Introduction to Biostats Flashcards

Make that A boi

1
Q

What is type 1 error?

A

Alpha error

Not accepting the null hypothesis when it is actually true and you should have accepted it

here really is no true differences between the groups being compared but you (in error) did not accept the Null Hypothesis thereby ultimately stating that you believe there is a difference between groups (when there really is NOT!)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is kurtosis?

A

A measure of the extent to which observations cluster around the mean. for a normal distribution the value or kurtosis stastistic is 0

Positive kurtosis = more cluster

Negative Kurtosis = less cluster

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is Skewness?

A

A measure of asymmetry of a distribution

Perfectly normal distribution is symmetric and has a skewness value of 0

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is the interval or ratio level?

A

Interval

  • Abritrary zero value (doesnt mean absence)
  • Temperatire = 0 = freezing

Ratio

  • Absolute (rational) zero value (0 means absence of meaurement) like physiological parameters

Yes order or magnitude

Yes consistency of scale or equal distances

Ex: living siblings (number) , personal age (in years) , blood pressure, Heart rate, Speed, amount of LDL

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What are the required assumptions of interval/ratio data ? (for proper selection of a parametric test)

A
  1. Normally distrubted
  2. equal variances
    1. multiple test availble to assess for equal variances between groups
    2. LEVENEs TEST
  3. Randomly derived and independent
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What other ways to look at interpretation of preset value?

A
  • The probability of making a type 1 error, if the null hypthesis is rejected
  • Probability of erroneously claiming a difference between groups when one does not really exist
  • Probability of obtaining group differences as great or greater if the groups were actually the same /equal
  • Probabiltt of obtaining a test statisitc as high/higher if the groups were actually the same or equal
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What are the three important steps in sample size determination?

A
  1. Minimum difference between groups deemed most significant
    1. smaller the difference b.w groups necessary to be consdered significant or importantm the greater the sample size (N) needed
  2. Expected variation of measurement
    1. KNown or estimated from past studies/ populations
  3. Alpha (type 1) and beta (type 2) error rates and confidence internval (usually ranged from 90% to 99%)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q
A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is a confidence interval?

A

CI’s are a value that describes both stastical significance and spread

Based on variation in sample (v/SD) and sample size (n)

journals are moving away from soley reporting P values or even showing them at all

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

How do you intreprete confidence intervals?

A

If the CI crosses 1.0 for ratios (OR/RR/HR)

or

0.0 for absolute differences = NOT SIGNIFICANT

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is type 2 error?

A

Beta error

Accepting the null hypothesis when it is actually false and you should have not accepted

There really IS a true difference between the groups being compared but you (in error) accepted the Null Hypothesis thereby ultimately stating that you believe there is no difference between groups (when there really IS!)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

how can the levels of measurement of data change?

A

After Data is collected we can appropriately go down in specificity/detail of data measurement (levels) but never Up!

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What are the three primary levels or groupings based for variables and there three key attributes? ***

A

NOMINAL

ORDINAL

INTERVAL or RATIO

3 Key attributes

  • Order/magnitude
  • Consistance of scale / equal distances
  • Rational Absolute zero
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is the levenes test?

A

It is a test used to access weter variences are equal b/w groups. Its goal is to find signficant difference between two compared group

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is standard deviation (SD) ?

A

square root of variance value (restores units of mean)

17
Q

What is a regression?

A
  • Provides a measure of relationship between the variables by allowing the prediction about the dependent, or outcome, variable (DV) knowning the value/category of independent variable
  • Also able to calulate OR for measure of Association
  • Nominal Regrsseion Test = logistic Regression
  • Odinal regression Test = Multinomial Logistic regression
  • Interval Regression Test = Linear Regression
18
Q

What is a survival test?

A
  • Compares the proportion of events over time , or time to events between groups
  • (repersented by a kaplan meier curve
    • Nominal Survival Test = Log-Rank test
  • Ordinal Survival Test = Cox-Proportional hazards Test
  • Interval Survival Test = Kaplan-Meier test
    • all can be repersented by a kaplan meier cruve
19
Q

What is variance (from mean) ?

A

The average the squared-differences in each individual measurement value (x) and the groups’ mean (x)

20
Q

When conducting studies (human) what are the three characteristics about your study data or measurements?

A
  1. Data is about desired variables
    1. dependent (outcome)
    2. independent
  2. Comparisons (stasical analyses)
  3. Inferences about data and comparisons
    1. null hypothesis
21
Q

Notes slide

A

Nominal ( dichomatoous & non ranked named categories)

  • Think No and NO

Ordinal (ordered, ranked categories)

  • Think Yes and No

Interval/Ratio ( equal distance numerical scales (units)

  • Think Yes and Yes

Nominal and ordinal are both discrete

Interval/Ratio continous

All stat tests are selected based on levels being compared

22
Q

What type of variable falls into the nominal level?

A

It has….

  • No order or magnitude
    • like it can not be ranked
  • No consistency of scale or equal distances
    • that there are no units and the difference values have no regular spacing
  • Simply labeled variables without quantative characteristics
    • like truck drivers and lawers have no numeral value attached to them

Ig there are only 2 categories then its is usally nominal. It can also include a variable that includes a multide of categories under it

Ex:

hair color -> brown, blonde, black, red, gray

Occupational -> truck driver , lawer, police officer

Political party -> Democrate, independent, republican

Sex -> Male or Female

All of these must fall under the rules above and they do. There is no measurable difference between brown and blonde hair, b/w republican or democrate, and b/w lawyer and trucker. If you were took at deeper level like inclusion of a finical aspect then it would no longer be Nominal.

23
Q

What is a negatively skewed data distrIbitution?

A

Asymmetrical distribution with on tail longer than another .

A distribution is skewed anytime median differs from the mean.

Negativly skewed = Mean is lower than the median.

tail points to left .

24
Q

What is a positively skewed data distribution?

A

Asymmetrical distrubution wiht on tail longer than another.

A distrubution is skewed anytime the median differs from the mean.

Positively skewed occurs when mean is higher than median.

tail points right

25
Q

What 4 key questions must you ask when selecting the correct statistical test?

A
  1. What DATA LEVEL is being recorded?
    1. Does data have order/magnitude ? (y/n)
    2. Does Data have equal, consistent distancess along the entire scale? (y/n)
  2. What type of comparison assessment is desired
    - Frewurncies/ counts/ proportions
    1. Outcome Prediction/Association (oR) = regression
    2. Even-occurrence/ Time-to-Event = survival test
    3. correlatoon = correlation test
  3. How many groups are being compared?
    • 2 or 3 or more groups
  4. is the data independent or related (paired) ?
    • ​data from the same (paired) or differrent groups (independent)
26
Q
A
27
Q

What describe characterisitcs of graph whose data is normally distributed?

A

Normally distrubted = symmetrical

MEan and median are equal or near equal

Statstical test useful for symmetrical data are called parametric

28
Q

What is the P value?

A

A statistical test determines possible error rate or likelihood of chance in comparing difference or relationship between variables

  1. a probability (p) value is obtained ; based on probability of observing, due to chance alone, a test statistic value as extreme or more extreme than actually observed if groups were similar
  2. probaility of value is selected by investigators before study begins.
29
Q

What type of variable can be described as Ordinal?

A

Yes order or magnitude

No consistency of scale or equal distances

there categories remain rankable but there NO consistance scale of measure from one category to the next

Ex:

Pain scale - there is no consistant or measurable difference between happy or sad from person to person. The level of sadness that someone registers is different from another person.

30
Q
A
31
Q

What is the null and alternative hypothesis?

A

Null hypothesis (H0):

  • A research perspective which states that wthee will be no true difference between the groups being compared
  • Superior, inferiority, equality, and its most common

Alternative hypothesis (H1):

  • A research perspective which states there will be a (true) difference between the groups being compared

_researchers either accept or dont accept the null based on stastical anaylsis_

32
Q

How do you handle data that is not normally distributed?

A

Use a stastical test that does not require the data to be normally-distrubuted (NON PARAMETRIC TESTS)

Or

TRansform fata to a standardized value (Z score log transformation)

Always run descriptive stastistics and graphs (not sure importance of this one)

33
Q

What are impacts to stastistical significance?

A

Power (1 - beta)

  • The statisical ability of a study to detect a true difference if one truly exhists between group comparisions and there the level of accuracy in correctly accepting / not accepting the null hypothesis

Sample size

  • The larger the sample size the greater the likelilhood (ability) of detecting a difference if one truly exists
  • Increase in power
34
Q

What is the correlation test?

A
  • Provides a quantitative measure of strength and direction of relation between variables ?
    • Values range from -1.0 to +1.0
  • Partial correlation = a correlation that contorls for confounding variables
  • nominal = Contingency Coefficient
  • Ordinal = Spearman Correlation
  • Interval = Pearson Correlation
    • P > .05 just means there is no linear correlation; there still may be non linear correlations presents
  • All corlations can be run as a partaiil correlation to control for confounding
35
Q

What are the two ways to reperesent data?

A
  • Qualitiative : words used to repersent data
  • Quantative : numbers used to repersent data