Exam 2 - Handout 5 Flashcards
What is QUALITATIVE data? Example?
Meaningful information collected in words
Not typically used for healthcare research w/ large populations
Ex. Written observation in medical records
What is QUANTITATIVE data? Examples?
Numerical/countable information
Ex. Age, weight, BP
What are discrete variables? Examples?
- Categorical
- Have a few possible values
- Often defined as “counts”
Ex. Sex, number of hospitalizations, yes/no
What are continuous variables?
Exist on a defined scale
Ex. Age, body temp, weight
(Think number lines)
What are the levels of measurement?
Nominal data
Ordinal data
Interval data
Ratio data
nominal data
Discrete categories with no particular order (e.g., sex)
(NOminal = NO order)
ordinal data
Discrete categories that can be ranked
e.g., Likert-type questions, pain scales
(likert-type questions are ones where you answer with “agree” “strongly disagree”)
(ORDinal = in ORDer)
interval data
continuous data with:
- a defined scale
- constant intervals
DOES NOT HAVE A TRUE ZERO POINT (this is what makes it different from ratio data)
ex. temperature (because even if the temp is 0, there is still a temp)
ratio data
continuous data with:
- defined scale
- constant intervals
- true zero point
ex. age, weight, income
Independent variable
The variable hypothesized to explain an observed clinical phenomenon
Think of it as the cause
Dependent variable
Variable that is predicted/explained by the independent variable
Think of it as the effect
Control variables
Other explanatory variables included to hold external conditions constant and isolate the effect of the independent variable
Measures of central tendency
Mean
Median
Mode
Mean
Arithmetic average of a set of values
Median
The middle value when data is arranged in order
Preferred when data has outliers that skew the mean
Mode
The value that appears most often
Useful for non-numerical, categorical values
Measures of dispersion
Range
Interquartile range
Variance of standard deviation
Skewness
Interquartile range
The difference between the 75th and 25th percentiles
Represents the middle 50% of the data
Range
The difference between the highest and lowest values
Variance
Represented by σ²
The average squared distance of values from the mean
(SD/mean)
Standard deviation
Represented by σ
The square root of the variance
Skewness
Indicates if data are evenly distributed around the mean
Coefficient of variation (CV)
Standardized measure of dispersion
σ/μ
Positive skew
More data concentrated to the LEFT of the mean
Negative skew
More data concentrated to the right of the mean
Box plots
Visually display the range, IQR, and median of a variable
Useful for comparing distributions across groups
Frequency tables
Organize discrete or continuous data by counting the frequency of each value
Should include clear titles, column names, and formatting for easy interpretation
Bar charts
Discrete, categorical data
Pie charts
Represent proportions or relative quantities of values
Should be limited to a small # of clearly labeled categories
Histograms
Continuous data divided into discrete categories
Proportions
The number of observations with a given characteristic divided by the total number of observations
Often reported as percentages (proportion x 100%)
Rates
Computed over a specific time period and use a multiplier
Ex. Per 1000
Examples include mortality and incidence rate
Sensitivity
The ability of a test to correctly identify individuals w/ the disease
Proportion of true positives out of all individuals w/ the disease
Negative predictive value
Probability an individual does NOT have the disease given a negative test result
Specificity
The ability of a test to correctly ID individuals WITHOUT the disease
Proportion of true negatives out of all individuals without the disease
Positive predictive value
Probability an individual HAS the diseases given a positive test
Receiver operating characteristic (ROC) curves
Illustrate the tradeoff between sensitivity and false positive rate at different decision thresholds
The area under the ROC curve indicates the overall discriminating power of the test