The Literature
Analyzing and Critiquing the
Literature
A review of literature should be integrated and critical.
Each study should be reviewed individually, summarized
and then juxtaposed with other studies in the area
showing how this research leads to the question in your
specific research.
The variables under study in your research should
emerge from the literature.
The design of the study should emerge from the review
of literature.
This chapter should lead the reader to an understanding
of how your study will contribute to the literature.
What is meta-analysis?
Meta-analysis is a statistical approach to
understanding the literature.
Meta-analysis uses effect size statistics to
evaluate the literature.
Criticisms of Meta-analysis
Coding of the data
Reporting of significant effects only or
lack of reporting of effects that are not
statistically significant
Descriptive Research Methods
Scientific Method
Decide on a problem
Gather facts to refine the problem
(narrow definition is best)
Develop hypotheses (through induction)
Test hypotheses
Definition of Descriptive Research
Descriptive research describes the
present status of people, attitudes, and
progress.
Forms of Descriptive Research
Ex post facto (after the fact)
Case studies
Correlation studies
Developmental studies
Survey studies
Sampling Techniques
Good sampling techniques are required of
all researchers.
Population
A population is defined as all members
that are described by the characteristics
selected by the experimenter.
◦ All students at SJSU.
◦ All women students at SJSU.
◦ All Kinesiology majors.
◦ All MA sport management students.
◦ All students in KIN 250.
Sample
A sample is a portion of the defined
population.
Different types of Samples
Simple random sample
Systematic sample
Stratified sample
Cluster sample
Proportional sample
Simple Random Sample
of Random Numbers
Table
Choosing numbers out of a hat
Systematic Sample
Systematically selects every nth person
Stratified Sample
A stratified sample assures a random
sample, however the sample has equal
numbers within a particular characteristic.
Cluster Sample
A sample is chosen because it is difficult
to sample the entire population, e.g.,
choosing all members of a particular class
rather than individuals.
A cluster sample is often easier and less
costly, but generalizability is limited
because of an N of 1.
Proportional Sample
Proportion out groups that you might
want in your sample.
The proportions should be logically based
in the literature.
Ex Post Facto Research
Ex post facto research is sometimes
called causal comparative research.
Ex post facto research is research that
takes place after the fact.
Often ex post facto research is used to
explain something in the present from
data collected sometime in the past.
Problems with Ex Post Facto Designs
Isthe cause that you hypothesize correct?
Many causes may be interrelated or the
result of more than one variable
interacting.
Extraneous variables may not be
accounted for.
Participants are self-selected. What puts
them into these categories?
Case Studies
Case studies are usually an examination
into one element of a population, e.g., one
school district, one school, one research
class, one person.
Case studies are often conducted in social
work and counseling for diagnosis and
recommendation purposes.
Case Studies
Observe – take notes on events and their
relationships by their location in time and
space.
Subjects’ recall – personal documents,
diaries, and letters
Measures may be physical, sociological, or
psychological
Case Studies
Must be careful about generalization
Data are limited to one unit
Case studies can be qualitative or
quantitative
Writing Hypotheses
Directional (H1)
◦ Physical activity program will affect body
composition such that physical activity
individuals will lose more fat than sedentary
individuals.
Null (HO)
◦ Physical activity will not affect body
composition.
Alternative
◦ Physical activity will affect body composition.
The researcher wants to accept the
directional hypothesis. The hypothesis is
accepted if the probability of finding a
statistically significant effect by chance
alone is less than 5 times in100 (p<.05).
The researcher wants to reject the null
hypothesis. The hypothesis is rejected if
the probability of finding a statistically
significant effect by chance alone is less
than 5 times in 100 (p<.05).
Hypothesis Testing
HO True HO False
Accept Correct decision Type II error
Reject Type1 error Correct decision
() ()
•Alpha () means the probability level acceptable for
statistical significance in a study. Type I error also means
the probability of rejecting a true null hypothesis.
•Beta ()) measures the Type II error. Type II error means
the probability of accepting a false null hypothesis.
Hypothesis Testing
Directional Null
H1 True H1 False HO True HO False
Accept Correct Type1 error Correct Type II error
decision () decision ()
Reject Type II error Correct Type1 error Correct
() decision () decision
•Alpha () means the probability level acceptable for
statistical significance in a study. Type I error also means
the probability of rejecting a true null hypothesis.
•Beta ()) measures the Type II error. Type II error means
the probability of accepting a false null hypothesis.
Correlational Studies
Correlational studiesexamine the
relationship between two or more
variables.
Correlations examine how variables
covary together.
Positive Correlation
From: Rothstein, A. L. (1985). Research Design & Statistics for Physical
Education. Prentice-Hall, Inc.: Englewood Cliffs, NJ.
Types of Correlations
Pearson correlation
Spearman correlation
Partialcorrelation
Multiple correlation
Multiple regression
Developmental Studies
Developmental studies are concerned
with changes that take place as a function
of time.
Kinesiology
Growth and Development
◦ Physical milestones
Motor Development
◦ Studies of the patterns of movement
◦ Motor learning across the lifespan
◦ Exercise physiology across the lifespan
◦ Biomechanics across the lifespan
◦ Sociocultural effects across the lifespan
Types of Developmental Studies
Longitudinal studies
◦ Long term, collect data from the same subject
over a number of years
Cross sectional studies
◦ Short term, usually 6 months or less and all
data are collected
Survey Design
Surveys are used to gather extensive
amounts of information for large groups
of individuals in short time spans.
Survey Designs
Public opinion
Attitudes
Achievement
Educational
Governmental
Industrial
Political
Survey Design
A good survey
◦ Wide scope
◦ Accuracy
◦ Ease of data collection
A bad survey
◦ Superficial
◦ Poor return rate
◦ Poor survey instruments
Things to Keep in Mind When
Creating Surveys
Define terms
Avoid statements with double
implications
Avoid leading questions
Beware of double negatives
Things to Keep in Mind When
Creating Surveys
Clearly identify the survey purpose
Outline the field of study
Avoid overlapping questions
Order questions in a logical format
◦ Simple to complex
Make sure questions are clear
◦ Eliminate ambiguities
◦ Eliminate all grammatical errors
Pre-code data for computation
Profile of Mood States
Bem Sex Role Inventory and the
Personality Attributes
Questionnaire
Bem (1974) constructed the Bem Sex Role
Inventory (BSRI) – It presented from 20 adjectives
viewed as desirable for men, 20 adjectives viewed as
desirable for women, and 20 gender-neutral adjectives.
Spence and Helmreich constructed the Personal
Attributes Questionnaire (PAQ).
Examples of Popular and Widely
Used Instruments
Beck Depression Inventory
Myers Briggs Indicator
Keirsey Temperment Sorter
Minnesota Multiphasic Personality Inventory
Leadership Opinion Questionnaire
Wechsler Adult Intelligence Scale IV (WAIS IV)
Kolb's Learning Styles Inventory
Measures of Self-Esteem; Self Efficacy
Dishman’s Exercise Adherence Measure
LeUnes,A.D. (2002). Bibliography on
psychological tests used in research and
practice in sport and exercise psychology.
Lewiston, N.Y. : E. Mellen Press.
See: HaPI - Health and Psychosocial
Instruments