Psychology 242, Introduction to Research Methods

Document Sample
Psychology 242, Introduction to Research Methods Powered By Docstoc
					       Psychology 242                                                                                  1
        Introduction
        to Research                Exam overview, Spring, 2011


      Go over the summary notes here
               These are general guides – get into the actual slide sets to
                understand them
      During lectures I have noted when you can expect a
       term or concept to be on the exam.
               Go back to your lecture notes, identify when I said something was
                “exam bait”, and make sure you understand those terms
      Give yourself the lectures!!
               Go through the slides and study the areas where you cannot easily
                describe the material to yourself
               Practice writing key terms
               Major focus: 2 statistics sections, within-subjects, and
                complex experiments.

    Psychology 242, Dr. McKirnan          Week 12-13, quasi-experimental designs.           Home
                                                                                    Back          Next
                                                                                            Page
   Psychology 242                                                                                             2
    Introduction
    to Research
                               Introduction to science, 1




 How do we know
  something?
 What does science
  do?
 The core features of a
  research study.
 Overall Research
  approaches.

Psychology 242, Dr. McKirnan                       Weeks 1 & 2; Introduction to science.           Home
                                                                                           Back          Next
                                                                                                   Page
   Psychology 242
    Introduction               Four basic sources of knowledge or information:                      3
    to Research




        Authority:                     “I believe what they tell me to”
                                        Credible / powerful people
                                        Important social institutions
                                        Simple tradition
        Intuition:                    “I believe my Gut feelings”
                                       Emotionality or a “hunch”
                                       “Emotional IQ”
        Empiricism:                    “I believe what I can see”
                                        Simple sensation or perception
                                        Direct observation; data

        Rationalism: “I believe what makes sense.”
                                       Logical coherence
                                       Articulation with other ideas
Psychology 242, Dr. McKirnan             Weeks 1 & 2; Introduction to science.           Home
                                                                                 Back          Next
                                                                                         Page
   Psychology 242
    Introduction
                                                                                            4
    to Research




        Authority:



        Intuition:


        Empiricism:
                                   Most central to
                                    Science
        Rationalism:


Psychology 242, Dr. McKirnan   Week 12-13, quasi-experimental designs.           Home
                                                                         Back          Next
                                                                                 Page
   Psychology 242                                                                                      5
    Introduction
    to Research                Basic features of a research study

      Basic features of research; be able to define these.
             Theory
             Hypothetical construct
             Hypothesis
             Replication
             Operational definition
             Internal & external validity
             Confound
             Independent v. Dependent variables
                       Which is the “cause” & which is the “effect”?
                       Which is measured & which is manipulated?
             Measurement v. experimental studies
Psychology 242, Dr. McKirnan                Weeks 1 & 2; Introduction to science.           Home
                                                                                    Back          Next
                                                                                            Page
   Psychology 242                                                                                    6
    Introduction
    to Research                Core features of a research study:
              Theory                         Hypothetical constructs
                                             In important relationship

                                              More specific variables
        Hypothesis                        

                                             Falsifiable prediction


                                              Operational definition
            Methods                       

                                             Internal & external validity


             Data &                          Numerical representation
                                             Normal distribution
            Analysis                         Probability


                                             Descriptive: Empirical question or exploration
              Results                        Hypothesis: Statistical significance


        Discussion                           Meaning of these results for the theory
                                             Limitations of methods: sample, setting, variables



Psychology 242, Dr. McKirnan           Weeks 1 & 2; Introduction to science.              Home
                                                                                  Back          Next
                                                                                          Page
    Psychology 242                                                                                     7
     Introduction
     to Research               Section 1 study guide



     Core elements of the
      research flow




    Each component of the
     research flow
     corresponds to a later
     component…




Psychology 242, Dr. McKirnan                Weeks 1 & 2; Introduction to science.           Home
                                                                                    Back          Next
                                                                                            Page
   Psychology 242                                                                                                       8
    Introduction
    to Research                Overall research strategies: Drug use

     Observation or Measurement                                                          Experiments
        Simple Description                           Correlational                   Quasi-           “True”
                                                       Studies                     experiments      experiments
  Qualitative                   Quantitative
Research                       Research              Research                    Research           Research
Question:                      Question:             Question:                   Question:          Question:
How does drug                  Who tends to use      What social or ψ                          What brain
                                                                                 Does one form of
use actually                   drugs, how often,     variables are               drug treatment centers control
occur?                         etc.?                 associated with                            “drug craving”?
                                                                                 work better than
                               (epidemiology of      drug use?                   another?      Methods:
Methods:                       drug use).                                Methods:                Experimental
                                                     Methods:
                               Methods:                                                          design:
Direct observation                                    hypothesis-        Experimental-like
                                                                                                Operationalize
of “shooting                    Surveys, face-        oriented surveys design comparing
galleries” or                    to-face                                                         drug “craving”
                                                       or interviews       two treatment
corner drug                      interviews,                                                     in rats (DV),
                                                       (potentially with   groups.
markets, in-depth                archival data                                                  Stimulate
                                                       targeted samples:
interviews with                  (e.g., drug arrests, people in rehab.,  Groups are non-        specific brain
drug users…                      ER visits..)          etc.).              equivalent (not       areas (IV) to
                                                                           blind, not randomly   map brain
                                Block by             Test ψ variables
                                 demographic                               assigned, self-       structure onto
                                                       (motivation,
                                 variables (age,       emotions,           selected…).           craving / drug-
                                 ethnicity…)                                                     seeking.
                                                       attitudes…)

Psychology 242, Dr. McKirnan                   Weeks 1 & 2; Introduction to science.                         Home
                                                                                                     Back          Next
                                                                                                             Page
       Psychology 242                                                                                                          9
        Introduction
        to Research      Overall Research strategies:          Validity
         Observation or Measurement                                                           Experiments
            Simple Description                           Correlational                   Quasi-             “True”
       Qualitative                   Quantitative          Studies                     experiments        experiments
Explore the actual                 Describe a            Relate measured              Test hypotheses     Test specific
process of a                       behavioral or         variables to each            in naturally        hypotheses via
behavior.                          social trend.         other to test                occurring events    controlled “lab”
                                                         hypotheses.                  or field studies.   conditions.


                        External validity                                                  Internal validity

                          Less control:                                                      More control:
    Observe / test phenomenon                                                  Isolate (or create) the
     under natural conditions.                                                   phenomenon in a controlled
                                                                                 environment
    More accurate portrayal of how it
     works in nature                                                            Addresses specific questions
                                                                                 or hypotheses
    Less able to interpret cause &
     effect                                                                     More ability to interpret cause
                                                                                 & effect
    Psychology 242, Dr. McKirnan                   Weeks 1 & 2; Introduction to science.                            Home
                                                                                                            Back          Next
                                                                                                                    Page
                                                                          10
                               Research ethics
   Psychology 242
    Introduction
    to Research




 The Common Rule: What can
  researchers “do” to get data?

 The Belmont Report and the
  Informed Consent document




Psychology 242, Dr. McKirnan         Research Ethics           Home
                                                       Back          Next
                                                               Page
    Psychology 242                                                                         11
                      The “Common Rule” criteria for Human
     Introduction
     to Research



                      Subjects Protection

   Minimization of risks
   Reasonableness of risks (cost – benefit analysis)
   Equitable subject selection
   Informed consent
   Documentation of consent
   Data monitoring for safety
   Provisions to protect vulnerable participants &
    maintain confidentiality

Psychology 242, Dr. McKirnan    Weeks 1 & 2; Introduction to science.           Home
                                                                        Back          Next
                                                                                Page
                                                                                12
                               Belmont Report
   Psychology 242
    Introduction
    to Research




1. Respect For Persons
       Right to exercise autonomy & make informed choices.

2. Beneficence
       Minimization of risk + maximization of social/individual benefit
      How much information should participants get from a blinded,
        randomized trial? See ethics of clinical trials

3. Justice
       Research should not unduly involve groups unlikely to benefit
        from subsequent applications.
       Include participants of all races & both genders
       Members of target population on design & research team
       Research & researchers contribute to study population studied
       Communicate research results & develop programs/ interventions
Psychology 242, Dr. McKirnan         Research Ethics                 Home
                                                             Back          Next
                                                                     Page
   Psychology 242                                                                 13
        Quasi-experimental designs
                               Quasi-experiments
    Introduction
    to Research




        Experimental designs for “studies in nature”.


Studying naturally occurring events
       Measurement studies
       Retrospective designs

Evaluate existing groups or program
      Single shot survey or measure
      Non-equivalent groups
      Time series designs
Psychology 242, Dr. McKirnan                 Research Ethics           Home
                                                               Back          Next
                                                                       Page
                                                                                              14
                        Quasi-experiments
    Psychology 242
     Introduction
     to Research



1. Study naturally occurring events that could not be
    brought into a lab or a true experiment.
             Measurement studies
             Retrospective designs


2. Evaluate existing groups or program(s)
            Simple survey of an intervention that already occurred
            Non-equivalent designs (due to…)
              Self-selection
              Non-random assignment
              Use of existing groups
              Participants not blind
            Time series designs, often with archival data


 Psychology 242, Dr. McKirnan    Week 12-13, quasi-experimental designs.           Home
                                                                           Back          Next
                                                                                   Page
     Psychology 242                                                                                         15
      Introduction
      to Research                True v. quasi-experimental designs, 3


True experiments:                                                Quasi-experiments:

Emphasize Internal Validity                                      Emphasize External Validity
 Assess cause & effect (in relatively                            Describe “real” / naturally

  artificial environment)                                          occurring events
 Test clear, a priori hypotheses                                 Clear or exploratory hypotheses



Groups Equivalent at baseline          Non-equivalent groups
 Random Assignment (or matching).      Non-random assignment

 Participants & experimenter Blind to  Existing groups
  assignment.                           Self-selection

                                        Participants not blind.



Control study procedures              Complete Control not Possible
 Create / manipulate the independent  May not be able to manipulate the
  variable                              independent variable
 Control procedures & measures        Partial control of procedures &

                                        measures
  Psychology 242, Dr. McKirnan                Week 12-13, quasi-experimental designs.            Home
                                                                                         Back          Next
                                                                                                 Page
                                                                                                   16
Quasi-experiments that do not have a control group:
     Psychology 242
      Introduction
      to Research




   Group                  Observe1            Intervention or event            Observe2

                          Observe1                    Confound                 Observe2


Threats to internal validity (confounds):
                                                  Historical / cultural events occur between
 History                                         baseline & follow-up.

                                                  Individual maturation or growth occurs
 Maturation                                      between baseline & follow-up.

                                                  People respond to being measured or
 Reactive measures                               being a measured a second time.

                                                  Extreme scores at baseline “regress” to a
 Statistical regression                          more moderate level over time.

                                                  People leave the experiment non-
 Mortality / drop-out                            randomly (i.e., for reasons that may affect
                                                  the results…).
  Psychology 242, Dr. McKirnan       Week 12-13, quasi-experimental designs.            Home
                                                                                Back          Next
                                                                                        Page
    Psychology 242                                                                                                 17
     Introduction
     to Research                Non-equivalent designs; pre- post-

#2 Two Group Pre- Post- Design

  Group                  Observe1                     Intervention or event                     Observe2

  Group                  Observe1                          Contrast group                       Observe2


Non-equivalent groups                                      Intervention & Assessments often
                                                               controlled by researcher in these
 Self-selection                                               designs.
 Non-random
  assignment                                                                             Similar to true
 Use of existing groups                                                                 experimental
 Participants not blind                                                               design, except for
                                                                                        non-equivalent
                                                                                            groups


 Psychology 242, Dr. McKirnan                Week 12-13, quasi-experimental designs.                    Home
                                                                                                Back          Next
                                                                                                        Page
   Psychology 242
    Introduction
                                                                                 18
    to Research




Research Sampling.

 Define your target population
 Operationally define your criteria for
  population “membership”
 Define your sampling frame
 Probability sampling
 Non-probability sampling


Psychology 242, Dr. McKirnan   Week 3; Experimental designs           Home
                                                              Back          Next
                                                                      Page
                                                                                                     19
                               Sampling overview
   Psychology 242
    Introduction
    to Research



Who do you want to generalize to?
          Who is the target population?
              broad – external validity

              narrow – internal validity

          How do you decide who is a member?
              demographic / behavioral criteria?

              subjective / attitudinal?

          What do you know about the population already – what
           is the “sampling frame”.
Is a Probability or random sample possible?
          “Hidden” population?
          Socially undesirable research topic?
          Easily available via telephone, door-to-door?
          Sampling frame adequate to choose selection method?

Psychology 242, Dr. McKirnan            Week 12-13, quasi-experimental designs.           Home
                                                                                  Back          Next
                                                                                          Page
                                                                                                            20
                               Sampling overview, 2
   Psychology 242
    Introduction
    to Research




Probability sampling                                             Most externally valid
    simple                                                      Assumes:
                                                                       Clear sampling frame
    multi-stage
                                                                       Population is available
    cluster or stratified
                                                                 Less externally valid for
                                                                  hidden groups.

Non-probability sampling                                         Less externally valid
   targeted / multi-frame
                                                                 High “convenience”
                                                                 Best when:
   snowball
                                                                       No clear sampling frame
   quota, etc.                                                        Hidden / avoidant
                                                                        population.



Psychology 242, Dr. McKirnan             Week 12-13, quasi-experimental designs.                 Home
                                                                                         Back          Next
                                                                                                 Page
                                                                                                  21
                               Descriptive research
    Psychology 242
     Introduction
     to Research




                                           Qualitative or
       Quantitative                                                    Existing data
                                           Observational
Describe an issue via                 Study behavior “in          Use existing data for
 valid & reliable                      nature” (high               new quantitative (or
 numerical measures                    ecological validity).       qualitative) analyses
Simple: frequency                     Qualitative                 Accretion
    counts of key                         In-depth interviews       Study “remnants” of
    behavior                              Focus (or other)           behavior
Blocking” by other
“                                          groups                    Wholly non-reactive
    variables                             Textual analysis       Archival
Correlational                             Qualitative              Use existing data to
    research: “what relates                quantitative               test new hypothesis
    to what”
                                      Observational                  Typically non-
Complex                                   Direct                     reactive
 modeling                                 Unobtrusive


Psychology 242, Dr. McKirnan              Descriptive Research.                        Home
                                                                               Back          Next
                                                                                       Page
                  Statistics: an introduction                  22
 Psychology 242
  Introduction
  to Research




 Using numbers in science

 Number scales & frequency distributions

 Central Tendency: Mode, Median, Mean

 Variance: Standard Deviation

 The Z score and the normal distribution

 Using Z scores to evaluate data

 Testing hypotheses: critical ratio.

                                                    Home
                                            Back          Next
                                                    Page
   Psychology 242
    Introduction               Plato’s cave as a core allegory of scientific statistics (“run              23
    to Research
                               show” and mouse over different areas for text).




Psychology 242, Dr. McKirnan                 Week 12-13, quasi-experimental designs.            Home
                                                                                        Back          Next
                                                                                                Page
       Psychology 242                                                                               24
        Introduction
        to Research
                                   Core assumptions of the scientific approach


     We never observe “nature” directly, we only see its
      manifestations or images:

              We study hypothetical constructs; basic
               “operating principles” of nature; e.g., learning, motivation,
               “health”…
              We can only observe the effects of these basic principles, not
               the processes themselves.
              Rational analysis – theory – helps us deduce what the “form”
               of these processes must be, and how they work. We then test
               predictions from these theories.

     Theories are tested & variables are measured only with
      samples of people, places, and psychological states,
      not the entire population of people, places, or events.
    Psychology 242, Dr. McKirnan            Exam #3 study guide                          Home
                                                                                 Back          Next
                                                                                         Page
      Psychology 242                                                                         25
       Introduction
       to Research                External validity: summary


                                       Is the sample typical of the
                                            larger population?

                                                     The
                                                   research                   Is this
  Is the
                                                   Sample:                  typical of
 outcome
 measure                        The               The study               “real world”
represen-                    Dependent           structure & The research    settings
  tative,                     Variable             context     Setting:    where the
  valid &                                                                 phenomenon
 reliable?                                           The                     occurs?
                                                 Independent
                                                   Variable

                       Does the experimental manipulation (or measured
                          predictor) actually create (validly assess…) the
                                                             are
                             phenomenon you sampling interested in?
   Psychology 242, Dr. McKirnan        Experimental Design &                      Home
                                                                          Back          Next
                                                                                  Page
                                                                                                         26
                               Types of numerical scales
    Psychology 242
     Introduction
     to Research



Ratio
            zero point grounded in physical property; “absolute” values
             continuous & equal intervals
            physical description: elapsed time, height
                                                                                     Continuous
Interval                                                                             scales
            no zero point; scale values relative                                     (scores on a
            continuous with equal interval                                           continuum)
            behavioral research, e.g., attitude or rating scales.
Ordinal
            rank order with non-equal intervals; no ‘0’ point
            Simple finish place, rank in organization...
Categorical
             ‘values’ = categories only
            inherent categories: ethnic group, gender, zip code

Central tendency                                             used for:
    Mode (most common score)                               categorical variables
    Median (middle of distribution)                        categorical or continuous variables
    Mean (average score)                                   continuous variables only
Psychology 242, Dr. McKirnan           Exam #3 study guide                                    Home
                                                                                      Back          Next
                                                                                              Page
   Psychology 242                                                                                      27
    Introduction
    to Research                Distributions
                                                                Mode
                                                                                         Mean

Normal distribution: mean = mode =                              Median
    median at center of the distribution




                                                                         Mean                   Median
Bimodal distribution Mean & median
   are similar, at the center.                                 Mode




Skewed distribution: Extreme scores in                                                  Mode
one direction make the median, and mean
                                                                                        Median
larger than the mode.
                                                                                                Mean




Psychology 242, Dr. McKirnan             Exam #3 study guide                             Home
                                                                                Back             Next
                                                                                         Page
    Psychology 242
     Introduction
                                                                                   28
     to Research
                           Measures of Dispersion or Variance


Two estimates of Variance:
1. The Range of the highest to the lowest score. Similar
        to “average” amount each score deviates from the M.
       Simple & easy to compute.
       Provides simple idea of where scores fall
       Very sensitive to any extreme score(s) (“outliers”).


2. Standard deviation of scores around the Mean
    Similar to “average” amount each score deviates from the M.
    “Standardizes” scores to a normal curve, allowing basic statistics
     to be used.
    More accurate & detailed than range:



Psychology 242, Dr. McKirnan            Exam #3 study guide             Home
                                                                Back          Next
                                                                        Page
    Psychology 242                                                                      29
     Introduction
     to Research               Introduction to normal distribution



 Properties of the normal distribution

    The normal distribution is a hypothetical
     distribution of cases in a sample
    It is segmented into standard deviation units.
    Each standard deviation unit (Z) has a fixed %
     of cases
    We use Z scores & associated % of the normal
     distribution to make statistical decisions about
     whether a score might occur by chance.


Psychology 242, Dr. McKirnan                Exam #3 study guide              Home
                                                                     Back          Next
                                                                             Page
   Psychology 242
    Introduction
                                                                                                        30
                    The Normal Distribution
    to Research




         34.13% of scores
        from Z = 0 to Z = +1                                      Z = # of standard deviation
               and                                                  units from mean;
        from Z = 0 to Z = -1                                      M = 0, each standard
                                                                    deviation = 1

           13.59% of scores
                                                                       Each segment of the curve
                                                                       (from Z=0 to Z=-1, and from
                                                                       0 to +1) takes up a fixed
 2.25% of scores                                                       percentage of area or % of
                                                                       cases.




                               -3   -2     -1       0   +1        +2   +3
                                                 Z Scores
                                         (standard deviation units)


Psychology 242, Dr. McKirnan                Exam #3 study guide                              Home
                                                                                     Back          Next
                                                                                             Page
    Psychology 242                                                                 31
     Introduction
     to Research                Normal distribution; Z scores

 Calculating a Z score to evaluate data

 To use the normal distribution to calculate how ‘good’ a score is…
 1.        Calculate how far the score (X) is from the mean (M); X–M.
 2.        “Adjust” X–M by how much variance there is in the sample; use
           the standard deviation (S).
 3. Z = X–M / S

How “good” is a score of ‘6' in two groups?

 A. The distance of the score from the M.
 B. The variance in the rest of the sample
 With low variance ‘6’ is higher (relative to other scores) then in a
     sample with higher variance.
 C. Criterion for “significantly good” score
                            the guide
 What % of the sample must#3 study score be higher than…
                       Exam
 Psychology 242, Dr. McKirnan                                           Home
                                                                Back          Next
                                                                        Page
    Psychology 242                                                                                         32
     Introduction
     to Research                Using the normal distribution, 2


A. The distance of the score from the M.
       The participant is 2 units above the mean in both
          tables.


B. The variance in the rest of the sample:
         Table 1, high variance                                     Table 2, low(er) variance
     Mean (M) = 4, Score (X) = 6                                      Mean (M) = 4, Score (X) = 6
    Standard Deviation (S) = 2.4.                                   Standard Deviation (S) = 1.15.
           (X-M = 6 - 4 = 2)                                               (X-M = 6 - 4 = 2)
       Z (X-M/S) = 2/2.4 = 0.88                                        Z (X-M/S) = 2/1.15 = 1.74
  About 70% of participants are                                    About 90% of participants are
  below this Z score                                               below this Z score



 Psychology 242, Dr. McKirnan                Exam #3 study guide                                Home
                                                                                        Back          Next
                                                                                                Page
    Psychology 242                                                                                33
     Introduction
     to Research                Evaluating scores using Z


C. Criterion for a “significantly good” score
X = 6, M = 4, S = 2.4, Z = .88

   X = 6, M = 4, S = 1.15, Z = 1.74




                                      70% of cases

                                           90% of cases
                                -3    -2      -1       0   +1        +2   +3
                                                    Z Scores
                                            (standard deviation units)
 Psychology 242, Dr. McKirnan                  Exam #3 study guide                     Home
                                                                               Back          Next
                                                                                       Page
    Psychology 242                                                                                  34
     Introduction
     to Research                Core research questions


 Data                                                Statistical Question
One participant’s score                             Does this score differ from the M for the
                                                    group by more than chance?

                                                    Does this M differ from the M for the
The mean for a group                                general population by more than
                                                    chance?


Means for 2 or more                                 Is the difference between these Means
                                                    more than we would expect by chance?
groups                                              -- more than the M difference between
                                                    any 2 randomly selected groups?

Scores on two                                       Is the correlation (‘r’) between these
measured variables                                  variables more than we would expect
                                                    by chance -- more than between any
                                                    two randomly selected variables?

 Psychology 242, Dr. McKirnan               Exam #3 study guide                          Home
                                                                                 Back          Next
                                                                                         Page
   Psychology 242                                                                 35
    Introduction
    to Research                Summary
 Numbers are important for representing “reality” in
  science (and other fields).
 Different measures of central tendency are useful &
  accurate for different data;
           Mean is the most common.
           Median useful for skewed data
           Mode useful for simple categorical data

 Variance (around the mean) is key to characterizing a
  set of numbers.
 We understand a set of scores in terms of the:
           Central tendency – the average or Mean score
           The amount of variance in the scores, typically the Standard
            Deviation.
Psychology 242, Dr. McKirnan      Statistics introduction 1            Home
                                                               Back          Next
                                                                       Page
       Psychology 242
        Introduction               Summary                                                                           36
        to Research




     Statistical decisions
      follow the critical ratio:

 Z is the prototype critical ratio:

                    How far is your score (X) from the mean (M)
                                                                                                    X–M
    Z=             How much variance is there among all the scores in the        =
                    sample [standard deviation (S)]                                                  S

 t is also a basic critical ratio used for comparing groups:

                How different are the two group Means
                                                                                  M1 – M2
t=             How much variance is there within each the two groups;
                                                                        =   Variance     grp1       Variance     grp2
                (“standard error of the mean”)                                                  
                                                                                n grp1                  n grp2



    Psychology 242, Dr. McKirnan            Statistics introduction 1                                    Home
                                                                                                Back            Next
                                                                                                         Page
 Psychology 242                                                                                            37
                                                                                              Revised 4/5/09
  Introduction
  to Research
                  Dr. McKirnan, Psychology 242


 Introduction to statistics # 2
 What can Plato’s Allegory of
  the Cave tell us about
  scientific reasoning?
 Was our hypothesis
  supported? The critical ratio
  and the logic of the t-test.
 The central limit theorem and
                                                          "The Allegory of the Cave" by Allison Leigh Cassel
  sampling distributions
 Correlations and assessing
  shared variance


                                                                                                Home
                             Statistics Introduction 2.                               Back            Next
                                                                                                Page
   Psychology 242                                                                     38
    Introduction
    to Research     Plato’s Cave, 6

  What does Plato’s Allegory of the Cave tell us
  about scientific reasoning?
We cannot observe “nature” directly, we only see its
 manifestations or images:


 We are trapped in a world
  of immediate sensation;
 Our senses routinely
  deceive us (they have error).
 We cannot get outside
  our limited sensations to
  see the underlying “form”
  of nature

                                      Statistics Introduction 2.           Home
                                                                   Back          Next
                                                                           Page
    Psychology 242                                                                     39
     Introduction
     to Research     Plato’s Cave, 7

    We study hypothetical constructs; basic
      “operating principles” of nature
           e.g., evolution, gravity, learning, motivation…

   Processes that we
    cannot “see” directly…
   …that underlie events
    that we can observe.
   We use rational analysis
    – theory – to deduce
    what the “form” of these
    processes must be, and
    how they work.

                                       Statistics Introduction 2.           Home
                                                                    Back          Next
                                                                            Page
    Psychology 242
     Introduction
                                                                     40
     to Research




     Why can’t we just observe “nature” directly?
1. We can only observe the effects of hypothetical
   constructs, not the processes themselves.
2. Our theory helps us develop hypotheses about
   what we should observe if our theory is “correct”.
3. We test our hypotheses to infer how nature works.
4. Our inferences contain error: we must estimate the
   probability that our results are due to “real” effects
   versus chance.




 Statistics Introduction 2.                               Home
                                                  Back          Next
                                                          Page
   Psychology 242
    Introduction               “Statistical significance”                                     41
    to Research




 Using numbers to test Statistical Significance
     We assume a score with less than 5% probability of
      occurring has not occurred by chance alone (i.e.,
      higher or lower than 95% of the other scores… p < .05)
     Z > +1.98 occurs < 95% of the time (p <.05).
     If Z > 1.98 we consider the score to be
      “significantly” different from the mean
     To test if an effect is “statistically significant”
               Compute a Z score for the effect
               Compare it to the critical value for p<.05; + 1.98


Psychology 242, Dr. McKirnan                   Statistics introduction 1           Home
                                                                           Back          Next
                                                                                   Page
    Psychology 242                                                                                          42
     Introduction
     to Research        Statistical significance & Areas under the normal curve


95% of scores are between Z = -1.98 and Z = +1.98.


                                       Z = -1.98                                  Z = +1.98




                                                                                             2.4% of
                                    2.4% of                                                   cases
                                     cases                    About 95%
                                                               of cases



                                              -3      -2      -1     0    +1    +2      +3
                                                                   Z Scores
                                                           (standard deviation units)

 Psychology 242, Dr. McKirnan           Exam #3 study guide                                      Home
                                                                                         Back          Next
                                                                                                 Page
       Psychology 242                                                                                                         43
        Introduction
        to Research        Statistical significance & Areas under the normal curve

                                                                               With Z > +1.98 or < -1.98
            In a hypothetical                                                  we reject the null
                distribution:                                                  hypothesis & assume
     2.4% of cases are higher                                                 the results are not by
      than Z = +1.98                                                           chance alone.

     2.4% of cases are lower
      than Z = -1.98                                                     34.13% 34.13%

     Thus, Z > +1.98 or < -1.98                            Z = -1.98       of
                                                                          cases
                                                                                   of
                                                                                 cases           Z = +1.98
      will occur < 5% of the time
                                                                                                                  2.4% of
      by chance alone.                                             95% of cases 13.59%                             cases
                                                            13.59%
                                                               of                          of
                             2.4% of                         cases                       cases
                              cases
                                                2.25%                                             2.25%
                                                  of                                                of
                                                cases                                             cases


                                        -3      -2        -1        0       +1   +2      +3
                                                                          Z Scores
                                                          (standard deviation units)
    Psychology 242, Dr. McKirnan             Statistics introduction 1                                             Home
                                                                                                          Back           Next
                                                                                                                   Page
       Psychology 242                                                                                          44
        Introduction
        to Research                Critical ratio

                                                      The strength of the results (our
                                                             direct observation of nature)
   Critical ratio =
                                                Amount of error variance (the odds
                                                    that our observation is due to chance)

                                                       Difference between Ms for the two groups
                                        t=
                                                                 Variability within groups (error)

                                                              Mgroup2
                   Mgroup1


                                                                 Within-group
Within-group                                                     variance, group2
variance, group1




  control group                         experimental group
    Psychology 242, Dr. McKirnan                    Exam #3 study guide                             Home
                                                                                            Back          Next
                                                                                                    Page
       Psychology 242                                                                                                45
        Introduction
        to Research                The Central Limit Theorem; large samples


Central Limit Theorem                                  True
                                                    Population M
   With many scores (high n)                                              Score Score      “True” normal
                                                                                             distribution
    “deviant” values are                                                   Score
                                                                               Score
                                                                       Score
    completely offset by other                                        Score Score
                                                                                  Score
    values.                                                         Score
                                                                          Score Score
                                                                    Score
                                                                   Score ScoreScore
                                                                                    Score
                                                                               Score
   The distribution is                                               Score Score
                                                                                    Score
                                                                 Score Score Score
    normal, with low(er)                                                             Score
                                                                  Score Score Score
    variance.                                                 Score Score
                                                                                      Score
                                                                                Score Score
                                                               Score Score
   The sampling                                         Score Score Score Score Score Score
                                                           Score Score Score Score Score Score
    distribution well                             Score      Score Score    Score   Score Score      Score
    approximates the                                   <-- smaller             M         larger --->
    population
    distribution
    Psychology 242, Dr. McKirnan               Exam #3 study guide                                        Home
                                                                                                  Back          Next
                                                                                                          Page
    Psychology 242                                                                               46
     Introduction
     to Research                Central limit theorem & evaluating t scores

Central Limit Theorem

 The same logic applies with samples we used
 to test hypotheses.
1. If the groups in an experiment have small samples, it is
   more likely that high error variance, rather than an
   actual experimental effect, led to differences between
   the Ms.
2. Smaller samples (lower df) = more variance, so t must
   be larger for us to consider it statistically significant (<
   5% likely to have occurred by chance alone).
3. Compare t to a sampling distribution based on df.
4. Critical value for t with p <.05 goes up or down
   depending upon sample size (df)
 Psychology 242, Dr. McKirnan                Exam #3 study guide                      Home
                                                                              Back          Next
                                                                                      Page
        Psychology 242
         Introduction
                         Central Limit Theorem; variations in sampling                                  47
         to Research
                         distributions


   As samples sizes (df) go                                     N > 120, t > + 1.98, p<.05
    down, the estimated
                                                                 df = 18, t > + 2.10, p<.05.
    sampling distributions of t
    scores based on them have                                    df = 8, t > + 2.30, p<.05.
    more variance, giving a
    more “flat” distribution.
                                                                            This makes the
                                                                             critical value for
                                                                             p<.05 increase.




                                    -2     -1           0       +1      +2
                                                      Z Score
      2.4% of cases below this value       (standard deviation units)   2.4% of cases above this value


     Psychology 242, Dr. McKirnan        Exam #3 study guide                                 Home
                                                                                     Back          Next
                                                                                             Page
    Psychology 242                                                                              48
     Introduction
     to Research            Summary: Statistical tests

t-test: Compare one group to another
    Experimental v. control                         (Experiment)
    Men v. women, etc.                              (Measurement)

   Calculate M for each group, compare them to determine
    how much variance is due to differences between
    groups.
   Calculate standard error to determine how much
    variance is due to individual differences within each
    group.
                                                             Difference between groups
   Calculate the critical ratio (t):                            standard error of M




 Psychology 242, Dr. McKirnan          Exam #3 study guide                           Home
                                                                             Back          Next
                                                                                     Page
    Psychology 242                                                                         49
     Introduction
     to Research                Statistics summary: correlation


Pearson Correlation: measures how similar the
   variance is between two variables (A.K.A. “shared
   variance”) within one group of participants.

            are people who are above or below the mean on one
             variable similarly above or below the M on the second
             variable.

            If everyone who is a certain amount over the M on one
             variable (say, Z = +1.5) is the same amount above M on the
             other variable (Z also = +1.5) the correlation would be
             +1.0.
Assess shared variance by multiplying                                  ZX * ZY
 the person’s Z scores for each of the                             r
 two variables / df:                                                    n1

 Psychology 242, Dr. McKirnan                Exam #3 study guide                Home
                                                                        Back          Next
                                                                                Page
    Psychology 242                                                                           50
     Introduction
     to Research             Inferential statistics: summary, Key terms


    Plato’s cave and the estimation of “reality”
    Inferences about our observations:
              Deductive v. Inductive link of theory / hypothetical constructs & data
              Generalizing results beyond the experiment
     Critical ratio (you will be asked to produce and describe this).
    Variance, variability in different distributions
    Degrees of Freedom [df]
    t-test, between versus within –group variance
    Sampling distribution, M of the sampling distribution
    Alpha (α), critical value
    t table, general logic of calculating a t-test
    “Shared variance”, positive / negative correlation
    General logic of calculating a correlation (mutual Z scores).
    Null hypothesis, Type I & Type II errors.
Statistics Introduction 2.                                                        Home
                                                                          Back          Next
                                                                                  Page
   Psychology 242
    Introduction
                                                     51
    to Research




Multiple independent variables    4/14/09



  Testing hypotheses about > 1
   independent variable
  Factorial Designs:
          Main effects,
          Additive Effects,
          Interactions

Psychology 242, Dr. McKirnan              Home
                                  Back          Next
                                          Page
    Psychology 242                                                                    52
     Introduction
     to Research        Designs with > 1 independent variable
1. Including a ‘control’ variable as an I.V.
       e.g., gender, age, race, etc.
       test if I.V. has the same effect within both groups


2. Testing hypotheses re: 2 or more I.V.s
       A. test separate, ‘main effects’ of each I.V. (Do
                each of these variables significantly affect the outcome?)

       B. test ‘additive’ effects of > 1 I.V.s
          simultaneously (What is the combined effect of these
                variables?)

       C. test interaction of 2 or more I.V.s (Does the
                effect of one I.V. on the outcome depend upon a second
                variable...?)
 Psychology 242, Dr. McKirnan    Exam #3 study guide                       Home
                                                                   Back          Next
                                                                           Page
    Psychology 242
     Introduction               Interaction example: Genetics, stress and depression, 1              53
     to Research



Interaction is very strong in an analysis of
  childhood trauma and depression.


                                                                    There is a
                                                                    general (main)
                                                                    effect whereby
                                                                    more trauma
                                                                    leads to greater
                                                                    likelihood of
                                                                    adult depression




 Psychology 242, Dr. McKirnan              Exam #3 study guide                            Home
                                                                                 Back           Next
                                                                                          Page
    Psychology 242                                                                                   54
     Introduction
     to Research
                                Interaction example: Genetics, stress and depression, 2


However … the effect of trauma interacts with genetics


                                                                    Childhood
                                                                    trauma has no
                                                                    effect in people
                                                                    who have no
                                                                    genetic
                                                                    vulnerability.
                                                                    With increasing
                                                                    vulnerability,
                                                                    increasing
                                                                    trauma increases
                                                                    the likelihood of
                                                                    depression
 Psychology 242, Dr. McKirnan              Exam #3 study guide                            Home
                                                                                 Back           Next
                                                                                          Page
    Psychology 242
     Introduction               Example of a 3-way interaction                              55
     to Research




Figure 3 Mean ratings of subjective stimulation and sedation on the BAES
     under 0.65 g/kg alcohol and placebo in women and men.

Alcohol (v. placebo) made                                 Alcohol made women much
men much more stimulated.                                 more sedated




 Psychology 242, Dr. McKirnan      Multiple independent variables                Home
                                                                         Back          Next
                                                                                 Page
   Psychology 242                                                                                                     56
 Alternate portrayal of 3-way mood interaction
    Introduction
    to Research




  Placebo conditions do not show                                            The alcohol conditions show a
     much effect                                                              classic “cross-over” effect for
                                                                              gender & mood;


                                50
                                45
                                                                                        Men get aroused
                                40
       M BAES subscale scores




                                35
                                30                                                        Men, Alcohol
                                                                                          Men, Placebo
                                25
                                                                                          Women, Alcohol
                                20                                                        Women, Placebo
                                15
                                10
                                                                                      Women get sedated
                                5
                                0
                                     Stimulation                  Sedation

Psychology 242, Dr. McKirnan                       Multiple independent variables                          Home
                                                                                                   Back          Next
                                                                                                           Page
    Psychology 242                                                                              57
     Introduction
     to Research                Multiple IVs; summary 1


Multiple Independent Variables / Predictors:

    Tell us much more than simple main effects
              Some variables may combine with others (additive
               effects…) to produce very strong effects
              Variables may interact with others, so that they
               only affect the outcome at one level of a second
               variable…
Drug use during sex leads to risk, primarily among those who have
 strong expectations that drugs decrease stress.

Stress strongly predicts depression, only among people who are
 genetically vulnerable.



 Psychology 242, Dr. McKirnan               Multiple independent variables           Home
                                                                             Back          Next
                                                                                     Page
    Psychology 242                                                                              58
     Introduction
     to Research                Multiple IVs; summary 2


Multiple Independent Variables / Predictors:
    Are critical to theory development and testing:
Changing sexual risk reduction requires that we understand both
  peoples’ psychological dispositions and their drug use patterns.

Stress or other environmental events can “switch on” genes that create
 psychological or other problems; genetic dispositions and environment
 are not separate processes.

    Establish key “boundary conditions” to theory:
     when and among whom does a basic
     psychological process operate?
Alcohol makes it more difficult to inhibit behavior, but primarily among
  men.

 Psychology 242, Dr. McKirnan               Multiple independent variables           Home
                                                                             Back          Next
                                                                                     Page
    Psychology 242                                                                           59
     Introduction
     to Research               Summary


    Key terms:
             Main effect
             Additive effect
             Interaction
             Cross-over interaction
             Factorial design
             Repeated measure




Psychology 242, Dr. McKirnan             Multiple independent variables           Home
                                                                          Back          Next
                                                                                  Page
   Psychology 242
    Introduction
                                                             60
    to Research




    Complex experiments:
    Within- subjects & blocking designs


 Own control
 Reversal designs
 Repeated measures
         & Randomized
         block designs


Psychology 242, Dr. McKirnan                      Home
                                          Back          Next
                                                  Page
   Psychology 242
    Introduction               Basic forms of within-subjects designs, 1
                                                                                              61
    to Research




Basic forms of within subjects designs;
1. Own control
         Each participant in control and
          experimental group.
         Optimally, order is counter-balanced

2. Reversal designs
3. Repeated measures & Randomized
      block designs


Psychology 242, Dr. McKirnan                 Exam #3 study guide                   Home
                                                                           Back          Next
                                                                                   Page
   Psychology 242
    Introduction               Basic forms of within-subjects designs, 3
                                                                                              62
    to Research




Basic forms of Within subjects designs;
1. Own control
2. Reversal designs
         Hypothesis: behavior controlled by clearly
                 bounded condition

         Design: “A – B – A”; impose – withdraw –
                 impose condition

3. Repeated measures & Randomized
      block designs

Psychology 242, Dr. McKirnan                 Exam #3 study guide                   Home
                                                                           Back          Next
                                                                                   Page
   Psychology 242
    Introduction               Basic forms of within-subjects designs, 2
                                                                                              63
    to Research




Basic forms of Within subjects designs;
1. Own control
2. Reversal designs
3. Repeated measures
                Multiple treatment conditions: each
                 participant gets each treatment.
                Longitudinal / time sampling: each
                 participant assessed over multiple time periods
                Randomized block designs; Repeated
                 measure combined with between-groups
                 variable.
Psychology 242, Dr. McKirnan                 Exam #3 study guide                   Home
                                                                           Back          Next
                                                                                   Page
    Psychology 242                                                                                            64
     Introduction
     to Research                Within subjects designs; own control, 2


1. Own Control Design
 Single               Control                     Observe1                Experimental        Observe2
 Group                Condition                                           Condition


       All participants get the Control                              All participants then get the
        Condition and measurement                                    experimental intervention and
                                                                     measurement.

 Experimental manipulation potentially very sharp for
  participants if there are no carry-over effects.
 Hypothesis tested by differences between conditions
  (Observation1 v. Observation2) within group.
 Internal validity: eliminate possible confound of group
  differences at baseline, since there is only one group.
 Statistical power increased: requires fewer subjects.

 Psychology 242, Dr. McKirnan                  Exam #3 study guide                                 Home
                                                                                           Back          Next
                                                                                                   Page
   Psychology 242
    Introduction
                                                                                    65
    to Research                Reversal designs


2. “REVERSAL” DESIGNS
                        Test at baseline in normal state,


    Test under temporary experimental condition


       Test again under normal state.
Examples:
 Role of incentives in enhancing performance

 Impact of anti-depressant drug on mood

 Effect of self-awareness on following social
  norms
Psychology 242, Dr. McKirnan               Exam #3 study guide           Home
                                                                 Back          Next
                                                                         Page
   Psychology 242                                                                                     66
    Introduction
    to Research                Examples of reversal designs


Test effect of, e.g., modeling (observation of
attractive experimental confederate) on
alcohol consumption.        If the model influences
                                                                     participant’s behavior:
    50
                                                                    Consumption will
    45
    40
                                                                     increase when the
    35
                                                                     model’s does…
    30                                                              Rate goes back down
    25
                                                                     when model’s does.
    20
    15                                                              Up again with model, etc..
    10
     5
     0
                    Lo




                                             Lo
                                   H




                                                        H
                                    ig




                                                         ig
                      w




                                               w
                                      h




                                                           h




Psychology 242, Dr. McKirnan
                                     Model's drinking rate
                                           Exam #3 study guide                             Home
                                                                                   Back          Next
                                                                                           Page
   Psychology 242
    Introduction               Basic forms of within-subjects designs, 4
                                                                                              67
    to Research




Basic forms of Within subjects designs;
1. Own control
2. Reversal designs
3. Repeated measures & Randomized
      block designs
          Combine blocking or grouping variable
           with repeated measure.
          Most common repeated / within-Ss design
                     Biomedical research
                     Behavioral intervention evaluations
Psychology 242, Dr. McKirnan                 Exam #3 study guide                   Home
                                                                           Back          Next
                                                                                   Page
       Psychology 242                                                                    68
        Introduction
        to Research                Randomized block designs


Blocking Variable; between - subjects factor
     “Person” variable; age, gender, ethnicity, etc.
               not a “true” IV since people not randomly assigned;
Or:
       Experimental condition; drug dose, treatment, etc.
               “True” IV with random assignment

Repeated measure: within-subjects factor
      Multiple treatment conditions: each participant
       is observed in each treatment condition (e.g., high v. low
       drug dose, different instructions…)
Or:
     Longitudinal / time sampling: measure D.V. over
      multiple time periods (Cohort studies)
    Psychology 242, Dr. McKirnan                Exam #3 study guide           Home
                                                                      Back          Next
                                                                              Page
    Psychology 242
     Introduction               Within subjects designs; own control, 3                                     69
     to Research




 Repeated measures / randomized block design
 Group 1              Baseline                Control                     Measure2      M3              M4..
                      Measure                 Condition

 Group 2              Baseline                Experimental                Measure2      M3              M4..
                      Measure                 Condition


Assignment                                   Treatment. Primary
Randomly or via                              Independent Variable. Control
 natural “blocks”                            group may receive Placebo.


     Baseline assessment                                             Follow-up. Long-term
        prior to intervention or                                      assessment of outcome or
        experimental condition.                                       Dependent Variable. Time may
                                                                      represent 2nd Independent
                                                                      Variable.

 Psychology 242, Dr. McKirnan                  Exam #3 study guide                               Home
                                                                                         Back          Next
                                                                                                 Page
                                Psychology 242                                                                                 70
                                 Introduction
                                 to Research
                                                 Unprotected anal intercourse (UAI) with HIV positive and
                                                 unknown sero-status partners among MSM, by study visit
RUN THIS SLIDE TO SEE THE INFO.
                                      60
                                      60
                                                                                                  Blocking
                                      50
                                      50                                                            Blocking
                                                                                                  variable

                                                                                                    variable
                                      40
                                      40
                                                                                                                PEP users
                                                                                                                PEP users
                                      30
                                      30
   Participants reporting (%)




                                                                                                                Non users
                                                                                                                Non users
                                      20
                                      20
                                                     Non-users at the end
                                                      Non-users safer
                                                     All men get of PEP
                                      10
                                      10             of thetime are safest                            Repeated
                                                      are study
                                                     overgenerally safer
                                                        (an additive effect)                          Measure
                                         0
                                         0
                                                 0
                                                 0       6
                                                         6     12
                                                               12    18
                                                                     18    24
                                                                           24     30
                                                                                  30           36
                                                                                               36
                                                                                                                 OR*(CI)        p
                                                             Month of study visit                   PEP         1.7(1.2-2.2) .001
                                                                                                    Visit       0.97( .96-.99) .001
 *Adjusted forDr. McKirnan site, drug use, Withineducation
  Psychology 242,
                  age, study               and Subjects Designs
                                                                                                                    Home
                                                                                                            Back          Next
                                                                                                                    Page
    Psychology 242                                                                71
     Introduction
     to Research               Summary

    Within subjects designs somewhat common in
     psychological research;
             Own control designs: create strong contrast for IV
             Eliminate problems in creating experimental v.
              control groups.
    Very common in bio-medical or public health
     studies;
             Most clinical studies are longitudinal; participants
              followed over time
             Intervention or experimental treatment is I.V. #1
              (blocking or grouping variable).
             Stability over time is I.V. # 2 (repeated measure)
Psychology 242, Dr. McKirnan         Within Subjects Designs           Home
                                                               Back          Next
                                                                       Page

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:9
posted:8/9/2012
language:
pages:71