Biostatistics Course Review
What is a population?
entire set of individuals or units to which data will be generalized (e.g., men with
type 2 DM)
____________________ is the process by which a researcher obtains
representative subjects (experimental units) from a target population.
Sampling
When does bias occur in a study?
extraneous factors occur disproportionately in the sample (e.g., all males and no
females, only 18 year old college students and no 21 year olds)
NOTE: these samples are referred to as NON-REPRESENTATIVE
____________________ describes the characteristics that subjects must have to
qualify for the study.
Inclusion criteria
__________________ define factors that will preclude participation in the study.
Exclusion criteria
Factors to consider when choosing a sample group:
control variables and ensure applicability to the larger population
,Example of controlling variables in a case study:
Tylenol vs Motrin- which is more effective in treating headaches.
Must account for: environment, causative factors, pain tolerance, dosage
What MUST a variable have in order to have a valid study?
the ability to be measured (i.e., assign a numeral to a variable)
What are variables?
factors that can have more than one value (i.e., anything can change)
What is the difference between continuous and discrete data?
1. Continuous data can have any conceivable value within any observed range
(i.e., body weight)
2. Discrete data is descriptive and categorical can only taken on certain values,
cannot be measured in intervals of less than 1 (i.e., gender)
What is the independent variable?
the variable that is changed and manipulated by the researcher (i.e., type of
medication)
NOTE: these changes can be qualitative or quantitative
What is the dependent variable?
variable that is measured and a response to the independent variable (i.e., body
weight)
,What are descriptive studies?
usually unplanned or natural experiments (noticed in nature)
i.e., no manipulation of a variable is needed
e.g., relationship between dietary calcium intake and bone health
What is the MOST important rule of statistics (in my opinion)?
CORRELATION DOES NOT GUARANTEE CAUSATION
(i.e., just because 2 results are related it does not mean that they will 100%
occur)
What are Extraneous variables?
variables that are not of interest in the study, but that can affect the dependent
variable
(i.e., gender or age-- should control these variables)
Types of measures in human research:
behavioral, physiological/anataomical, self-report
What defines behavioral statistics?
direct observation of anything a person (or animal) does
NOTE: each behavior should be classified with use of a code for each behavior
What are examples of physiological/anatomical data measurements?
, electrical recordings, range of motion, heart, BP, blood analysis, reflexes
What are self-reports in statisitics?
-responses on questionnaires, responses during interviews
-may relate to how subjects acts, feels or their experience
What are constructs?
general abstract concepts that cannot be directly measured or observed in
statistics
What are direct measurements?
direct observation of the desired variable that is being measured (i.e., body
weight)
What are examples of indirect measurements?
hunger, depression, strength and language
(i.e., measuring different things that are representing something else or desired
variable)
EXAMPLE: on a scale from 1-10, when was the last time they ate measures
HUNGER
What are the four types of measurement scales?
nominal, ordinal, interval, ratio
*All require the assignment of a numeral, and require rules of measurement
What are nominal scales?
What is a population?
entire set of individuals or units to which data will be generalized (e.g., men with
type 2 DM)
____________________ is the process by which a researcher obtains
representative subjects (experimental units) from a target population.
Sampling
When does bias occur in a study?
extraneous factors occur disproportionately in the sample (e.g., all males and no
females, only 18 year old college students and no 21 year olds)
NOTE: these samples are referred to as NON-REPRESENTATIVE
____________________ describes the characteristics that subjects must have to
qualify for the study.
Inclusion criteria
__________________ define factors that will preclude participation in the study.
Exclusion criteria
Factors to consider when choosing a sample group:
control variables and ensure applicability to the larger population
,Example of controlling variables in a case study:
Tylenol vs Motrin- which is more effective in treating headaches.
Must account for: environment, causative factors, pain tolerance, dosage
What MUST a variable have in order to have a valid study?
the ability to be measured (i.e., assign a numeral to a variable)
What are variables?
factors that can have more than one value (i.e., anything can change)
What is the difference between continuous and discrete data?
1. Continuous data can have any conceivable value within any observed range
(i.e., body weight)
2. Discrete data is descriptive and categorical can only taken on certain values,
cannot be measured in intervals of less than 1 (i.e., gender)
What is the independent variable?
the variable that is changed and manipulated by the researcher (i.e., type of
medication)
NOTE: these changes can be qualitative or quantitative
What is the dependent variable?
variable that is measured and a response to the independent variable (i.e., body
weight)
,What are descriptive studies?
usually unplanned or natural experiments (noticed in nature)
i.e., no manipulation of a variable is needed
e.g., relationship between dietary calcium intake and bone health
What is the MOST important rule of statistics (in my opinion)?
CORRELATION DOES NOT GUARANTEE CAUSATION
(i.e., just because 2 results are related it does not mean that they will 100%
occur)
What are Extraneous variables?
variables that are not of interest in the study, but that can affect the dependent
variable
(i.e., gender or age-- should control these variables)
Types of measures in human research:
behavioral, physiological/anataomical, self-report
What defines behavioral statistics?
direct observation of anything a person (or animal) does
NOTE: each behavior should be classified with use of a code for each behavior
What are examples of physiological/anatomical data measurements?
, electrical recordings, range of motion, heart, BP, blood analysis, reflexes
What are self-reports in statisitics?
-responses on questionnaires, responses during interviews
-may relate to how subjects acts, feels or their experience
What are constructs?
general abstract concepts that cannot be directly measured or observed in
statistics
What are direct measurements?
direct observation of the desired variable that is being measured (i.e., body
weight)
What are examples of indirect measurements?
hunger, depression, strength and language
(i.e., measuring different things that are representing something else or desired
variable)
EXAMPLE: on a scale from 1-10, when was the last time they ate measures
HUNGER
What are the four types of measurement scales?
nominal, ordinal, interval, ratio
*All require the assignment of a numeral, and require rules of measurement
What are nominal scales?
