Lecture 7 - Validity, Accuracy and Sensitivity

Question 1

what is validity

Answer 1

the extent to which a method gives you the correct answer

i.e measures what you want it to measure

Question 2

what is absolute validity

Answer 2

reference method requires surreptitious measurement of intake

Question 3

what is relative validity

Answer 3

reference method is another indirect method of assessing intake

Question 4

what is repeatability

Answer 4

the extent to which a method gives you the same answer each time you use it

Question 5

what is repeatability also known as

Answer 5

reproducibility, reliability and precision

Question 6

what are the issues to consider in a FFQ validation study

Answer 6

• appropriate reference method = that does not require memory
• study sample = needs to be accurate of what you are wanting to asses
• timeframe = reference and FFQ should be over same timeframe
• methods of statistical analysis

Question 7

accuracy is not a word commonly used in nutrition, but what is it

Answer 7

accuracy is best used in a restricted statistical sense to describe the extent to which the measurement is close to the true value,

it has an important effect on the validity of the study

Question 8

what is measurement error

Answer 8

difference between the measured value and the true value

Question 9

observed intake is equal to

Answer 9

true intake + measurement error

Question 10

what are the two types of measurement error

Answer 10

random error and systematic error

Question 11

measurement error leads to

Answer 11

• attenuation of relationships
• bias

Question 12

random error is due to

Answer 12

chance or normal variation

Question 13

does random error change the mean

Answer 13

no it doesnt

Question 14

random error increases….., therefore

Answer 14

variability around the mean, therefore decreased repeatability

Question 15

what are the main sources of random error

Answer 15

• daily variation in consumption
• inaccurate portion size estimation

Question 16

can random error be removed

Answer 16

not totally

Question 17

systematic error is far more

Answer 17

concerning than random error

Question 18

systematic error causes

Answer 18

results to depart from the true value in constant direction (bias)

Question 19

systematic error changes ……, therefore

Answer 19

changes the mean, therefore decreased validity

Question 20

what are the main types of systematic error

Answer 20

• selection bias (non-representative sample)
• measurement bias (incorrect calibration)

Question 21

can systematic error be removed

Answer 21

can’t be removed by statistical analysis

Question 22

what is non response bias

Answer 22

people dont volunteer to participate

people dont adhere to intervention

people drop out

Question 23

in non response bias, a non representative sample may

Answer 23

may under or over estimate effects

Question 24

what are the strategies to minimise non response bias

Answer 24

• minimise response burden
• mail or phone reminders
• offer material rewards
• train interviewers so warm and trusted
• identify non responders and characterise
• over sample groups at risk of non-responding

Question 25

what is respondent bias

Answer 25

bias introduced by the respondent

Question 26

what are the three main types of respondent bias

Answer 26

• low energy reporting
• over reporting
• social desirability and approval bias

Question 27

what are the strategies to minimise respondent bias

Answer 27

• pre test methods
• interviewer training
• private interviews
• identify flawed data
• identify participants at risk

Question 28

what is interviewer bias

Answer 28

bias introduced by the interviewer

Question 29

interviewer bias may be caused by

Answer 29

incorrect recording

intentional omissions

poor cultural sensitivity

Question 30

what are the strategies to minimise interviewer bias

Answer 30

standardised computer interviews

train interviewers

identify problem interviewers

Question 31

what are the two forms of respondent memory lapses

Answer 31

errors of omission or errors of commission

Question 32

what are errors of omission

Answer 32

like when you forget something you ate

Question 33

what are errors of comission

Answer 33

can’t remember but will fill it in with something else

Question 34

respondent memory lapses may be more likely …

Answer 34

• longer time period to be recalled
• men
• age extremes
• distracting environments

Question 35

what are strategies to avoid respondent memory lapses

Answer 35

multiple pass interviewing techniques
- like when they have uninterrupted recall and then probing questions

minimise time between intake and recall

work with information retrieval

Question 36

incorrect portion estimate is probably

Answer 36

the largest source or error

Question 37

incorrect portion estimation is when they fail to

Answer 37

accurately quantify amount eaten

Question 38

incorrect portion estimation impacts

Answer 38

all ages, BMI, SES and genders

Question 39

what are the strategies to minmising incorrect portion estimation

Answer 39

• measurement aids
• train interviewers
• train respondents

Question 40

what is the definition of supplements

Answer 40

there is no uniform definition of supplements

Question 41

omission of supplements leads to

Answer 41

• systematic underestimation of nutrient intake
• overestimation of prevalence inadequacy

Question 42

what are the strategies to minimising omission of supplements

Answer 42

structured questionnaire on long term intake

closed ended questions on : brand, amount per unit, frequency of use, duration of use

interviewer sights supplement

Question 43

what is sensitivity

Answer 43

how good the test / method is at correctly identifying people who have the disease / low status

Question 44

what is specificity

Answer 44

how good the test / method is at correctly identifying people who are well / have adequate status