Categorical data Flashcards

1
Q

what is abo blood group an example of

A

nominal variable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

what is tumour stage an example of

A

ordinal variable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

what is a catergorical varibale taht has two levels called

A

binary variable

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

what does chisqaured do

A

It compares observed results to expected

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

what are importanyt things about the chi squared test

A
  • make sure it is counts in cells not proportions
  • make sure each independasnt unit appears once on the table
    -make sure the expected values are not too small
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

what would you use when the expected i smaller thsan 5 casts

A

fisher exact test

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

why is the fishers exact test not always used

A
  • too many tables when the counts are large
  • no need chi sqaured is usually adequate
  • exact test doesnt need approxiamtion but corresponsing confidence intervals are often too wide
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

what are the 3 ways of finding a difference between parameters pi1 and pi2

A
  • the abosultue difference
    -the relative risk
    -the odds ratio
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

when would you not use an agreement statitsic

A
  • when comparing measurements taken in differnet subjevcts
  • when there has been an intervention that is likely to have affected the measured trait
  • measurements are likely to be different
  • hypothesis testing
    -to find evidence taht a differnce does exist
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

what test allows inter rate agreement

A

kohens kappa

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

what are the 3 steps of chisquared

A
  • make a table of what you would expect
    find how. far from obserevd
  • work out how suprosiing the difference is
How well did you know this?
1
Not at all
2
3
4
5
Perfectly