Biostatistics Lecture Notes

Vocabulary flashcards for Biostatistics lecture review.

Biostatistics Data

Raw material or set of values for one or more variables.

Variable

Characteristic or attribute that can take different values for different subjects.

Quantitative variable

Numerically valued variable.

Qualitative variable

Non-numerically valued variable.

Population

Collection of all individuals or items under consideration in a statistical study.

Sample

Subset of a population from where information is taken.

Standard deviation

A difference in value compared to the mean value.

Parameter

A summary value that characterizes the nature of the population.

Statistic

A summary value calculated from a sample.

Mean

Average; sensitive to outliers.

Median

Middle value when data is ranked; resistant to outliers.

Mode

Value that occurs with the greatest frequency.

Negatively Skewed

Tail to the left.

Positively Skewed

Tail to the right.

Harmonic mean

Used to average values that change in time.

Quartile

Each set of data has three of these values. Divides data into four parts.

Interquartile range

Q3 - Q1

Percentile

Divides a set of ordered data into 100 equal parts.

Range

Max - min

Variance

Square of standard deviation.

Coefficient of variation

(SD/mean) x 100

Frequency distribution

Shows classes or intervals of data.

Class width

Difference between lower limits of both groups in frequency distribution.

Midpoint

(lower + upper limit of a group)/2

Relative frequency

% of data that falls into that class (class frequency/sample size)

Cumulative frequency

Sum of the frequency for that class and all previous classes.

Class boundaries

Numbers that separate the classes without forming gaps between them.

Bar & pie chart

Summarize qualitative data.

Histogram

Bar chart for representing frequency distribution of quantitative variables.

Box plot

Median & first + third quartiles of the distribution are used.

Line graph

Individual data points are connected by a line.

Error bars

Indicate variability of data.

Marginal table

Consists of frequency & percentage for qualitative data or central tendency & dispersion for quantitative data.

Cross table

Shows 2 or more variables simultaneously in table format and can include sub-groups.

Scatter plot

Shows relationship between 2 variables when both are measured on a numerical scale.

Conditional probability

Occurrence of one event depends on whether the other event has occurred.

Sensitivity

Ability of a test to correctly identify a diseased person.

Specificity

Ability of a test to correctly identify healthy people.

False positive

Probability of having a disease for someone who is not actually sick.

False negative

Probability of not having a disease for someone who is actually sick.

Bayes theorem

Describes the proper way to incorporate new evidence into prior probabilities to form an updated probability estimate.

Census

Collect data of the entire population.

Simple Random Sampling

Small, homogenous population, equal chance of getting selected.

Stratified Random Sampling

Frame organized into separate strata, each stratum sampled independently (best results with homogeneous elements).

Systematic Sampling

Elements put into a list & every kth element is chosen (every 3rd/4th) (random start).

Cluster Sampling

Population divided into clusters a subset of clusters is randomly selected (based on geographical areas etc, epidemiologic research) – (best results with heterogeneous elements).

Discrete probability distributions; binomial and Poisson

Variable takes only integer values.

Continuous probability distribution; normal

Variable has values measured on a continuous scale.

Binomial

Only 2 outcomes = probability of either occurring is calculated - applicable when the outcome is the “number of times an event occurs”.

Poisson

If average number of occurrences of the event is given, associated probabilities can be calculated - applicable when the outcome is the “number of times an event occurs”.

factors that influence sample representative-ness

Sampling procedure• Sample size• Participation (response)

SAMPLING METHODS

probability and non_probability sampling

What do we need to determine a sample size?

Population mean- Population proportion

Normal

• NORMAL IS +2 SD (95%)

Z distribution

A normal distribution with a mean of 0, and standard deviation of 1 = standard normal distribution.