Test 1 Flashcards
Nominal Scale of Measurement
Based on classification of an observation according to the group to which it belongs
Ex. Gender, political party, marital status
Ordinal Scale of Measurement
Classification of an observation according to its relationship to other observations
Ex. Poor-fair-good rating scale
Continuous (Interval) Scale of Measurement
Measurement scale with equal units of measurement. Distance between two numbers is known. Zero point is arbitrary.
Ex. Fahrenheit and centigrade
Continuous (Ratio) Scale of Measurement
Characterized by equal units of measurement and a true zero point.
Ex. Mass, time
Populations
Greek symbols; not a variable
Samples
Roman characters; variables
Mode
Most frequent measurement. Most useful with nominal scale.
Median
Middle value when measurements arranged in order of magnitude. Most useful with ordinal scale.
Mean
Sum of measurements divided by total measurements. Most useful with interval or ratio.
Range
Difference between the largest and smallest measurements
Interquartile Range
The range between a quarter of the data.
Variance (MS)
Average of the square of deviations of measurements about the mean. Considered in terms of the distance of each measurement from the mean. “Degrees of freedom”
Calculate mean -> subtract mean from each measurement and squaring result -> add these deviations to get some of squares (SS) -> divide SS by sample number-1
Standard Deviation (SD)
Positive square root of the variance
Coefficient of Variation (CV)
Unitless measurement of the percentage of the spread.
Standard error of the mean (SE or SEM)
Relates to sample size. Used to assess how accurately a sample mean reflects a population mean.
SE=SD/square root of N
Efficacy
The extent something produces results under IDEAL CONDITIONS
Effectiveness
Extent to which something produces results when deployed under routine conditions
REAL WORLD
Hierarchy of Evidence
- Systematic review and meta-analysis
- Clinical trial
- Longitudinal cohort studies
- Case-control studies
- Descriptive and cross-sectional studies
- Case reports and case series
- Personal opinion, subjective impressions, anecdotal accounts
Impact Factor
Total number of citations to articles appearing in journal / total number of articles published
Systematic Review and Meta-analysis
- Analysis of multiple research articles
Clinical Trial
- Given population is given an agent to see if it is effective. Placebos are often used.
Longitudinal Cohort Studies
- Groups of individuals defined based on exposure to suspected risk factor. Subjects followed over period of time. Groups should have people that are identical other than the variable being studied
Case-control studies
- Subjects chosen based on outcome and are looked at working backwards to see if certain things helped lead to those outcomes.
Descriptive and cross-sectional studies
- Surveys given to multiple groups. Often saw as a snapshot in time. Less reliable but can help you narrow down a question.
Case reports and case series
- Reports on individual cases
Central Limit Theorem
Sample mean will be normally distributed amongst the population.
Confidence Interval
Used to say how confident the sample mean represents the population mean.