Lecture #17- Chance 1

Question 1

What is external validity? How does one do that?

Answer 1

The extent to which the findings of the study can be applied to a broader pop - also known as GENERALISABILITY

Judgement call depending on what is bring studied and who it is being applied to (no easy metric about who the studies can be applied to)

Question 2

What is internal validity? What are three elements when it comes to assessing the internal validity?

“Are our study findings _____ and don’t accurately reflect ‘___ _____’”?

“Are there other………?”

Answer 2

The extent to which the findings of the study are free of chance, bias and confounding (alternative explanation of the findings)

wrong…..the truth

explanations for the study findings, apart from them bring right?

Question 3

Why does chance arise?

Answer 3

Because we sample

Question 4

Sampling

1. We want our study to say something about what?
2. We take a ____ of the whole population

Answer 4

-

Question 5

Estimating population parameters

1. There is a _____ ____ out there, it’s called the ______. Will we ever know what it is? It’s the measure in the population and what we measure is of the sample so we are _____ the parameter. Thus the two are called: ____ and e______

Answer 5

-

Question 6

Fundamental concepts:

1. Parameter = (definition)
2. Estimate = (definiton) and sometimes called what?

Answer 6

-

Question 7

Study samples _____ so we will get variation in e_____ too. Explain the basic fundamental concept of this.

Answer 7

-

Question 8

Sampling error

1. If you repeatedly sample randomly from the same pop, most of the time you’d get……….
2. But some of the samples would be………
3. This is sampling error and is a form of what kind of error?
4. Commonly just called?

Answer 8

-

Question 9

What can be done about sampling error?

1. Increase……..
   - Reduces sample ______ (standard deviations etc)
   - Increases…….
   - Increases…….

But not too huge of a sample because….

Answer 9

-

Question 10

Can you eliminate sampling error? Why?

Answer 10

-

Question 11

Fundamental concept #2

1. What we mean by chance and sampling error is that……
2. Can’t eliminate sampling error but can……..

Answer 11

-

Question 12

Three definitions of CI - what are they? Technical, less precise and then our intuitive one. Can the CI ever be 100%

Answer 12

-

Question 13

Interpretation of the CI:

1. Why do we not include the actual parameter in the answer?
2. What is the interpretation?

Answer 13

Just don’t need to state it

Question 14

Can the CI be applied to any numerical value?

Answer 14

-

Question 15

If they give you an RR and a CI (that doesn’t cross 1), what three statements do you need to make?

Answer 15

1. Interpret RR
2. CI interpretation
3. P value

Question 16

Precision

1. What feature of the CI indicated precision?
2. “Precision is important” - explain that

Answer 16

-

Question 17

Fundamental concept #3

1. Increasing the sample size does what to the CI? and what does that have to do with precision of estimate? But don’t want it too large because?

Answer 17

-

Question 18

Clinical importance

1. RCTs often state what?
2. How do CI help with deciding whether the study findings are clinically important?

Answer 18

-