Research: Lecture 4 (quiz 3)

Question 1

Parameter

Answer 1

descriptive value for a population

Question 2

Statistic

Answer 2

descriptive value for a sample

Question 3

Mean

Answer 3

average
-most commonly used
-only used with interval/ratio
-influence by outliers
-toward the tail opposite of mode

μ, x

Question 4

When shouldn’t you report the mean?

Answer 4

if you have outliers/extreme scores, the mean will be pulled towards the extremes (towards the tail) and will not provide a central value.

Question 5

Variance

Answer 5

SD^2 or (distance from mean)^2/ n-1

σ^2

Question 6

Standard Deviation

Answer 6

the standard (average) distance between a score and the mean
- Square root (distance from mean)^2/ n-1

σ

Question 7

Frequency Distribution

Answer 7

organized picture of an entire set of scores

histogram, smooth curve, stem and leaf

Question 8

Smooth Curve

Answer 8

emphasizes the fact that the distribution is NOT showing the exact frequency for each category
-want it to be symmetrical (normal curve, mean and median are equal)

Question 9

1 SD in a normal distribution

Answer 9

68.26% (34.13%)

Question 10

2 SD in a normal distribution

Answer 10

95.44% (13.59%)

Question 11

3 SD in a normal distribution

Answer 11

99.72% (0.13%)

Question 12

Histogram

Answer 12

shows all the frequencies of the distribution

Question 13

Positive vs. Negative Skew

Answer 13

non-symmetrical distribution
-named for tail

Positive: scores pile up at low values, tail points to high values

Negative: scores pile up at high values with tail at low

Question 14

Kurtosis

Answer 14

peakedness of the distribution

Question 15

Leptokurtic

Answer 15

skyscraper
-higher and thinner peak
-low variability
-easier to get significance

Question 16

Platykurtic

Answer 16

hill
-lower peak
-higher variability
-harder to get significance

Question 17

Stem-And-Leaf Display

Answer 17

preserves the original data values
It’s especially useful for small to moderately sized data sets.
-each score divided into a stem (first digit) or leaf (last digit)

Question 18

Central Tendency Measures

Answer 18

describes the center of the distribution and represents the entire distribution of scores as a single number (mode, median, mean)

Question 19

Mode

Answer 19

most frequent
-used in all data
-located on one side near peak, other farthest from mean
-bimodal, multimodal

Question 20

Median

Answer 20

middle: 50% of the scores in the distribution have values that are equal or less than the median
-used for ordinal, interval, or ratio
-unaffected by outliers
-can’t show significant difference
-between mean and mode

Question 21

what is the difference in the location of the mean, median, and mode in a symmetric distribution versus a skewed distribution

Answer 21

symmetric: mean, median, and mode all located equally at the peak

skewed:
- mode: located at peak
- median: located in between mode and mean
- mean: towards the tail away from peak

Question 22

Variability

Answer 22

how spread out the data is
-descriptive (how spread out) and inferential stats (how accurate to population)
-measured by range or SD

Question 23

what is the difference between large and small variability

Answer 23

small: good representation

large: distorted representation

Question 24

How to calculate variance

Answer 24

1: find the mean
2: subtract values from the mean (deviation)
3: square the deviations
4: find sum of squared deviations
5: for sample: divide by n-1
for population: divide by N

Question 25

Range

Answer 25

total distance covered by the distribution (highest score - lowest score)

Question 26

SD in Normal Distribution

Answer 26

70% of scores 1 SD of mean (35+/-) -95% of scores 2 SD of mean -99% of scores 3 SD of mean ## Footnote standardized, mean is 0

Question 27

how to calculate SD

Answer 27

take the square root of the variance

Question 28

Z Score

Answer 28

where a score is located relative to other scores -# of SD above or below mean -descriptive (where in curve) and inferential stats (reference to population) ## Footnote z= score-mean/SD

Question 29

how to calculate z-scores (transforming X to z)

Answer 29

(score - mean)/ standard deviation

Question 30

Inferential Statistics

Answer 30

infer things about the population based on sample

Question 31

Probability

Answer 31

proportion under the curve -z score creates % as body or tail

Question 32

You receive a score that is exactly one standard deviation above the mean. What percentage of scores are below your score?

Answer 32

84 | 34+34+13.5+2.14+0.13

Question 33

According to the central limit theorem, if you have number, not letters _______ participants in your study normality can be assumed.

Answer 33

30

Question 34

If a research study is under powered which of the following threats to the validity of this research study exists?

Answer 34

statistical validity

Question 35

If a research study is claims they studied fitness in DPT students as measured by BMI which of the following threats to the validity of this research study exists?

Answer 35

construct validity

Question 36

Which of the following types of internal validity threats exist in a research study that is investigating improving ambulation in children 12 months to 24 months in which all participants receive PT once per week for 6 months.

Answer 36

maturation regression towards the mean

Question 37

face validity

Answer 37

does an instrument measure what it is supposed to measure based on simple **observation**

Question 38

content validity

Answer 38

does the instrument used cover the entire domain to be measured (what experts agree it measures)

Question 39

criterion based validity

Answer 39

the degree to which the outcomes of one test correlate with the outcomes of a **gold standard test**

Question 40

concurrent validity

Answer 40

the degree to which the outcomes of one test correlate the outcomes of **another test **

Question 41

predictive validity

Answer 41

can an instrument be used to predict some future performance/outcome

Question 42

construct validity

Answer 42

the degree to which a theoretical construct is measured by an instrument Does the test actually measure what it claims to measure?

Question 43

test-retest reliability

Answer 43

can successive measurements by an instrument be consistent

Question 44

intra-tester reliability

Answer 44

can the **same** rater get the same score?

Question 45

inter-rater reliability

Answer 45

can **different** raters get the same score?

Question 46

descriptive studies

Answer 46

simply describe data - surverys - retrospective data - normative data - qualitative data

Question 47

exploratory studies

Answer 47

looking at relationships between variables - correlational - predictive - methodological - case control - quasi-experimentals - single subject

Question 48

experimental studies

Answer 48

true experimental design with randomization

Question 49

ROC

Answer 49

a measures to help establish cut off scores for a specific measure - want more area under the curve for higher sensitivity/specificity -plot of sensitivity vs. specificity

Question 50

alpha level

Answer 50

preset significance level - the point at which you would consider the result highly unlikely - 0.05: 5% risk of committing a type 1 error (saying something is significant when it isn't) false positive

Question 51

p-value

Answer 51

actual probability that the results occurred just because of sampling error p<0.05: reject null hypothesis (significance) p>0.05: fail to reject null hypothesis (no significance)

Question 52

null hypothesis vs. alternative hypothesis

Answer 52

null hypothesis: proposes there is no significant difference between groups alternative: statement about the expected outcome or relationship between variables of a study

Question 53

meta-analysis

Answer 53

a systematic review that statistically combines the results of multiple studies into a single, pooled estimate

Question 54

cohen's D

Answer 54

a measure of effect size the larger differences in groups, the larger the effect size less variability = larger effect size (mean 1 - mean 2) / mean of SD's 0.2 = small 0.5 = medium 0.8 = large