Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
37 Cards in this Set
- Front
- Back
|
FIELD TAGS
|
WITHIN BRACKETS ABREVIATIONS FOR SEARCH CATEGORIES
|
|
|
MeSH INDEXERS MUST:
|
- Have at least a bacelor's degree in biomedical science
- reading in one or more foreign language - special training in MEDLINE indexing |
|
|
STATISTICS ARE TO SAMPLES AS A _______ IS TO A POPULATION
|
PARAMATER
|
|
|
INDEPENDENT VARIABLE
|
A TREATMENT OR INTERVENTION, PRECEDE THE DEPENDENT IN TIME
|
|
|
DESCRIPTIVE STASTICS SUMMARIZE DATA BUT CAN NOT BE USED TO ________
|
TEST A HYPOTHESES
|
|
|
VARIABILITY
|
EXTENT VALUES ARE SPREAD ABOVE AND BELOW THE MIDDLE, a.k.a. DISPERSION
|
|
|
MODE:
|
MOST FREQUENT OCCURING VALUE
|
|
|
MEDIAN:
|
THE MIDDLE VALUE (NOT AVERAGE)
|
|
|
LEVELS OF MEASUREMENT: NOIR
|
NOMINAL, ORDINAL, INTERVAL, RATIO
|
|
|
NOMINAL
counting |
CODED BY NUMBER, NAME, OR LETTER
NO ORDER ex: gender (male or female) Mode |
|
|
ORDINAL
< > |
LIKE NOMINAL BUT ORDERED BY RANK - MORE OR LESS BUT NO TRUE VALUE OF VARIABILITY
ex: military rank (lieutenant, captain...) Median |
|
|
INTERVAL
+ - |
MEASURES ORDERED WITH EQUAL INTERVALS - NO TRUE ZERO
ex: fahrenheit symetrical Mean |
|
|
RATIO
+ - x / |
EQUAL INTERVALS - TRUE ZERO
ex: ROM, lifting capacity (15kg, or 30kg) symetrical Mean |
|
|
Normal curve AREA
|
68.3% +/- 1SD
95.5% +/- 2SD |
|
|
S'(2)
VARIANCE |
MEASURE OF SPREAD
|
|
|
Z-SCORE
Z=(x-u)/sigma x=raw score, u=mean, sigma=SD |
The number of SDs above or below the mean
|
|
|
z-table
|
portion of the area under the curve for different values of z
|
|
|
INFERENTIAL STATISTICS
|
DATA MAKE INFERECES ABOUT POPULATIONS, TAKE MANY MEANS AND THEN TAKE A MEAN OF THE MEANS
|
|
|
SAMPLE MEAN IS
|
AN ESTIMATE OR POINT ESTIMATE
|
|
|
STANDARD ERROR OF MEAN = SEm
lower the better |
SEm = SD/square root of (n)
|
|
|
SEm is high when____
|
SD is large or sample is too small
|
|
|
CONFIDENCE INTERVAL
typical 95% |
range likely to contain population mean
within 95% |
|
|
calculate CI range
|
z-score @95% multiply by SEm
+/- product to sample mean for limits |
|
|
SIZE OF CI _____
|
INDIRECT W/ SAMPLE SIZE
SMALL SAMPLE - LARGE CI (BAD) |
|
|
null Hypothesis testing
|
H(null) either accepted or rejected
|
|
|
H(null) accepted
|
no difference between two groups
|
|
|
H(null) rejected if outside CI
|
accept H1 - 95% that difference is true
|
|
|
alpha error (type 1)
|
reject H(null) = accept H1 (but false difference)
|
|
|
beta error (type 2)
|
accept H(null) = reject H1 (but true difference)
|
|
|
as alpha <0.05 =
|
type 1 decreases = type 2 increases
|
|
|
Power = 1-beta
|
probability of correctly rejecting a false H(null)
|
|
|
standard power value
|
0.80
|
|
|
if value of difference is w/in CI then
|
H(null) should not be rejected = no statistical difference
|
|
|
t-test
|
find whether the means of two groups are statiscally different
|
|
|
one-tailed or two-tailed
|
two is better because splits alpha between two tails
|
|
|
t =
|
difference between the group means/variability of the data
|
|
|
big t =
|
small p-value & statistically significant
|
