[manphnx]

I couldn't understand this question!
the parameters of sensitivity and specificity are used for assessing?
1. criterion validity
2. construct validity
3. discriminant validity
4. content validity

and while your at it... here is another question.... just a random one.
calculation of a sample size for a prevalence study all of the following are necessary except..
1. prevalence of the disease in population
2. power of the study
3. significance level
4. desired precision
:notsure::notsure::notsure:

 

[patelMD]

a) 4

b) 4

Not too sure about B) though

 

[khanar]

Not high yield

Where did you get these questions from for God's sake!
The various types of validity have never been tested in the USMLE exam ever.
Even the most expert statisticians are confused about these.
Moreover, I think the question is wrong because specificity and sensitivity are other measures of the validity of the test and they are not categorized into any of the options you listed.

 

[patelMD]

Where did you get these questions from for God's sake!
The various types of validity have never been tested in the USMLE exam ever.
Even the most expert statisticians are confused about these.
Moreover, I think the question is wrong because specificity and sensitivity are other measures of the validity of the test and they are not categorized into any of the options you listed.

I think with the first question, its basically asking you whether you know what information is gained from Specificity and Sensitivity, which I think is not tested, but everyone should know anyways.

The question two is just weird, I guessed! )

On another note:

Is biostats and Epidemio THAT important on Step 1? I sucked at the kaplan practice questions (in the lecture notes)

 

[khanar]

I think with the first question, its basically asking you wether you know what information is gained from Specificity and Sensitivity, which I think is not tested, but everyone should know anyways.

The question two is just weird, I guessed! )

On another note:

Is biostats and Epidemio THAT important on Step 1? I sucked at the kaplan practice questions (in the lecture notes)

Yes it's high yield, but absolutely not such kind of questions.

 

[drarnab]

I couldn't understand this question!
the parameters of sensitivity and specificity are used for assessing?
1. criterion validity
2. construct validity
3. discriminant validity
4. content validity

and while your at it... here is another question.... just a random one.
calculation of a sample size for a prevalence study all of the following are necessary except..
1. prevalence of the disease in population
2. power of the study
3. significance level
4. desired precision
:notsure::notsure::notsure:

first one is easy
3. Discrimination, also read up on ROC curves.. receiver operator characteristics.

second one is a weird. significance level makes no sense for a prevalence study !! LOL.. a significance level means that a "difference" is not by chance..

anyway stop doing these confusing questions. I did these during my MPH at johns hopkins... these are not USMLE level.

 

[manphnx]

first one is easy
3. Discrimination, also read up on ROC curves.. receiver operator characteristics.

second one is a weird. significance level makes no sense for a prevalence study !! LOL.. a significance level means that a "difference" is not by chance..

anyway stop doing these confusing questions. I did these during my MPH at johns hopkins... these are not USMLE level.

well! i guess i owe u guys an apology... i just posted these questions just for discussion purposes. i thought there might be a hidden concept that i m missing....

any ways the answer to the first question is 1. (as given in the book, just matter of factly)
and answer to the second question is 'power of study'
explanation:
factors determining the sample size include.
1. design of study
2. plan of statistical analysis
3. accuracy of measurement to be made
4. degree of precision required
5. degree of significance
6. ratio of cases to controls in a case control study or prevalence of a disease in a cohort study

power of the study is the ability of a statistical test to detect a null hypothesis that is false and the probablility that a false null hypothesis is rejected