Class 29: Biostats (Don't Screw This Up, This'll Be Weird) Flashcards Preview

Epidemiology > Class 29: Biostats (Don't Screw This Up, This'll Be Weird) > Flashcards

Flashcards in Class 29: Biostats (Don't Screw This Up, This'll Be Weird) Deck (8)
Loading flashcards...
1
Q

What are study/data measurements?

A
    • It’s all blue.
  • Data’ will be collected on desired ‘variables’
  • Dependent variable(s) [outcome variables]
  • Independent variables
  • Comparisons are commonly made (statistical analyses)
  • Inferences will be made about the sample-derived ‘data’ and their comparisons (Null Hypothesis)
2
Q

What is null hypothesis?

A
  • Null Hypothesis (H0)
  • ** Researchers either ACCEPT or REJECT this perspective, based on STATISTICAL ANALYSIS
  • A research perspective which states there will be no (true) difference between the groups being compared
  • Most conservative and commonly utilized
  • Various statistical-perspectives can be taken by the researcher:
    • Superiority
    • Noninferiority
    • Equivalency
  • Alternative Hypothesis (H1)
  • A research perspective which states there Will Be a (true) difference between the groups being compared
3
Q

Name the three primary levels (groupings) for variables based on answers to 2 key attributes

A
  • NOMINAL
  • ORDINAL
  • INTERVAL
4
Q

What are the 2 key attributes of data management

A
  1. Magnitude (or Dimensionality)
  2. Consistency of scale (Fixed Interval)
    - Equal, measurable spacing between UNITS of scale

Unofficial third: 3. Rational/Absolute Zero (Ratio group)

Each attribute can be assessed with a “YES” or “NO” response to the inquiry of ‘Does it have IT’

5
Q

Describe the Nominal Level

A
  • NOMINAL (Dichotomous/Binary; Non-Ranked Named Categories)
  • NO Magnitude / NO Consistency of scale
  • NOMINAL VARIABLES ARE SIMPLY LABELED-VARIABLES WITHOUT QUANTITATIVE CHARACTERISTICS (OR DICHOTOMOUS)
  • Example:
  • Male/Female, Democrat/Rebpublican, Truck Driver/Lawyer

Non-example: Being a second year med student means suggests you’ve been in med school longer so that can’t count. No numerical magnitude allowed. (Correction: Turns out this last paragraph/sentence is totally not true)

6
Q

Describe the ordinal level

A
  • ORDINAL (Ranked Categories; Non-Equal-Distance)
  • YES Magnitude / NO Consistency of scale
  • This one is like nominal, but has to be ranked or ordered. One thing is larger than another.
  • Examples include Satisfaction surveys (From 1 being hated it to 5 being worshiped it)
7
Q

Describe the interval/ration Level

A
  • INTERVAL/RATIO (Order & Magnitude & Equal Intervals-of-scale (UNITS))
  • YES Magnitude / YES Consistency of scale
  • Living Siblings (number) & Personal Age (in years)
  • Can you describe others?

(These have to be exact, consistent across everyone, and measurable. You’ll probably screw this up somehow.)

  • See Slides 12-13 for breakdown.
8
Q

What is the hierarchy for changing levels of measurement data

A
  • Nominal - Attributes are only named; weakest level
  • Ordinal - Attributes can be ordered
  • Interval - Distance is meaningful
  • Ratio - Absolute Zero
  • After data is collected, we can appropriately go down in specificity/detail of data measurement (levels), but never up!
  • We can change glucose numbers to (prediabetic, diabetic as fuck, etc…), but we can’t change the categories into numbers without knowing said numbers.