Professional Documents
Culture Documents
CLINICAL RESEARCH
• a structured process of investigating facts
and theories and exploring connections
between them
EXAMPLE:
In an elderly patient two-months post stroke, is
partial body weight-supported treadmill training
more effective than traditional gait training with
full weight-bearing for improving walking speed,
endurance and balance?
LEVELS OF
EVIDENCE
LEVELS OF EVIDENCE
SYSTEMATIC REVIEW
◉Comprehensive review of the medical
literature
◉Advantage:
Increase power by increasing sample size
Improves estimates of effect size
Resolves uncertainty
Improve generalizability of findings
META-ANALYSIS
RANDOMIZED CONTROLLED
TRIAL (RCT)
◉Experimental research
◉Uses randomization
◉Relative effect of a specific intervention compared
to a control condition
Experi
Control
mental
group
group
RANDOMIZED CONTROLLED
TRIAL (RCT)
LONGITUDINAL STUDY
◉Repeated observations of the same
variables over long periods of time
◉Advantage: can accumulate data through
documentation of growth and change on
the same individual
◉Collected in time sequence
COHORT STUDY
◉Longitudinal, observational study
\ A-B design
◉2 phases
◉Pretreatment/baseline phase followed by
treatment phase
SINGLE SUBJECT
EXPERIMENTAL DESIGN
\ A-B-A design (multiple baseline design)
◉3 phases
◉Baseline phase -> treatment phase -> 2nd
baseline phase
◉Qualitative data
◉Quantitative data
SCALES OF MEASUREMENT
NOIR
•Nominal
•Ordinal
•Interval
•Ratio
SAMPLING
PROBABILITY SAMPLING
•Simple random sampling - subjects have an
equal chance of being selected
PROBABILITY SAMPLING
•Systematic sampling- taking every nth
subject from the population
PROBABILITY SAMPLING
•Stratified random sampling - population is
divided into homogenous subgroups/strata
PROBABILITY SAMPLING
•Cluster sampling- population is divided into clusters
NON-PROBABILITY SAMPLING
•Convenience sampling - Subjects are
chosen based on their availability
\ Content Validity
◉measurement reflects the meaningful
elements of a concept
VALIDITY
\ Construct Validity
◉To measure abstract concept
◉A theoretical construct is measured by a test
or measurement.
a. Convergent
b. Divergent
VALIDITY
\ Concurrent Validity
◉Comparing measurements to a gold standard
\ Predictive Validity
◉Measurement is considered to be valid
because it is predictive of a future
behavior/event
VALIDITY
\ Internal Validity
• the intervention being evaluated (independent
variable) is the cause of the outcome and not a
result of extraneous factors
\ External Validity
◉Results of the study are generalizable to the
population
THREATS TO
VALIDITY
THREATS TO VALIDITY
Sampling Bias:
• systematic sampling error
Experimenter bias
THREATS TO VALIDITY
Hawthorne effect:
• the subject's knowledge of participation
in an experiment influences the results
of a study.
Placebo effect:
• subjects respond to a dummy treatment
with positive effects
RELIABILITY
\ Intrarater reliability
• repeated measures made by the same
person over time
\ Interrater reliability
• measurements made by more than one
person
RELIABILITY
\ Test-retest reliability
• repeated measures made by the same
individual on separate occasions
\ Split-half reliability
• Test is split in half
• Reliability of the first half is compared to the
second half.
STATISTICAL
MEASURES OF
VALIDITY
STATISTICAL MEASURES OF
VALIDITY
\ Sensitivity
• Percentage of people who test positive for a
specific disease among a group of people who
have the disease
\ Specificity
• percentage of people who test negative for a
specific disease among a group of people who
do not have the disease
STATISTICAL MEASURES OF
VALIDITY
\ True Positive
• Individuals are correctly identified as
having the target condition
\ True Negative
• Individuals are correctly identified as
not having the target condition
STATISTICAL MEASURES OF
VALIDITY
\ False Negative
• Indicates that a person does not have a specific
disease or condition when the person does have
the disease or condition
\ False Positive
• Indicates that a person has a specific disease or
condition when the person actually does not have
the disease or condition
STATISTICAL MEASURES OF
VALIDITY
\ Negative Predictive value
• Ability of a diagnostic test to correctly determine
the proportion of patients without the disease
from all patients with negative results
\ Dependent variable
–Response or outcome assumed to be caused
by the effect of the independent variable
HYPOTHESIS
•A tentative statement to explain certain
observations or facts that requires further
experimentation to be verified
•False positive
•“mayabang”
TYPE II ERROR (BETA ERROR)
•Wrongly deciding to accept the null
hypothesis
•False negative
•“pa-humble/martyr”
DECREASING ERRORS:
•Increase sample size
•Normal- bell-
shaped curve
–mean, median and
mode are the same
NORMAL DISTRIBUTION
•68% of all values fall within one SD above and
below the mean
a. 63 to 65 inches
b. 54 to 74 inches
c. 39 to 89 inches
d. 59 to 69 inches
INFERENTIAL
STATISTICS
INFERENTIAL STATISTICS
•Use sample data to make inferences
about the population
Parametric Nonparametric
• Interval and ratio • Nominal or ordinal
data data
• Population is • Non-normal
distribution, small
normally sample size
distributed
• Do not assume
• Homogeneity of Homogeneity of
variance variance
STEPS
1. What are the independent and
dependent variables?
2. What is the scale of measurement of
the study variables?
3. How many samples/groups are in
the design?
WHAT ARE THE INDEPENDENT AND
DEPENDENT VARIABLES?
•Independent Variable
–Manipulated variable
•Dependent Variable
–Response variable
WHAT IS THE SCALE OF MEASUREMENT
OF THE STUDY VARIABLES?
•Parametric
–Interval/ratio
•Nonparametric
–Nominal, ordinal
PARAMETRIC
STATISTICS
T-TEST
•T-test assesses whether the means of two
groups are statistically different from each
other.
2 INDEPENDENT SAMPLE T-TEST
•used when testing 2 independent
groups
EXAMPLE
•Researchers compare the mean blood
pressure of patients who received a
new drug treatment vs. those who
received a placebo (a sugar pill).
PAIRED/DEPENDENT T-TEST
•Compares the difference between 2
matched samples
•Repeated measures or paired data
•Within subjects design
EXAMPLE
•A researcher compares the mean test
scores before (pre-test) and after (post-
test) the subjects completed in a test
preparation course. They want to see if our
test preparation course improved people's
score on the test.
ANALYSIS OF VARIANCE (ANOVA)
•Test the equality of means between 3 or
more populations by analyzing sample
variance
•(N) population distribution and
population variance is equal
ONE-WAY ANOVA
• Similar to independent t-test
• Sample data are separated into different
groups on the basis of 1 characteristic
EXAMPLE
•Researchers compares the mean level of
prejudice of first-years, sophomores,
juniors, and seniors. Mean level of
prejudice is our dependent variable, and
years in college is our independent
variable.
REPEATED-MEASURES ANOVA
•Uses the same subjects with every
condition of the research, including
the control
•Within subjects
EXAMPLE
•Suppose we are interested in whether the
effect of a weight loss method will weaken
over time. 30 subjects' weight information
were recorded 3 times. Their weight loss
after adopting this method for one month,
three months and half a year was stored in
a worksheet.
TWO-WAY ANOVA
•examines the influence of different
categorical independent variable on
one dependent variable
EXAMPLE
•Researchers are testing the depression level of
people who are taking anti-depressant drugs at
different dosages. They also want to know if
gender plays an influence in it.
0 mg 25 mg 50 mg
9 7 4
MEN
8 7 2
7 6 3
8 7 4
WOMEN
7 6 3
9 6 3
ANCOVA
•A parametric test used to compare two
or more treatment groups or conditions
while also controlling for the effects of
intervening variables
EXAMPLE
◉We wanted to compare the effect of two teaching
strategies on clinical performance of students in their first
year of clinical training. We hypothesize that training with
videotaped cases will be more effective than discussion and
reading groups. We randomly assigned 12 students to 2
groups. We are concerned however, that the students’
academic performance would be a potential confounding
factor in making this comparison, based on the assumption
that there is correlation between academic and clinical
performance. Therefore we would want to know if GPA is
evenly distributed.
PEARSON CORRELATION
•Measures the magnitude and direction
of the linear relationship of two
variables on the interval scale (or ratio)
NONPARAMETRIC
STATISTICS
CHI-SQUARE TEST
•Nonparametric test of significance
used to compare data in the form of
frequency counts occurring in two or
more mutually exclusive categories
EXAMPLE
•Polling responses from people of
different nationalities to see if one's
nationality affects the response
MANN-WHITNEY U TEST
•2 independent samples
•Ordinal data
EXAMPLE
A researcher is interested in the effect of body
position on a person’s ability to relax, as measured
by EMG biofeedback from the frontalis muscle. To
study this question, 11 subjects are randomly
assigned to two groups in a pretest-posttest design,
with one group positioned supine, the other sitting.
Results are recorded as changes in microvolt activity.
The researcher hypothesizes that the positions will
facilitate different levels of relaxation (a
nondirectional hypothesis)
WILCOXON SIGNED RANK TEST
•Compares 2 dependent samples
•Within subjects