Lecture 7 Flashcards
Validity, accuracy, sensitivity and friends
There will always be error in dietary assessments. The challenge is to …..
understand and estimate the error structure during analysis
What is validity
The extent to which a method gives you the “correct” answer ie measures what you want it to measure.
P1 and P2 of validity
P1- Analysing anthropometric data
P2- Anthropometry assignment
What is absolute validity
reference method requires surreptitious measurement of intake
what is relative validity
reference method is another indirect method of assessing intake
what is repeatability
the extent to which a method gives you the same answer each time you use it
what is repeatability also known as
reproducibility, reliability and precision
what are the issues to consider in a FFQ validation study
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
Accuracy
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
what is measurement error
difference between the measured value and the true value
observed intake is equal to
true intake + measurement error
what are the two types of measurement error
random error and systematic error
measurement error leads to
attenuation of relationships
bias
random error is due to
chance or normal variation
does random error change the mean
no it doesnt
random error increases….., therefore
variability around the mean, therefore decreased repeatability
what are the main sources of random error
- daily variation in consumption
- inaccurate portion size estimation
can random error be removed
not totally
systematic error is far more
concerning than random error
systematic error causes
results to depart from the true value in constant direction (bias)
systematic error changes ……, therefore
changes the mean, therefore decreased validity
what are the main types of systematic error
- selection bias (non-representative sample)
- measurement bias (incorrect calibration)
can systematic error be removed
can’t be removed by statistical analysis
what is non response bias
people dont volunteer to participate
people dont adhere to intervention
people drop out
in non response bias, a non representative sample may
may under or over estimate effects
what are the strategies to minimise non response bias
- 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
what is respondent bias
bias introduced by the respondent
what are the three main types of respondent bias
- low energy reporting
- over reporting
- social desirability and approval bias
what are the strategies to minimise respondent bias
- pre test methods
- interviewer training
- private interviews
- identify flawed data
- identify participants at risk
what is interviewer bias
bias introduced by the interviewer
interviewer bias may be caused by
incorrect recording
intentional omissions
poor cultural sensitivity
what are the strategies to minimise interviewer bias
standardised computer interviews
train interviewers
identify problem interviewers
what are the two forms of respondent memory lapses
errors of omission
errors of commision
what are errors of omission
like when you forget something you ate
what are errors of comission
can’t remember but will fill it in with something else
respondent memory lapses may be more likely …
- longer time period to be recalled
- men
- age extremes
- distracting environments
what are strategies to avoid respondent memory lapses
multiple pass interviewing techniques
- like when they have uninterrupted recall and then probing questions
minimise time between intake and recall
work with information retrieval
incorrect portion estimate is probably
the largest source or error
incorrect portion estimation is when they fail to
accurately quantify amount eaten
incorrect portion estimation impacts
all ages, BMI, SES and genders
what are the strategies to minmising incorrect portion estimation
- measurement aids
- train interviewers
- train respondents
what is the definition of supplements
there is no uniform definition of supplements
omission of supplements leads to
- systematic underestimation of nutrient intake
- overestimation of prevalence inadequacy
what are the strategies to minimising omission of supplements
structured questionnaire on long term intake
closed ended questions on : brand, amount per unit, frequency of use, duration of use
interviewer sights supplement
what is sensitivity
how good the test / method is at correctly identifying people who have the disease / low status
what is specificity
how good the test / method is at correctly identifying people who are well / have adequate status