Week 3 Flashcards
What is ratio
Releationship between 2 numbers,, numerator not necessarily included in the deminontator 5:2= 2.5 sex ratio
What is proportion
Numererator necessarily included in the denominator ranges between 0 and 1 how many women in the population 2/4 50%
What is Odd
The probability of an event occuring relative to it not occurring different than proprtion its how many are divided by how many are not. If 25 people in 100 people are diabetic odd is 25/75=0.33.
What happens when a disease is rare?
Odd esitmates porpotion
What is rate
Speed of occurrence of an event over time it is porportion but need a time factor
What is prevalence rate
The proprtion of the population that has a given disease at a specified time
What is point prevalence rate
with disease at specific time divided by population at same time can times by 1000 to get per 1000 people
What is measures of incidence
The proportion of the population at risk that develops a given disease or other attribute during a specified time period
How do you measure incidence
Remove people who are not at risk you remove them from the population you can’t get cancer if you already had it
What is the formula for incidence rate
new events during specified time period divded by population at risk if 9000 residents were studied and 245 of them already had cancer and a total of 179 new cases has been digonsed what is the ir rate of cancer per 10 000 people 179 divided by 9000-245 times 10,000
What is relative risk
How many times as likely it is that someone who is exposed to something will experince a particular health outcome compares to someone who is not exposed
What is the formula for relative risk
the top is incidence in exposed a/ a+b divided by incidence in unexposed c divided by c+d
What dose it mean if for relative risk you got the number 1.76
If you smoke the chance of you getting the disease is 76%
What if you wanted to caculate odds ratio in a table
odds of disease in exposed (a/b) divided by odds of disease in unexposed c/d not a percentage
What is random error
Error due to chance
What is systematic error
Error due to recognizable source
Can you have both random and systematic error
Yes
What is precision vs accuracy is measurment
If you have high precision it is reliable with high accuracy is valid
Why can insufficient precision happen
The measurment tool is not precise enough
Two indpendent interviewers rate the same differtly using the same scale
The same in
What makes a valid measure
It measures what it is suppose to measure
Does not include unnecessary questions
Is able to do what it is designed to do predict discriminate etc
What is Misclassification
Putting people into the wrong categories
What is non-differential
Usually weakens associations brings estimants closer to the null valuse dilutes the estimate, data collected from one group is inacuarate while the other group is accurate
What is differential
Different in different study group effect estimates may change in any direction depending on the particular error both groups are inaccurate
What can you do to fix information bias once it occured
Nothing needs to be reduced during study design
