|
|
| Ron Jones Bio |
| CorporateWellness |
| Coach & Train |
| Exercise Library |
| Handouts |
| Health & Fitness |
| KETTLEBELLS |
| Products by RJ |
| Site Map |
|
|
|
|
|
|
-95.jpg){kind=small}
.jpg){kind=small}
KIN
605: Exam #1 Study Guide
(Last Updated On: 10-23-01)
NOTE:
This document was originally composed on MS Word. When it gets inserted into
the web program Front Page, the formatting gets funky. The words are still
the same; it just doesn't "look as good."
Text: Research Methods in Physical Activity by Thomas & Nelson
*These
notes define the bulleted topics in order of appearance from the “Expectations
for Exam 1” outline posted 10-20-01 on Blackboard.Com
Sequence of Steps for Developing a Sound Research Study: (Chapter #1)
- Developing
the Problem (Define & Delimiting) (p.12)
- Researcher needs to be very specific about what is to be studied and to what extent it will be studied.
- Need to identify the independent variable and dependent variable i.e. the researcher must define exactly what will be studied and what will be the measured effect then the experimental design can be determined.
·
Independent Variable:
The part of the experiment that the researcher is manipulating; also
called the experimental or treatment variable.
o
The one the researcher is trying
to understand.
o
A Categorical Variable
(also called a Moderator Variable) is a kind of independent variable except that
it cannot be manipulated, for example, age, race, or sex.
(This contradicts what Dr. M said in the 10-16 lecture where she stated
this was something “controlled” by experimenter i.e. temperature was the
example I wrote down.)
·
Dependent Variable:
Depends on something i.e. what you are manipulating.
o
The effect or consequent of the
IV; also called the yield.
- Formulating
the Hypothesis (p.12-13)
- When a person sets out to do a study, they generally have an idea what the outcome or anticipated solution will be based on some theoretical construct, results of previous studies, or past observations or experience.
·
Hypothesis:
The anticipated outcome or expected results
of a study or experiment.
o
Hypothesis must be testable and
designed so it can be either supported or refuted.
o
Deduced
statement always in present tense that specifies the direction the data will
take when the study is conducted.
o
Researcher’s
best guess based on reading and thinking at specifying the direction that he or
she believes the data will go.
o
Deduced
from theory or induced from other empirical studies and real world observations.
o
Based on
logical reasoning and, when predictive of the study’s outcome, are labeled
research hypotheses.
- Gathering
the Data (p. 13)
- Researcher must decide on proper methods of acquiring necessary data to be used in testing the research hypothesis i.e. reliability of testing instruments, controls employed, overall objectivity and precision of data gathering process.
- Related to the research design and controls are internal and external validity.
·
Internal Validity:
The extent to which the results of a study can be attributed to the
treatments used in the study.
·
External Validity:
The generalizability of the results of a study i.e. to what extent the
results apply to the “real world.”
- Analyzing
& Interpreting Results (p. 13)
- Statistical analysis is used here along with considerable knowledge, experience, and insight.
- Researcher must provide evidence for the “support or rejection” of hypothesis.
- Inductive reasoning is employed here whereas deductive reasoning is primarily used in the problem statement.
·
Inductive Reasoning:
(p. 27-29) Individual observations are tied together into specific
hypotheses that are grouped into more general explanations that are then united
into theory.
o
Many individual studies are
needed to move from the specific hypotheses.
o
Deductive Reasoning moves down
from the theory to specific hypotheses.
·
Deductive Reasoning:
(p. 29) Moves from a theoretical explanation of events to specific
hypotheses that are tested against (or compared with) reality to evaluate
whether the hypotheses are correct.
Writing
a Clear Problem Statement, Null Hypothesis, & Research Hypothesis: (Chapters
#2,3)
1.
Problem Statement of Research: (p.
31, 52-54)
·
Follows the
introduction unless a literature review is included in the introduction.
Then a brief problem statement should appear soon in the intro section.
·
Should be
succinct (compact precise expression without wasted words).
·
Specify
research questions you are asking i.e. by carefully defining the research
problem, the researcher can keep the literature search within reasonable limits.
·
Try to specify what research
questions you are asking.
·
ID
variables in study including the IV, DV, and categorical variables (if any).
Some control variables can also be identified here.
·
An
effectively constructed introduction leads smoothly to the study’s purpose.
This is expressed as the statement of the problem and should be as clear
and concise as the subpurposes, or variables, allow it to be i.e. “very
succinct.”
o
Example:
2.
Null Hypothesis (p.
54,135)
·
Null Hypothesis:
Hypothesis used primarily in the statistical test for the reliability of
the results that says that there are no differences among treatments (or no
relationships among variables).
o
Example:
Self-efficacy will not influence exercise adherence.
3.
Research Hypothesis (p. 54)
·
Hypothesis:
The anticipated outcome or expected results of a study or experiment.
- When a person sets out to do a study, they generally have an idea what the outcome or anticipated solution will be based on some theoretical construct, results of previous studies, or past observations or experience.
o
Hypothesis must be testable and
designed so it can be either supported or refuted.
o
Deduced
statement always in present tense that specifies the direction the data will
take when the study is conducted.
o
Researcher’s
best guess based on reading and thinking at specifying the direction that he or
she believes the data will go.
o
Deduced
from theory or induced from other empirical studies and real world observations.
o
Based on
logical reasoning and, when predictive of the study’s outcome, are labeled
research hypotheses.
o
Example:
It is hypothesized that self-efficacy is a strong determinant in
exercise adherence to the point of exercise.
Recognize
the Difference Between Inductive & Deductive Development (Chapter #2)
1.
Inductive Reasoning: (p. 27-29)
·
Individual observations are tied together into specific hypotheses that are grouped into more
general explanations that are then united into theory.
·
Many
individual studies are needed to move from the specific hypotheses.
·
Deductive
Reasoning moves down from the theory to specific hypotheses.
2.
Deductive Reasoning: (p. 29)
·
Moves from
a theoretical explanation of
events to specific hypotheses that are tested against (or compared with) reality
to evaluate whether the hypotheses are correct.
Steps
of a Literature Review (Chapter #2)
·
Literature Review:
(p. 29-48) 3 basic parts of actual literature review are: intro, body, summary
and conclusions that demonstrate that your problem needs investigation and that
you have considered the value of relevant past research in developing your
hypothesis and methods. Allows you
to: identify the problem, develop hypotheses, and develop the method.
·
Don’t
isolate the studies; put all individual studies together for a central finding
or position.
·
Major part
of developing research problem is reading what has already been published about
the problem. This helps to
“identify” the specific problem.
·
Introduction:
Should explain the purpose of the RV and how and why of its organization.
·
Body:
Should be organized around important topics.
·
Summary:
Summarize important implications and suggest directions for future research.
Steps of Lit Review
1.
Write the Problem Statement (p. 31)
·
Specify
which questions you are asking; keep lit search to reasonable limits.
2.
Consult Secondary Sources (p. 31-32)
·
If you are
not knowledgeable about the topic, you should consult: textbooks, encyclopedias,
or research reviews.
3.
Determine Descriptors (p.
32)
·
These are
terms that help to locate sources pertaining to the topic. Note that various databases have their own descriptors for
topics.
4.
Search Preliminary Sources (p.
32-33)
·
Use
preliminary (or general) sources to find primary sources via computer-aided
searches. These primarily consist
of abstracts (from conventions, summits, etc.) and indexes (like PE Index, Index
Medicus, etc.) but could also be bibliographies, library information systems,
computer searches (like ERIC, Medline, etc.).
5.
Read & Record the
Literature (p. 40-42)
·
As a
researcher, you should note the following information on each study you read:
o
Statement of the problem (and
maybe hypotheses)
o
Characteristics of the
participants
o
Instruments and tests used
(including reliability and validity information if provided)
o
Testing procedures
o
IV and DV
o
Treatments applied to
participants (if an experimental study)
o
Design and statistical analyses
o
Findings
o
Questions raised for further
study
o
Citations to other relevant
studies not located.
·
It is also
advised that you: photocopy if relevant to your study, write complete citation
on the title page if journal does not provide this.
6.
Write the Literature Review (p.
42-48)
·
3 Basic
Parts: Intro, Body, and Summary & Conclusions.
o
Intro:
Should explain the purpose of the review and the “how and why” of its
organization. Important to make the
intro interesting!
o
Body:
Should be organized around important topics and summarize important
implications and suggest directions for future research.
Reviews purpose is to demonstrate that your problem needs investigation
and that you have considered the value of relevant past research in developing
Hypoth and methods. Needs to be
clear and interesting.
o Summary & Conclusions:
Define
Limitations, Delimitations, Assumptions, & Significance of a Study (Chapter
#3)
1.
Limitations: (p.
56,57)
·
Possible
shortcoming or influence that either cannot be controlled or is the result of
the delimitations imposed by the investigator.
·
Can’t control these i.e.
weather, attitudes, truthfulness of N, etc.
·
Researcher
tries to reduce limitations that might stem from faulty procedures.
2.
Delimitations: (p.
56,57)
·
Limitation
imposed by the researcher in the scope of the study; a choice the researcher
makes to define a workable research problem.
·
Limits of
gender, age, etc.
3.
Assumptions: (p.
56)
·
Assumption
that certain conditions exist and that the particular behaviors in question can
be observed and measured.
·
Without
certain fundamental assumptions or premises, a study could not proceed or take
place.
4.
Justifying the Significance of a Study (p.
57-59)
·
So what
good is your study and how is this of any importance to your profession?
·
Worth of
research study is judged by whether it is basic or applied science.
·
Basic Science Research:
Does not have immediate social significance.
May have limited direct application but in the researcher has careful
control of the conditions.
·
Applied Research:
Addresses immediate problems for improving practice.
The worth of applied research must be evaluated on the basis of its
contribution to the solution of some immediate problem.
Has direct value to practioners but in the researcher has limited control
over the research setting
Distinguish
Between Dependent, Independent, & Control Variables (Chapter #1,3)
1.
Dependent Variable:
(p. 12) Depends on something i.e. what you are manipulating.
·
The effect
or consequent of the IV; also called the yield.
2.
Independent Variable:
(p. 12) The part of the experiment that the researcher is manipulating;
also called the experimental or treatment variable.
·
The one the
researcher is trying to understand.
·
A Categorical
Variable (also called a Moderator Variable) is
a kind of independent variable except that it cannot be manipulated, for
example, age, race, or sex.
3.
Control Variables:
(p. 53) Factor that could possibly influence the results and that is kept
out of the study.
Specify the Institutional Review
Process & Steps for Protecting Human & Animal Subjects/Participants
(Chapter #5)
1.
Institutional Review Process (Blackboard.Com CSUN Link/Week #2)
·
All students and faculty conducting research that utilizes humans or
animals must complete a protocol form and submit it to ORSP (Office of Research and Sponsored Projects).
The SACPHS
(Standing Advisory Committee for the Protection of Human Subjects) and IACUC (Institutional Animal Care and Use Committee) then
review and approve all projects to insure that no undue risks are involved.
2.
Protecting Humans (p.
76, 84-86)
·
Human Subjects Committee:
(p. 76) Sets ethical standards and guidelines for research participants.
·
Must balance degree of risk, participants’ rights, and potential value
of research in contributing to knowledge, development of technology, and to
improvement of people’s lives.
·
What
Research Participants Should Expect: (p. 84)
o
Right
to privacy or nonparticipation (don’t ask for unnecessary info; obtain direct
consent)
o
Right
to remain anonymous (focused on group data and use ID numbers)
o
Right
to expect experimenter responsibility (inform of who has direct access to
original data)
·
Informed Consent:
(p. 85) The critical elements are:
o
Fair explanation of procedure to
be followed, including an ID of those that are experimental.
o
Description of the attendant
discomforts and risks.
o
Description of the benefits to be
expected.
o
Disclosure of appropriate
alternative procedures that would be advantageous for the participant.
o
An offer to answer any inquires
concerning the procedures.
o
An instruction that the
participant is free to withdraw consent and to discontinue participation in the
project or activity at any time.
o
No exculpatory language through
which the participant is made to waive, or appear to waive, any legal right or
to release the institution or its agents from liability or negligence.
3.
Animal Rights (p.
86,87)
·
If animals are well treated, their use is justified if it falls into one
of five categories:
o
Drug
testing such as the development and testing of AIDS drugs
o
Animal
models of disease such as development of animal models of arthritis, diabetes,
iron deficiency, auto-immune dysfunction, and aging
o
Basic
research, focused on examining and elucidating mechanisms at a level of definition not possible in human models.
o
Education
of undergrad and grad students in labs and lectures with experience and info
gained from use of animal models
o
Development
of surgical techniques used extensively in training of med students and testing
of new surgical devices and procedures.
·
If use means animals are incapacitated, they must be sacrificed
humanely.
·
If used for exercise science, institutions require adherence to the
Guide for the Care and Use of Laboratory Animals published by US Dept. HHS as
detailed by the Animal Welfare Act and most institutions support the rules and
procedures for recommended care of lab animals as outlined by the American
Association for Accreditation of Laboratory Animal Care.
Consent Form Mandatory Critical Parts
(Chapter #5)
Informed
Consent: (p.
85) The critical elements are:
o
Fair explanation of procedure to
be followed, including an ID of those that are experimental.
o
Description of the attendant
discomforts and risks.
o
Description of the benefits to be
expected.
o
Disclosure of appropriate
alternative procedures that would be advantageous for the participant.
o
An offer to answer any inquires
concerning the procedures.
o
An instruction that the
participant is free to withdraw consent and to discontinue participation in the
project or activity at any time.
o
No exculpatory language through
which the participant is made to waive, or appear to waive, any legal right or
to release the institution or its agents from liability or negligence.
Conduct
Appropriate Randomization Methods (Chapter #6)
·
Randomization Methods (p.315)
o
I do
not understand this concept and have no class notes that make sense to me.
Please explain this to me if you understand how to do it!
- Random Sample: A sample taken from a population where every member of the population has an equal chance of being selected in the sample.
- Random Number Generator: www.randomizer.org
Types
of Research (Chapters # 11-18)
1.
Historical
(#11)
(p. 203-216)
- Basic
Methods are:
o
Must have access to
libraries and archives.
o
Labor to identify and
understand patterns in the evidence.
o
Expect to construct
meaningful generalizations from historical evidence or data.
- Good
history has a meaningful paradigm or framework.
o
Paradigm:
Outstandingly clear or typical example or archetype; framework of
a scientific school or discipline within which theories, laws, and
generalizations and the experiments performed in support of them are formulated.
The literature assists researcher to shape paradigm that directs or
constrains the research.
o
Paradigm Crisis Phenomenon: Development of
discrepancies in a paradigm leading to proposals of a new paradigm that better
explains the data.
·
Primary Sources: (p. 207)
Firsthand source of data in research; the original study and usually refers to
journal articles.
- Secondary
Sources: (p. 205,206)
Existing body of books, articles, and other media that are histories. Note “direct” and “ancillary” literatures.
- Main
Focus: (p. 216) Working
with evidence which includes external and internal criticism of the primary
sources.
- Internal
Criticism: (p. 216)
Deals with trustworthiness of the information found in primary source of
historical research.
- External
Criticism: (p. 216)
Establishes the authenticity of a primary source.
- Must
always seek to interpret in the context in which the person lived or in
which the event took place.
2.
Philosophic
(#2,12)
- Fundamental
Goal: (p. 219, 220) To
examine reality by using reflective procedures and not the empirical tools
of science. Philosophers
reflect on exercise and utilize such things as ideas and ideals, meaning,
lived experience, values, logical relationships, and reasons in attempting
to shed some light on it. This
allows philosophers to answer questions that empirical methodologies cannot
solve.
- Locating
Research Problem:
(p. 222,223) This is easy to do, but good philosophic research begins with
definitions, descriptions, clarifications, and disclaimers, particularly
when the object of analysis is too large or excessively vague.
·
Inductive Reasoning:
(p. 27-29, 224) Individual observations are tied together into specific
hypotheses that are grouped into more general explanations that are then united
into theory.
o
Thought that moves from a limited
number of specific observations to general conclusions about the thing or class
of thing that was observed. It
relies on the power of reason to identity common elements or similarities at an
abstract level.
o
Many
individual studies are needed to move from the specific hypotheses.
o
Deductive Reasoning moves down
from the theory to specific hypotheses.
·
Deductive Reasoning:
(p. 29, 225) Moves from a theoretical explanation of events to specific
hypotheses that are tested against (or compared with) reality to evaluate
whether the hypotheses are correct.
o
A companion technique of
inductive thinking, and many philosophers adroitly (skill, clever, resourceful)
and spontaneously intermix the two. Deduction
requires intellectual movement in the opposite direction from induction.
Whereas inductive thinking has philosophers working from particulars to
general abstractions, deduction has philosophers start with general claims to
see what particulars follow.
3.
Research
Synthesis/Meta Analysis (#13)
- We
have not been assigned this reading.
4.
Survey
(#14)
- Survey
Defined: (p. 261)
Technique of descriptive research that seeks to determine present practices
or opinions of a specified population; can take the form of a questionnaire,
interview, or normative survey.
5.
Other
Descriptive Methods (#15)
- Developmental
Research: (p. 277) Study
of changes in behavior across years.
- Longitudinal
Studies: (p. 277)
Research in which the same participants are studied over a period of years.
- Cross-Sectional
Study: (p. 278) Research
in which samples of participants from different age groups are selected in
order to assess the effects of maturation.
- Cohort
Problem: (p. 277)
Problem in cross-sectional design of whether all the age groups are really
from the same population.
- Case
Study: (p. 280) Form of
descriptive research in which a single case is studied in depth to reach a
greater understanding about other similar cases.
- Job
Analysis:
(p. 282) Type of case study that determines the nature of a particular job
and the types of training, preparation, skills, and attitudes necessary for
success in the job.
- Observational
Research:
(p. 289) Descriptive technique that involves the qualitative and
quantitative analysis of observed behaviors and attempts to study what a
person “actually does” as opposed to the survey method that relies on
self reports about how a person behaves.
- Correlation
Research:
(p. 288) Research that explores relationships among variables and that
sometimes involves prediction of a criterion variable.
6.
Epidemiology
(#16)
- Epidemiology
Defined: (p. 293) The
study of the distribution and determinants of health related states or
events in specified populations and the application of this study to the
control of health problems.
- Major
Components of Epi Research:
(p. 293)
o
Distribution:
Frequency (prevalence, incidence, mortality rate) and patterns (person, place,
and time).
o
Determinants:
Defined characteristics associated with change in health.
o
Application:
Translation i.e. knowledge to practice.
- Epi
Study Designs:
(p. 296-305)
o
Cohort:
A specified group of individuals who are followed over a period of time.
o
Cross-Sectional
Designs: (p. 297) Perhaps
the most frequently conducted type of study examining the relationship between
physical activity and health outcomes where “known risk factors for a
disease” are used as intermediate end points.
o
Ecological Designs:
(p. 298) Use existing data sources for both exposure and disease outcomes to
compare and contrast rates of disease by specific characteristics of an entire
population such as with census data, vital stats records, employment records, or
national figures for health info such as food consumption.
o
Analytical
Designs: (p. 299) These case
control and cohort studies are designed to test specific hypotheses regarding
causal links between various exposures and mortality and incidence outcomes
using purely observational methods.
o
Case
Control Studies: (p. 302)
Like the cohort design, these aim to identify factors that are causally related
to a disease outcome where a population of individuals with disease (i.e. cases)
and without disease (i.e. controls) is recruited into the study over the same
period of time.
o
Recall
Bias: (p. 303) Systematic
errors introduced by differences in the recall accuracy between comparison
groups (i.e. between cases and controls).
o
Selection Bias:
(p. 304) Systematic errors introduced by difference in the characteristics of
participants entering and not entering a study.
o
Bias:
(p. 304) Systematic deviation of a calculated (estimated) value from the true
value.
o
Confounding Factor:
(p. 304) Factor that obscures the true relationship between an exposure-outcome
association of interest.
7.
Experimental
& Quasi-Experimental (#17)
- Most
KIN studies today are this type i.e. lab studies.
Some surveys are done with recreation and some historical studies are
done in pedagogy.
- Observe
some outcome variable.
- This
is truly “experimental” and not just observe and describe like
descriptive surveys. Surveys have a lot of room for bias.
Validity “can be” valid but are you truly getting to the true
cause with a survey? The “true cause” can only be found by experimental
design.
- Attempts
to establish cause-and-effect relationships (need good theoretical
framework) where an independent variable is manipulated to judge its effect
on a dependent variable when three criteria is present:
o
1-Cause must precede the
effect in time.
o
2-Cause and effect must
be correlated with each other.
o
3-Correlation between
cause and effect cannot be explained by another variable.
- Can
never say “cause and effect” with just one study.
Need hundreds of studies to draw this conclusion.
- Key
Terms (10-16-01 class
lecture)
o
IV:
What you manipulate.
o
DV:
What is not manipulated.
o
Categorical Variable:
Grouping variables if subjects broken into groups.
o
Control Variable:
Things controlled by experimenter.
o
Extraneous Variable:
Over and above and usually not directly related to IV and DV although can be
brought into the study.
- Distinguishing
Between Types of Validity: (10-16-01
class lecture)
o
Internal Validity:
Did treatments (IV) cause the change in outcome (DV)? IV is the ability of the
researcher to capture “what is really there.”
o
External Validity:
(p. 351) To what populations, settings, or treatments can the outcome be
generalized? When you control EV,
you remove the “real world” value that is sometimes needed in order to
gradually work into real world applications.
What is more important? More
control or more real world?
Ø
There is a trade off
between internal validity and external validity.
Ø
One
of strongest arguments for EV of qualitative research is user generalizability
i.e. when reader of study evaluates the descriptions and analysis and
determines what things apply to his or her situation.
8.
Qualitative
(#18)
- Defined:
(p. 333) Strives to develop hypotheses from the observations i.e. emphasizes
“induction” as opposed to quantitative research that emphasizes
deduction.
- Researcher
is primary instrument for data collection and analysis and researcher’s
sensitivity and perception are crucial.
- Methods:
(p. 350) Field observations, case studies, ethnography, narrative reports
gathered in natural settings.
- Does
not have preconceived hypotheses i.e. inductive reasoning is stressed
whereby the researcher develops hypotheses from observations where
participant rapport and trust are critical.
- Data
Analysis:
(p. 351) Involves organizing, abstracting, integrating, and synthesizing
where analytical narrative is the foundation i.e. gives the reader a sense
of being present and conveys a holistic meaning to the situation.
- Triangulation
of Data: (p. 351) Used
to establish validity and reliability.
- Grounded
Theory: (p. 351) Theory
that evolves from data.
- Internal
and external validity are important concepts.
Describe and Build a Sound &
Appropriate Research Design (Chapter #17)
Types of Designs (p. 318-329)
1.
Preexperimental Designs:
(p. 318) One of three types of research design that control very few of
the sources of invalidity and that do
not have random assignments
of participants to groups:
·
one-shot study (T, O)
·
one-group pretest—posttest design (O1, T, O2, Stats Analysis?)
·
static group comparison (T, O1/O2>Stats Analysis?)
2.
True Experimental
Designs: (p. 319) Any design used
in experimental research in which groups are randomly formed and that controls
most sources of invalidity.
·
Randomized-groups design
·
Pretest-posttest randomized-groups design
·
Solomon four-group design
3.
Quasi-Experimental
Designs:
(p. 323) Research designs in which the experimenter tries to fit the
design to real-world settings while still controlling as many of the threats to
internal validity as possible. Tries
to be as “real world” as possible and not just describe the
environment—but actually “manipulate” the environment.
Often used when it is difficult or impossible to use true experimental
designs or when a true experimental design significantly limits external
validity.
·
Time-series design
·
Reversal design
·
Nonequivalent-control-group design
·
Ex
post facto design
(something happens and then you observe)
o
Experimenter
was NOT in control of the experiment.
o
Variables
are NOT controlled.
o
Not a
strong method but sometimes useful i.e. NYC attacks and how they affect the
American psyche.
·
Switched-replication design (when treatment occurs might affect outcome)
·
Single-subject design
Reducing Risks of Internal &
External Validity (Chapter #17)
Threats to Internal Validity (p. 315-317, 10-16-01 class notes)
·
History:
Events not part of treatment.
·
Maturation:
Events due to passage of time.
·
Testing:
Effects of more than one test administration.
·
Instrumentation:
Change in calibration of measurements.
·
Statistical
Regression:
Selected based on extreme score.
·
Selection
Biases:
Non-random participant selection.
·
Experimental
Mortality:
Loss of participants.
·
Selection-Maturation
Interaction:
Passage of time influencing groups differently.
·
Expectancy:
Influence of experiments on participants i.e. they try to please the
experimenter. The attitiude of
researcher can greatly affect outcome, etc.
A researcher tries to treat each participant identically but this is hard
to do. You need rigid guidelines
i.e. don’t talk politics, religion, or about emotional issues, etc.
Controlling Threats to Internal Validity (10-16-01
class notes)
1.
Randomization:
You can never get a true “real” random sample with the entire N
population.
·
Matched pairs (not matched groups)
·
Randomizing treatments or counter balancing i.e. random assignments like
cross over design which is the most often used type in KIN.
Need to make sure numbers and groups are balanced i.e. male vs. female.
2.
Placebos:
3.
Blind Setups:
Participants don’t know what they are getting i.e. placebo vs. green tea vs. creatine
ergo supplement.
4.
Double Blind Setup: Participants
and experimenter don’t know what is being given i.e. an independent
experimenter numbers, labels, and administers the samples.
Threats to External Validity (10-16-01
class notes)
1.
Reactive and Interactive
Effects of Testing: Pretest may
make participants sensitive to treatment.
2.
Interaction of Selection
Biases and Treatment: Treatment may
work only on participants selected on specific characteristics.
3.
Reactive Effects of
Experimenter Arrangements: Setting
constraints may influence generalizability i.e. crossover design where half
doesn’t receive treatment and half does receive treatment then switch and
repeat (this allows the means to get averaged).
4.
Multiple
Treatment Interference: One treatment
may influence the next treatment.
Controlling Threats to External Validity (p.
317,318)
·
Generally controlled by selecting participants, treatments, experimental
situation, and tests to represent some larger population.
Random selection is the key to controlling most threats to EV.
RonJones.Org
| Back to Top | Back to CSUN Homepage
| Site Map
Be Strong.