Effectiveness pelvic floor muscles exercises in

    control of incontinence of urine.

    Effectiveness of infection control policy to

    decrease infection in postnatal ward.

  • 8/2/2019 Experimental &Quasi Experimental Design

    64/70

    4. PRE-TEST & POSTTEST DESIGNS WITHA COMPARISON GROUP

    In this design the comparison groupis not randomly selected. Theanalysis is made on examining the

    differences between comparisongroup and experimental grouppretest, between pretest and post

    test, between comparison group andexperimental group posttest.

  • 8/2/2019 Experimental &Quasi Experimental Design

    65/70

    Experimental group

    Pre- test Treatment Post-test

    Non equivalent control group

    Pre-test Post-test

    Example:

    Effectiveness of education on partograph in

    assessing labour progress among nurses

    working in labour rooms

    INTURRUPTED TIME SERIES DESIGN

  • 8/2/2019 Experimental &Quasi Experimental Design

    66/70

    INTURRUPTED TIME SERIES DESIGN

    It is similar to descriptive time series design

    except that a treatment is applied at s givenpoint in time and the outcome is measured atdifferent point of time In simple interruptedtime series design there is no control or

    comparisongroup

    Use of multiple methods to measure theoutcome greatly strengthenthe design.

    Experimental groupPre-test T Post test

    O(1)O(2) O(3) o(1) o(2) o(3)

    EXAMPLE

  • 8/2/2019 Experimental &Quasi Experimental Design

    67/70

    EXAMPLE

    Effectiveness of selected nutritionalinterventions on growth anddevelopment pattern of childrenattending under fives clinic at

    Omayal Achi community health centreat Tiruvalluvar district 2006- 2008

  • 8/2/2019 Experimental &Quasi Experimental Design

    68/70

    Interrupted Time series designs witha comparison group

    Comparison group allows examination ofdifferences in trends between groups afterthe treatment and of the persistence of

    treatment effects overtimeExperimental group

    Pre-test T Post test

    o(1) o(2) o(3) o(1) o(2) o(3)

    Comparison groupPre-test T Post test

    o(1)o(2) o(3) o(1) o(2) o(3)

    ADVANTAGES & DISADVANTAGES OF

  • 8/2/2019 Experimental &Quasi Experimental Design

    69/70

    ADVANTAGES & DISADVANTAGES OFQUASI EXPERIMENTAL STUDIES

    Advantages:From the point of view of controls for

    internal validity quasi experimentaldesigns are thought to be superior to pre

    experimental design.Disadvantages:

    Cannot test causal hypothesis

    Do little to ensure external generalizability.

  • 8/2/2019 Experimental &Quasi Experimental Design

    70/70